US20100326673A1

Movatterモバイル変換

Info

Publication number: US20100326673A1
Application number: US12/874,998
Authority: US
Inventors: Justin Borgstadt
Original assignee: Halliburton Energy Services Inc
Current assignee: Halliburton Energy Services Inc
Priority date: 2008-06-16
Filing date: 2010-09-02
Publication date: 2010-12-30
Anticipated expiration: 2028-06-16
Also published as: SG10201508215QA; SG10201508222VA; CA2795175C; CA2794538C; EP2310625A1; EP2392978A1; EP2392767A1; US7819183B2; EP2392767B1; SG192411A1; EP2383422A1; US7866403B1; US7845420B1; CA2795548A1; CA2795175A1; EP2310625B1; US7841414B1; CA2728172A1; SG10201508218PA; US20100326653A1

Abstract

Well bore servicing equipment is provided. The well bore servicing equipment comprises a first manipulator to grip a well bore work string, to raise the work string, and to lower the work string. The well bore servicing equipment further comprises a controller to receive a work string trajectory input and to automatically control the first manipulator to raise and lower the work string substantially in conformance with the work string trajectory input.

Description

CROSS-REFERENCE TO RELATED APPLICATIONS

This is a Continuation Application of U.S. patent application Ser. No. 12/140,191, filed Jun. 16, 2008 and published as US 2009/0308603 A1, and entitled “Work String Controller,” which is hereby incorporated by reference herein in its entirety.

STATEMENT REGARDING FEDERALLY SPONSORED RESEARCH OR DEVELOPMENT

Not applicable.

REFERENCE TO A MICROFICHE APPENDIX

Not applicable.

BACKGROUND

Automated control systems attempt to drive physical characteristics of a system, for example a process or object, to achieve system objectives. Automated control systems may be able to improve on the control provided manually by a human operator, for example by providing a higher frequency response, by taking account of a greater number of system parameters, and/or by providing a higher accuracy of control. Automated control systems may take many forms and may be designed to use continuous time controllers and/or discrete time controllers. Control systems may be both designed and described with control diagrams representing processing blocks. Generally, a control system may be built and implemented from the control system diagram.

A well bore may be serviced using a work string. A work string may include continuous coiled tubing which is fed continuously into the well bore from large spools. The longer the continuous tubing, the greater the tensile strength of the tubing may be to support the weight of the longer tubing. On the other hand, the greater the tensile strength of the tubing, the less flexible the tubing may be and the greater stress that may be produced in the tubing as it flexes going into the well bore and coming back out of the well bore. An advantage of coiled tubing is that it can be fed relatively rapidly and continuously into a well bore. A work string may also be composed of interconnected pieces or joints of pipe, for example joints of pipe about 10 meters long with a male threaded end and an opposite female threaded end. The pipe joints are connected together by threading two pipe joints together tightly. Various tools may be attached to the end of the work string —either coiled tubing or interconnected joints of pipe—to accomplish a variety of well bore operations.

SUMMARY

Further disclosed herein is an automated hydraulic flow regulator, comprising a proportional-integral (PI) controller portion to determine a flow error and to generate a corrective signal based on the flow error; a command feed-forward portion to determine a command feed-forward signal based on a flow rate command and an estimate of a hydraulic pump motor angular velocity; and a pump control gain to amplify the sum of the corrective signal and the command feed-forward signal to produce a commanded hydraulic pump motor signal, whereby the automated hydraulic flow regulator controls a hydraulic flow. The flow error may be determined as the difference of the flow rate command and a sensed flow rate. The command feed-forward signal may be determined by amplifying the flow rate command by a feed-forward gain, wherein the feed-forward gain may be inversely proportional to the estimate of the hydraulic pump motor angular velocity. The hydraulic flow regulator may further include an axial piston pump that produces the hydraulic flow, wherein the commanded hydraulic pump motor signal controls the axial piston pump by controlling the angular displacement of a swash plate of the axial piston pump.

Further disclosed herein is an automated force modulator, comprising a proportional gain controller portion to determine a force error and to generate a corrective signal based on the force error; a command feed-forward portion to determine a command-feed forward signal based on a velocity command and a directional area gain, wherein the directional area gain has a first value when the velocity command has a first polarity and the directional area gain has a second value when the velocity command has a second polarity; and a summation junction to output a commanded flow signal based on the corrective signal and the command feed-forward signal. The force error may be determined as the difference of a force command and an estimated force. The estimated force may be determined based on an indication of a rod side pressure of a hydraulic actuator and a piston side pressure of the hydraulic actuator. The first value of directional area gain may be proportional to the area of a piston of a hydraulic actuator, and the second value of directional area gain may be proportional to the area of the piston of the hydraulic actuator minus the cross-section area of a rod of the hydraulic actuator. The summation junction may output the commanded flow signal based further on a flow disturbance term. The automated force modulator may further comprise a directional valve modulator to output a directional valve command based on the polarity of a sum of the corrective signal and the command feed-forward signal.

Further disclosed herein is an automated manipulator controller, comprising a proportional-integral-derivative (PID) controller portion to determine a manipulator position error and a manipulator velocity error and to generate a corrective signal based on the manipulator position error and the manipulator velocity error; a command feed-forward portion to determine a command feed-forward signal based on a manipulator acceleration command and a manipulator mass gain; a manipulator damping gain to determine a manipulator damping force based on a manipulator damping gain and an indication of manipulator velocity; and a summation junction to output a first manipulator force command based on the corrective signal, the command feed-forward signal, the manipulator damping force, and a feed forward work string load command. The manipulator position error may be determined as the difference of a manipulator position command and an indication of manipulator position, and the manipulator velocity error may be determined as the difference of a manipulator velocity command and the indication of manipulator velocity. The summation junction may further output the first manipulator force command based on a manipulator weight. The summation junction may further output the first manipulator force command based on a difference between the manipulator force command and an estimate of a manipulator force disturbance.

Further disclosed herein is a pressure estimator comprising a directional valve command feed forward component to produce a rod side flow command and a piston side flow command based on a directional flow valve command and a feed forward flow rate command; a piston side pressure observer to produce an estimated piston side pressure based on the piston side flow command, a sensed piston side pressure, an actuator velocity, and an actuator position; and a rod side pressure observer to produce an estimated rod side pressure based on the rod side flow command, a sensed rod side pressure, the actuator velocity, and the actuator position. The pressure estimator may further comprise a directional valve disturbance flow component to estimate a disturbance flow based on the directional flow valve command, the estimated piston side pressure, and the estimated rod side pressure. When the directional flow valve command has a piston side value, the piston side flow command may be proportional to the feed forward flow rate command and the rod side flow command may be proportional to the negative value of the feed forward flow rate command. When the directional flow valve command has a rod side value, the piston side flow command may be proportional to the negative value of the feed forward flow rate command and the rod side flow command may be proportional to the value of the feed forward flow rate command. When the directional flow valve command has a piston side value, the rod side flow command further may be proportional to an effective rod side area of the piston divided by the effective piston side area of the piston. When the directional flow valve command has a rod side value, the piston side flow command further may be proportional to the effective piston side area of the piston divided by the effective rod side area of the piston.

Further disclosed herein is a pressure observer comprising a first proportional-integral (PI) controller portion to determine a first pressure error and to generate a first corrective signal based on the first pressure error; a first feed-forward portion to determine a first flow rate based on a velocity and a first gain; a first integrator to integrate the sum of the first corrective signal and the first flow rate; and a second gain to amplify an output of the first integrator by a first model gain and to produce a first pressure estimate. The first proportional-integral controller portion may further determine a first flow disturbance signal based on the first pressure error. The first pressure error may be determined as the difference of a first sensed pressure and the first pressure estimate. The first gain may be proportional to the difference of an area of a piston of a hydraulic actuator and a cross-section area of a rod of the hydraulic actuator. The first model gain may be proportional to a bulk modulus of a hydraulic fluid and may be inversely proportional to a volume comprising a rod side chamber of a hydraulic actuator, wherein the volume of the rod side chamber may change as a rod of the hydraulic actuator moves. The pressure observer may further comprise a second proportional-integral (PI) controller portion to determine a second pressure error and to generate a second corrective signal based on the second pressure error; a second feed-forward portion to determine a second flow rate based on the velocity and a second gain; a second integrator to integrate the sum of the second corrective signal and the second flow rate; and a second gain to amplify an output of the second integrator by a second model gain and to produce a second pressure estimate.

Further disclosed herein is a work string controller comprising a first proportional-integral-derivative (PID) controller portion to determine a first position error and a first velocity error and to generate a first simulated force feedback based on the first position error and the first velocity error; a second proportional-integral-derivative (PID) controller portion to determine a second position error and a second velocity error and to generate a second simulated force feedback based on the second position error and the second velocity error; a third proportional-integral-derivative (PID) controller portion to determine a work string position error and a work string velocity error and to generate a force term based on the work string position error, based on the work string velocity error, based on the first simulated force feedback, and based on the second simulated force feedback; a gain to amplify the force term by a work string mass gain to produce an acceleration term; a first integrator to integrate the acceleration term to produce an estimated work string velocity; a second integrator to integrate the estimated work string velocity term to produce an estimated work string position; and a commands generator to produce a first slip bowl command, a first acceleration command, a first velocity command, a first position command, a second slip bowl command, a second acceleration command, a second velocity command, and a second position command based on the estimated work string velocity and the estimated work string position. The commands generator may be further operable to produce the first and second slip bowl commands based on a collar indication. The work string mass gain may be inversely proportional to a mass of a work string. The work string controller may further comprise a first slip bowl portion to determine a first slip bowl error based on a first estimated slip bowl position, a first commanded slip bowl position, and a first slip bowl model; a first proportional-integral (PI) controller portion to generate a third simulated force feedback based on the first slip bowl error; a second slip bowl portion to determine a second slip bowl error based on a second estimated slip bowl position, a second commanded slip bowl position, and a second slip bowl model; and a second proportional-integral (PI) controller portion to generate a fourth simulated force feedback based on the second slip bowl error, wherein the third proportional-integral-derivative controller portion generates the force term further based on the third simulated force feedback and the fourth simulated force feedback.

Further disclosed herein is a collar locator, comprising a first digital camera to capture a first image; and a collar detector coupled to the first digital camera and operable to generate a first thresholded image of the first image, to generate a first edge detection analysis of the first thresholded image, and to determine the location of a collar based in part on the first edge detection analysis. The collar locator may further include a first light source, whereby the light source may illuminate a work string that is at least a portion of the first image. The collar locator may further comprise a second digital camera to capture a second image, wherein the collar detector is coupled to the second digital camera, and the collar locator may further generate a second thresholded image of the second image, generate a second edge detection analysis of the second thresholded image, and determine the location of the collar based in part on the second edge detection analysis. In an embodiment, the first and second digital cameras may each capture a plurality of images at a periodic rate, and the collar detector may determine a position and a velocity of the collar based on the plurality of images. In an embodiment, the first and second digital cameras may each capture about 30 images per second.

Further disclosed herein is a method of servicing a well bore with a work string comprising placing the work string in a well bore; receiving a work string trajectory input; determining a simulated force feedback of the work string on a first manipulator; determining an estimated work string velocity based at least on the simulated force feedback of the work string on the first manipulator; determining an estimated work string position based at least on the simulated force feedback of the work string on the first manipulator; determining a first manipulator position command and a first manipulator velocity command based on the estimated work string position and the estimated work string velocity; and automatically controlling the first manipulator based at least on the first manipulator position command and the first manipulator velocity command. Determining the estimated work string velocity may comprise determining a corrective signal based on a work string position command and a work string velocity command, combining the simulated force feedback of the work string on the at least one manipulator and the corrective signal to determine a force term, converting the force term to an acceleration term, and integrating the acceleration term to determine the simulated work string velocity. Determining the estimated work string position may comprise integrating the estimated work string velocity to determine the simulated work string position. The simulated force feedback of the work string on the at least one manipulator may be determined as the sum of a plurality of a simulated force feedback of the work string on a manipulator, and determining the simulated force feedback of the work string on the manipulator may comprise determining a manipulator position error term, determining a manipulator velocity error term, amplifying the manipulator position error term to produce a first proportional term, integrating and amplifying the manipulator position error term to produce a first integral term, amplifying the velocity error term to produce a first derivative term, and summing the first proportional term, the first integral term, and the first derivative term. In an embodiment, determining the simulated force feedback of the work string on the manipulator may further comprise determining a slip bowl position error, amplifying the slip bowl position error by a second proportional gain to produce a second proportional term, integrating and amplifying the slip bowl position error to produce a second integral term, and summing the second proportional term, the second integral term, the first proportional term, the first integral term, and the first derivative term. In an embodiment, the first integral gain may be used to produce the first integral term, and a second integral gain may be used to produce the second integral term, wherein the second integral gain may be at least about ten times larger than the first integral gain. Controlling the first manipulator may be further based on a collar indication input. Controlling the work string may further include determining a second manipulator position command and a second manipulator velocity command based on the simulated force feedback of the work string on the at least one manipulator, the estimated work string position, the estimated work string velocity, and the work string trajectory input; and automatically controlling a second manipulator based on the second manipulator position command and the second manipulator velocity command. The method may further comprise determining a simulated force feedback of the work string on an a second manipulator; determining the estimated work string velocity further based on the simulated force feedback of the work string on the second manipulator; determining the estimated work string position further based on the simulated force feedback of the work string on the second manipulator; determining a first feed forward work string load command, a second manipulator position command, a second manipulator velocity command, and a second feed forward work string load command based on the estimated work string position and the estimated work string velocity; further automatically controlling the first manipulator based on the first feed forward work string load command; and automatically controlling the second manipulator based at least on the second manipulator position command, the second manipulator velocity command, and the second feed forward work string load command.

These and other features will be more clearly understood from the following detailed description taken in conjunction with the accompanying drawings and claims.

BRIEF DESCRIPTION OF THE DRAWINGS

For a more complete understanding of the present disclosure, reference is now made to the following brief description, taken in connection with the accompanying drawings and detailed description, wherein like reference numerals represent like parts.

FIG. 1 is a block diagram of a control system according to some embodiments of the disclosure.

FIG. 2A is an illustration of a single actuator hydraulic power system according to some embodiments of the disclosure.

FIG. 2B is an illustration of a two actuator hydraulic power system according to other embodiments of the disclosure.

FIG. 3 is an illustration of a dual-jacking mechanism according to some embodiments of the disclosure.

FIG. 4 is a block diagram of a work string control system according to some embodiments of the disclosure.

FIG. 5 is an illustration of a feedback vector according to some embodiments of the disclosure.

FIG. 6 is a block diagram of a manipulator control system and a manipulator physical system interface according to some embodiments of the disclosure.

FIG. 7 is control diagram of a flow regulator according to some embodiments of the disclosure.

FIG. 8 is a control diagram of a force modulator according to some embodiments of the disclosure.

FIG. 9 is a control diagram of a traveling head controller according to some embodiments of the disclosure.

FIG. 10 is a block diagram of a drive observer according to some embodiments of the disclosure.

FIG. 11 is a control diagram of a rod side pressure observer and a piston side pressure observer according to some embodiments of the disclosure.

FIG. 12 is a control diagram of a traveling head position and velocity observer according to some embodiments of the disclosure.

FIG. 13 is a block diagram of a work string controller according to some embodiments of the disclosure.

FIG. 14 is a control diagram of a manipulator simulated force feedback component according to an embodiment of the disclosure.

FIG. 15 is an illustration of a kinematic model according to an embodiment of the disclosure.

FIG. 16 is an illustration of a pipe collar locator according to an embodiment of the disclosure.

FIG. 17 is a flow chart of a method of controlling a work string according to an embodiment of the disclosure.

FIG. 18 illustrates an exemplary general purpose computer system suitable for implementing the several embodiments of the disclosure.

DETAILED DESCRIPTION

It should be understood at the outset that although illustrative implementations of one or more embodiments are illustrated below, the disclosed systems and methods may be implemented using any number of techniques, whether currently known or in existence. The disclosure should in no way be limited to the illustrative implementations, drawings, and techniques illustrated below, but may be modified within the scope of the appended claims along with their full scope of equivalents.

In the figures and in the text herein below, variable names and variable symbols that are associated with an asterisk, for example F*, generally represent commanded values. Variable names and variable symbols that are associated with a caret, for example F̂, generally represent estimated values. Variable names and variable symbols that are not associated with any additional symbol, for example F, generally represent measured and/or actual values. Turning now toFIG. 1, acontrol system architecture8 suitable for some of the embodiments of the present disclosure is discussed. Thecontrol system architecture8 includes asystem controller10, a plurality ofmanipulator controllers20, amanipulator observer system30, a plurality ofdrive modulators40, adrive observer system50, a plurality of manipulatorphysical systems60, and a manipulatedphysical system70. An embodiment of thesystem controller10 is shown inFIG. 13 and described in more detail hereinafter. An embodiment of themanipulator controller20 is shown inFIG. 9 and described in more detail hereinafter. An embodiment of themanipulator observer system30 is shown inFIG. 12 and described in more detail hereinafter. An embodiment of thedrive modulator40 is shown inFIG. 7 andFIG. 8 and described in more detail hereinafter. An embodiment of thedrive observer system50 is shown inFIG. 10 andFIG. 11 and described in more detail hereinafter. An embodiment of the manipulatorphysical system60 is shown inFIG. 2A,FIG. 2B,FIG. 3,FIG. 6, andFIG. 7 and described in more detail hereinafter. An embodiment of the manipulatedphysical system70 is shown inFIG. 3 andFIG. 16 and discussed in more detail hereinafter.

WhileFIG. 1 depicts threeseparate manipulator systems60, in different embodiments, different numbers ofmanipulators60,drive modulators40, andmanipulator controllers20 may be implemented. Also, whileFIG. 1 depicts a second manipulator physical system60-b, a second drive modulator40-b, and a second manipulator controller20-bas well as a third manipulator physical system60-c, a third drive modulator40-c, and a third manipulator controller20-cwithout an associatedmanipulator observer system30 and without an associateddrive observer system50, in other embodiments amanipulator observer system30 and/or adrive observer system50 may be associated with these system control components. Thesystem controller10 receives a systemcontroller command signal12, aphysical system feedback16 from the manipulatedphysical system70, and amanipulator feedback14 from each of the manipulatorphysical systems60 as input, for example a first manipulator feedback14-afrom the first manipulator physical system60-a, a second manipulator feedback14-bfrom the second manipulator physical system60-b, and a third manipulator feedback14-cfrom the third manipulator physical system60-c. In an embodiment, the systemcontroller command signal12 may comprise a work string trajectory input. Thesystem controller10 outputs a manipulatorcontroller command signal22 to each of themanipulator controllers20. In an embodiment, the manipulatorcontroller command signal22 may comprise a manipulator command trajectory that is based on a work string trajectory input to thesystem controller10. In an embodiment, a first manipulator controller command signal22-amay comprise a first manipulator command trajectory and a second manipulator controller command signal22-bmay comprise a second manipulator command trajectory, wherein both the first manipulator command trajectory and the second manipulator command trajectory are based on the work string trajectory input. The general purpose of thecontrol system architecture8 is to drive thephysical system70 according to the systemcontroller command signal12. In an embodiment, the systemcontroller command signal12 may be a trajectory describing the position, velocity, and acceleration of the manipulatedphysical system70 at different times. The systemcontroller command signal12 may further include various mode and commands. In an embodiment, thecontrol system architecture8 comprises well bore servicing equipment, for example a control system automatically controlling a plurality of manipulators to raise and lower a well bore work string in and out of a well bore to accomplish a well bore servicing job.

Each of themanipulator controllers20 receives a manipulatorcontroller command signal22 from thesystem controller10 and one of a manipulatorobserver feedback signal34 from themanipulator observer system30 or amanipulator feedback24 as inputs. Each of themanipulator controllers20 outputs a drivemodulator command signal42 to thedrive modulator40. The first manipulator controller20-aalso outputs a first manipulatorcontroller output signal32 to themanipulator observer system30. With respect to describingFIG. 1, the term “signal” may mean either a single signal or a vector of signals. For example, in an embodiment, thecommand signal22 may comprise a commanded traveling head position signal, a commanded traveling head velocity signal, a commanded traveling head acceleration signal, a commanded traveling head force signal, and a commanded slip bowl position signal.

Each of thedrive modulators40 receives the drivemodulator command signal42 from themanipulator controller20 and amanipulator feedback signal44 from the manipulatorphysical system60. The first drive modulator40-aalso receives a first drive observersystem output signal58 from the firstdrive observer system50. Each of thedrive modulators40 outputs a manipulator physicalsystem command signal62 to the manipulatorphysical system60. The first drive modulator40-aalso outputs a first drivemodulator output signal52 to the firstdrive observer system50.

Themanipulator observer system30 receives the first manipulatorcontroller output signal32 and a first manipulator feedback24-a. Themanipulator observer system30 outputs the first manipulatorobserver feedback signal34 to the first manipulator controller20-aand a first manipulator observersystem output signal54 to the firstdrive observer system50.

The firstdrive observer system50 receives the firstobserver output signal54 from themanipulator observer system30, the first drivemodulator output signal52 from the first drive modulator40-a, and a firstmanipulator feedback signal56 from the first manipulator physical system60-a. The firstdrive observer system50 outputs the first drive observersystem output signal58 to the first drive modulator40-a.

Each of the manipulatorphysical systems60 receives the manipulator physicalsystem command signal62 from thedrive modulator40. Each of the manipulatorphysical systems60 outputs a plurality of feedback signals14,24, and44 to thesystem controller10, themanipulator observer system30 or themanipulator controller20, and thedrive modulator40, respectively. Additionally, the first manipulator physical system60-aoutputs a first manipulator feedback signal to the firstdrive observer system50. Each of the manipulatorphysical systems60 also interacts with the manipulatedphysical system70, represented inFIG. 1 by the manipulated physical system interaction withmanipulator72. For example, in an embodiment, the manipulatedphysical system70 may be a work string for servicing and/or drilling a well bore and themanipulators60 may be a plurality of hydraulic jacks coupled to slip bowls configured to grip the work string. The work string may comprise a plurality of connected segments (e.g., drill string, tubing string, casing string, etc.) or a continuous length of oilfield tubular such as coiled tubing. The work string may have one or more associated or connected tools, for example one or more down hole tools positioned at or near a terminal end of the work string. In some embodiments, the several manipulatorphysical systems60 that manipulate the manipulatedsystem70 may be of different types. The manipulatorphysical system60 may include hydraulic actuators, electric motor actuated screw jacks, robot lever arms, slip bowls, and other devices. In another embodiment, the manipulatedphysical system70 may be some other object manipulated by one or more manipulatorphysical systems60 that are robotic arms. It should be understood that thecontrol system architecture8 depicted inFIG. 1 is suitable to a number of alterations, modifications, and arrangements of components, all of which are contemplated by the present disclosure.

Turning now toFIG. 2A, a single actuatorhydraulic system100 is described. The single actuatorhydraulic system100 comprises ahydraulic pump102, adirectional flow valve104, a hydraulicpressure supply line106, ahydraulic return line107, ahydraulic fluid reservoir108, ahydraulic supply line109, ahydraulic actuator110, a rod sidehydraulic line112, a piston sidehydraulic line114, a first counterbalance valve115-a, and a second counterbalance valve115-b. In an embodiment, thehydraulic pump102 provides pressurized flow of hydraulic fluid at an effective pressure and rate of flow to drive thehydraulic actuator110 according to operational control regimes. In different embodiments, differenthydraulic pumps102 may be selected to provide different flow/pressure capacities and/or different pump ratings. Pressurized hydraulic fluid flows through the hydraulicpressure supply line106 to thehydraulic actuator110 under the control of thedirectional flow valve104. The hydraulic fluid is returned from thehydraulic actuator110 through thehydraulic return line107 tohydraulic fluid reservoir108 under the control of thedirectional flow valve104. Thehydraulic pump102 draws hydraulic fluid from thehydraulic fluid reservoir108 via thehydraulic supply line109. Thedirectional flow valve104 directs pressurized hydraulic fluid to the rod sidehydraulic line112 when in a first control state, to the piston sidehydraulic line114 when in a second state, and to thehydraulic return line107 when in a third state.

In an embodiment, thedirectional flow valve104 has four ports which connect to the hydraulicpressure supply line106, thehydraulic return line107, the rod sidehydraulic line112, and the piston sidehydraulic line114. Thedirectional flow valve104 has an internal diverting spool that is electrically actuated by use of a first and a second solenoid. When the first solenoid is energized, the internal diverting spool is displaced to a first position, connecting the hydraulicpressure supply line106 with the rod sidehydraulic line112 and connecting thehydraulic return line107 with the piston sidehydraulic line114. When the second solenoid is energized, the internal diverting spool is displaced to a second position, connecting the hydraulicpressure supply line106 to the piston sidehydraulic line114 and thehydraulic return line107 to the rod sidehydraulic line112. When neither the first solenoid or the second solenoid is energized, the internal diverting spool remains in a neutral position, and all four hydraulic lines—the hydraulicpressure supply line106, thehydraulic return line107, the rod sidehydraulic line112, and the piston sidehydraulic line114—are connected together, effectively routing the hydraulicpressure supply line106 to thehydraulic return line107 and bypassing both the rod sidehydraulic line112 and the piston sidehydraulic line114.

Thehydraulic actuator110 comprises arod116 attached to apiston118. Therod116 is supported and retained by an end cap (not shown) of thehydraulic actuator110 and by thepiston118. Thepiston118 is supported by the interior of thehydraulic actuator110. Therod116 may be coupled to a weight bearing structure (not shown) to manipulate or move the weight bearing structure. The interior of thehydraulic actuator110 includes arod side chamber120 and apiston side chamber122. By directing hydraulic fluid at different pressures into therod side chamber120 from the rod sidehydraulic line112 and into thepiston side chamber122 from the piston sidehydraulic line114, therod116 is driven under force in different directions. The force exerted by therod116 may be calculated to be the difference of the product of an area of thepiston118 multiplied by the hydraulic pressure, P_PS, in thepiston side chamber122 and a product of the hydraulic pressure, P_RS, in therod side chamber120 multiplied by an area determined as the area of thepiston118 minus a cross-sectional area of therod116. In different embodiments, differenthydraulic actuators110 may be selected having different stroke lengths, different diameters, different piston sizes, different chamber volumes, and other different specifications, capacities, and/or dimensions.

In an embodiment, a first counterbalance valve115-ais installed in the rod sidehydraulic line112 and a second counterbalance valve115-bis installed in the piston sidehydraulic line114. The first and second counterbalance valves115-a, bare cross-connected to each other. The purpose of the first and second counterbalance valves115-a, bis to hold any overrunning loads of thehydraulic actuator110. For example, if thehydraulic actuator110 is extended and bearing a heavy weight, as may be the case when thehydraulic actuator110 is supporting a long pipe string or coiled tubing, and then the single actuatorhydraulic system100 is controlled or commanded to direct hydraulic fluid and pressure to therod side chamber120 and to return hydraulic fluid from thepiston side chamber122, while not choking back the flow out of thepiston side chamber122, thehydraulic actuator110 and the load it supports may fall uncontrolled at the rate which the hydraulic fluid can escape thepiston side chamber122. The first and second counterbalance valves115-a, bpromote maintaining controlled movement of thehydraulic actuator110 by holding back hydraulic fluid so that thehydraulic actuator110 does not run away or fall.

Turning now toFIG. 2B, a two actuatorhydraulic system130 is described. The two actuatorhydraulic system130 is substantially similar to the single actuatorhydraulic system100, with the difference being that the two actuatorhydraulic system130 contains twohydraulic actuators110, a first hydraulic actuator110-aand a second hydraulic actuator110-b, plumbed in parallel. The first hydraulic actuator110-acomprises a first rod116-aattached to a first piston118-a. The interior of the first hydraulic actuator110-aincludes a first rod side chamber120-aand a first piston side chamber122-a. Similarly, the second hydraulic actuator110-bcomprises a second rod116-battached to a second piston118-b. The interior of the second hydraulic actuator110-bincludes a second rod side chamber120-band a second piston side chamber122-b. The two rod side chambers120-a, bare plumbed in parallel from the common rod sidehydraulic line112, and the two piston side chambers122-a, bare plumbed in parallel from the common piston sidehydraulic line114. Because thehydraulic actuators110 are plumbed in parallel, the function of thedirectional flow valve104 and thecounterbalance valves115 remain the same as described with respect toFIG. 2A. The use of multiplehydraulic actuators110, for example two hydraulic actuators110-a,110-bas illustrated inFIG. 2B, may provide increased force to apply to a manipulated object. Additionally, the use of multiplehydraulic actuators110 may promote ease of force transfer from thehydraulic actuators110 to the manipulated object or objects.

The two hydraulic actuators110-c, dare substantially similar to the hydraulic actuators110-a, bdescribed above with reference toFIG. 2B. Because in some control regimes or operational modes the two hydraulic actuators110-a, band the two hydraulic actuators110-c, dmay be commanded independently, the two hydraulic actuators110-a, bmay be coupled to a first directional flow valve104-a(not shown) and a first hydraulic pump102-a(not shown) and the two hydraulic actuators110-c, dmay be coupled to a second directional flow valve104-b(not shown) and a second hydraulic pump102-b(not shown).

In one case, for example when the weight of thework string152 is moderate, thehydraulic actuators110 may be commanded so that while thefirst slip bowl154 grips thework string152 and the first and second hydraulic actuators110-a, bmove thework string152 in a positive Z direction, thesecond slip bowl158 is disengaged from thework string152 and the third and fourth hydraulic actuators110-c, dmove in a negative Z direction. At about the limits of travel, thehydraulic actuators110 are commanded to stop, bringing thework string152, both slip bowls154,158, and both traveling

heads

156,160 to a stop with zero velocity. Thesecond slip bowl158 is engaged to grip thework string152, and thefirst slip bowl154 is disengaged from thework string152. This operation may be referred to as a “hand-off” from one

slip bowl

154,158 to another and may involve additional movements of one or more of the traveling heads156,160 for thesecond slip bowl158 to fully engage thework string152 and for thefirst slip bowl154 to fully disengage from thework string152. The third and fourth hydraulic actuators110-c, dmay then be commanded to move thework string152 in the positive Z direction, while the first and second hydraulic actuators110-a, bare commanded to move in the negative Z direction. At about the limits of travel, thehydraulic actuators110 may be commanded to stop, bringing thework string152, both slip bowls154,158, and both traveling

heads

156,160 to a stop with zero velocity. A hand-off is then performed transferring thework string152 from the grip of thesecond slip bowl158 to the grip of thefirst slip bowl154. This cycle may be repeated, moving thework string152 further into the well or reversed and repeated, moving thework string152 out of the well. In some contexts, this mode of operation may be referred to as a high speed sequential mode.

Using the high speed sequential mode, the four hydraulic actuators110-a, b, c, dmay cooperate to maintain a sequence of start-stop movements of thework string152, driving thework string152 at an average commanded velocity V to a commanded depth Z. In this mode, the two pairs of hydraulic actuators—hydraulic actuators110-a, band hydraulic actuators110-c, d—can operate out-of-phase, handing off between each other, thereby achieving higher velocities of thework string152 than may be possible with a single pair of hydraulic actuators, for example110-a, b. In some contexts herein, the commanded velocity V and commanded depth Z may be referred to as a trajectory of thework string152. Alternatively, a plurality of commanded velocities V and commanded depths Z may be concatenated into a series that may also be referred to as a trajectory of thework string152.

In another case, for example when thework string152 may be too heavy for manipulation by one pair ofhydraulic actuators110 at a time, the four actuators110-a, b, c, d and the two slip bowls154,158 may be commanded to share the load of manipulating thework string152. In this case, the first and second slip bowls154,158 may be engaged to grip thework string152 and the four hydraulic actuators110-a, b, c, dmay be commanded to move thework string152 in the positive Z direction to about one limit of their travel. The four hydraulic actuators110-a, b, c, dthen may be commanded to bring thework string152 to a stop, zero velocity V, whereupon thestationary slip bowl162 may be engaged to grip and hold the weight of thework string152. When thestationary slip bowl162 is supporting thework string152, the first and second slip bowls154,158 may be disengaged, and the four hydraulic actuators110-a, b, c, dmay be commanded to move in the negative Z direction to about the opposite limit of their travel. The four hydraulic actuators110-a, b, c, dmay then be commanded to come to a stop, zero velocity V, whereupon the first and second slip bowls154,158 may be commanded to engage and grip thework string152. Thestationary slip bowl162 then may be commanded to disengage, and the cycle may be repeated. In some contexts, this mode of operation may be referred to as a distributed load or high capacity sequential mode. This mode may be characterized by a coordinated sequence of start and stop operations between the two pairs of hydraulic actuators—hydraulic actuators110-a, band hydraulic actuators110-c, d—working in unison and using thestationary slip bowl162 to bear thework string152 during stops. This mode may provide a capacity to manipulate higher loads, for example aheavier work string152, but with lower velocity than may be possible with the high speed sequential mode.

In another case, for example when the weight of thework string152 is within the handling capacity of one pair ofhydraulic actuators110, for example the hydraulic actuators110-a, b, the four hydraulic actuators110-a, b, c, dand the two slip bowls154,158 may be synchronized to maintain a substantially continuous movement of thework string152. In other words, the hydraulic actuators110-a, b, c, dand the slip bowls154,158 may move in such a way as to provide substantially constant velocity of thework string152 or substantially constant acceleration or deceleration of thework string152. The hand off of thework string152 between the hydraulic actuators110-a, b, c, dand the slip bowls154,158 may be coordinated “on the fly” to promote this continuous motion. In some contexts, this mode of operation may be referred to as a high speed continuous mode. This mode may have the advantage of reducing the power and/or energy consumption requirements necessary to start and stop thework string152 with each motion as described in the high speed sequential mode. Additionally, this mode may reduce stress and strain on the various components of the dual-jackingsystem150 and/or the hydraulic components, thereby possibly extending the service life of the same.

In another case, for example when the weight of thework string152 is within the handling capacity of one pair ofhydraulic actuators110, for example the hydraulic actuators110-a, b,constraints other than the maximum handling capacity of the pairs ofhydraulic actuators110 may constrain thework string152 velocities and/or accelerations. For example, mechanical, electrical, and/or safety constraints may limit the speed at which a hand-off “on the fly” can be accomplished. This maximum hand-off speed may be slower than the desired or commanded velocity of thework string152. In order to achieve an overall “average” work string speed which matches the commanded velocity of thework string152, the four hydraulic actuators110-a, b, c, dand the two slip bowls154,158 may be synchronized. When synchronized, the four hydraulic actuators110-a, b, c, dand the two slip bowls154,158 may decelerate thework string152 to the maximum hand-off speed during hand-offs and accelerate thework string152 to a velocity between hand-offs such that the overall average velocity of thework string152 is the commanded velocity of thework string152. In some contexts, this mode of operation may be referred to as a high speed constrained mode.

In another case, for example when the weight of thework string152 is within the handling capacity of one pair ofhydraulic actuators110, for example the hydraulic actuators110-a, b, mechanical, electrical, or safety constraints or considerations may make it desirable to use thestationary slip bowl162 during the hand off between thehydraulic actuators110, for example between the hydraulic actuators110-a, band the hydraulic actuators110-c, d. The four hydraulic actuators110-a, b, c, dand the three

slip bowls

154,158, and162 may be synchronized such that after a stroke of manipulating thework string152 is completed by the hydraulic actuators110-a, bthefirst slip bowl154 hands off thework string152 to thestationary slip bowl162. At this point, the slip bowls154 and158 are both disengaged from thework string152. Thestationary slip bowl162 then hands off thework string152 to thesecond slip bowl158 with the hydraulic actuators110-c, din position for a full stroke of manipulating thework string152. The hydraulic actuators110-c, dthen manipulate thework string152 while the hydraulic actuators110-a, b reposition thefirst slip bowl154 to a starting position for the next hand off. In some contexts, this mode of operation may be referred to as a high speed sequential mode with the stationary slip. Generalizing, this mode of operation may be referred to as a three manipulator high speed sequential mode as opposed to the two manipulator high speed sequential mode described above.

In another case, for example when the weight of thework string152 exceeds the handling capacity of one pair ofhydraulic actuators110, for example the hydraulic actuators110-a, b, three or more pairs ofhydraulic actuators110 may be employed to provide a high capacity continuous mode of operation. For example, hydraulic actuators110-a, b, c, d, e, fmay be operated with hydraulic actuators110-a, bforming in part a first manipulator, hydraulic actuators110-c, dforming in part a second manipulator, and hydraulic actuators110-e, fforming in part a third manipulator. The first, second, and third manipulators may be coupled to slip bowls by a traveling head as depicted inFIG. 3. The first and second manipulator may grip thework string152 and move the work string while the third manipulator is repositioned. The third manipulator may then grip thework string152, and the second manipulator may release thework string152. The first and third manipulators may move thework string152 while the second manipulator is repositioned. The second manipulator may then grip thework string152, and the first manipulator may release thework string152. The second and third manipulators may move thework string152 while the first manipulator is repositioned. The first manipulator may then grip thework string152, and the third manipulator may release thework string152. The cycle may be repeated to provide substantially continuous movement of thework string152. It will be appreciated by one skilled in the art that other numbers of manipulators may be combined in a fashion similar to that described above to provide substantially continuous motions while distributing the weight of thework string152 over multiple manipulators.

The mode of operation chosen determines how the commanded trajectories of the manipulators (traveling head position, traveling head velocity, slip bowl position) are computed through kinematics. Additionally, all of these modes are relevant with any number, type, or combination of manipulators. For example, a system which consists of four manipulators could be used where the weight of thework string152 exceeds the capacity of a single manipulator but does not exceed the capacity of two manipulators working in unison. A mode can be envisioned where the manipulators are paired into two sets for capacity reasons and the paired sets of manipulators are operated in a high speed continuous mode to maximize pipe speed.

Another example would be the same four manipulator system where the weight of thework string152 exceeds the capacity of two manipulators but does not exceed the capacity of three manipulators working in unison. A mode can be envisioned where there are always three manipulators in contact with thework string152 for capacity reasons and the manipulators are operated in a version of the high speed continuous mode to maximize the velocity of thework string152.

With multiple manipulators in a particular system a series of modes can be envisioned where the control system automatically changes from mode to mode to optimize performance and prevent failure as the weight of thework string152 increases or decreases.

In combination with the present disclosure, one skilled in the art may recombine or extend the scenarios described above to describe other controlling regimes, all of which are contemplated by the present disclosure and system.

Turning now toFIG. 4, a workstring control system200 is described. The workstring control system200 is one embodiment of thecontrol system architecture8 described with reference toFIG. 1 above. The workstring control system200 comprises acontroller202, a commandedtrajectory input204, a firstflow rate command210, a firstflow direction command212, a first slipbowl position command214, a secondflow rate command216, a secondflow direction command218, a second slipbowl position command220, a thirdflow rate command222, a thirdflow direction command224, a third slipbowl position command226, afirst feedback vector230, asecond feedback vector234, and athird feedback vector238. In an embodiment, any number of flow rate commands and flow direction commands may be provided by thecontroller202. Similarly, in an embodiment, any number of feedback vectors may be received by thecontroller202. Thecontroller202 is configured to automatically generate the flow rate commands210,216, and222; the flow direction commands212,218, and224; and the slip bowl position commands214,220, and226 based on the commandedtrajectory input204 and the

feedback vectors

230,234, and238. In an embodiment, each one of the commands may depend upon all of the input vectors. For example, the firstflow rate command210 may depend not only on thefirst feedback vector230 but also on thesecond feedback vector234, because the motions of the first and second traveling heads154,158 may not be commanded entirely independently of each other. In an embodiment, the third slipbowl position command226 may control thestationary slip bowl162 and the thirdflow rate command222 and the thirdflow direction command224 may remain zero.

In an embodiment, thecontroller202 is implemented on a general purpose computer system using digital control methods based on discrete time processing of sampled inputs. In another embodiment, however, thecontroller202 may be implemented as a combination of the general purpose computer system and some analog processing components and at least part of thecontroller202 may use analog control methods based on continuous time processing of analog inputs. Analog feedback control system components are well known to those skilled in the art and may be implemented, for example, using differential amplifiers, capacitors, and resistors to compose integrators, differentiators, amplifiers, and other common analog feedback control system components. General purpose computer systems are discussed in further detail hereinafter. It will be understood by one skilled in the art that the several components of thecontrol system architecture8 including the manipulator controllers20-a, b, c, the drive modulators40-a, b, c, themanipulator observer system30, and thedrive observer system50 may be conceptually aggregated as thecontroller202 and implemented in one or more coordinated computer programs or software components that are executed on one or more computer systems.

In an embodiment, thecontroller202 is coupled to the dual-jackingsystem150 described above and is operable to control the actuators110-a, b, c, dand to control the first and second slip bowls154,158 to manipulate thework string152 to achieve the commandedtrajectory input204. In this embodiment, the firstflow rate command210 may be coupled to a first hydraulic pump102-a(not shown) associated with the first and second hydraulic actuators110-a, b, the firstflow direction command212 may be coupled to a first directional flow valve104-a(not shown) associated with the first and second hydraulic actuators110-a, b, and the first slipbowl position command214 may be coupled to thefirst slip bowl154. In this embodiment, the secondflow rate command216 may be coupled to a second hydraulic pump102-b(not shown) associated with the third and fourth hydraulic actuators110-c, d, the secondflow direction command218 may be coupled to a second directional flow valve104-b(not shown) associated with the third and fourth hydraulic actuators110-c, d, the second slipbowl position command220 may be coupled to thesecond slip bowl158, and the third slipbowl position command226 may be coupled to thestationary slip bowl162.

In this embodiment, velocity and position sensors coupled to one or both of the first and second hydraulic actuator110-a, band pressure and flow sensors coupled to a first rod side hydraulic line112-a(not shown) and a first piston side hydraulic line114-a(not shown) connected in parallel to the first and second hydraulic actuators110-a, bmay be coupled to thecontroller202 as thefirst feedback vector230. In this embodiment, velocity and position sensors coupled to one or both of the third and fourth hydraulic actuators110-c, dand pressure and flow sensors coupled to a second rod side hydraulic line112-b(not shown) and a second piston side hydraulic line114-b(not shown) connected in parallel to the third and fourth hydraulic actuators110-c, dmay be coupled to thecontroller202 as thesecond feedback vector234. In an embodiment, the velocity and position sensors may be coupled to therod116 of thehydraulic actuator110.

Turning now toFIG. 5, anexemplary feedback vector250 is described. Thefeedback vector250 may comprise a pistonside pressure feedback252, a rodside pressure feedback254, aflow rate feedback256, a travelinghead velocity feedback258, a travelinghead position feedback260, a slipbowl position feedback261, and a work stringcollar location feedback262. In other embodiments, some of the depicted feedbacks may not be present and other feedbacks not depicted may be present in thefeedback vector250. In another embodiment, velocity of thework string152 and position of thework string152 may be present in thefeedback vector250.

Turning now toFIG. 6, amanipulator control system300 is described. Themanipulator control system300 is one embodiment of portions of thecontrol system architecture8. In the following descriptions some control components associated with specific embodiments are associated with their corresponding generic component in thecontrol system architecture8 depicted inFIG. 1, for example by enclosing one or more components in a dotted line box and referring to the dotted line box with a label from thecontrol system architecture8 depicted inFIG. 1. Thecontrol system architecture8 depicted inFIG. 1 should not be limited by the following description, because the specific embodiments of control system components described hereinafter are only some of the wide variety of possible embodiments of thecontrol system architecture8 that are contemplated by the present disclosure.

Themanipulator control system300 automatically controls a physical system, for example the manipulatorphysical system60, through a manipulatorphysical system interface302. In an embodiment, the physical system associated with the manipulatorphysical system interface302 may be substantially similar to portions of the dual-jackingsystem150 and portions of the two actuatorhydraulic system130. In another embodiment, however, themanipulator control system300 may control a different manipulatorphysical system60. Themanipulator control system300 may be said to have two degrees of freedom because it controls thefirst slip bowl154 and it controls the first travelinghead156, for example by controlling the first directional valve104-aand the first hydraulic pump102-a. In an embodiment, the manipulatorphysical system interface302 comprises thefirst slip bowl154, the first directional flow valve104-a, the first hydraulic pump102-a, aflow sensor310, a first pressure sensor312-a, a second pressure sensor312-b, a travelinghead position sensor314, and a travelinghead velocity sensor316. Themanipulator control system300 may be implemented on a general purpose computer system using digital control methods based on discrete time processing of sampled inputs.

In another embodiment, however, thecontrol system300 may include some analog components which process continuous time inputs according to analog feedback control methods.

In an embodiment, themanipulator control system300 comprises aflow regulator320, aforce modulator322, apressure estimator324, a travelinghead controller326, and a traveling head position andvelocity observer328. In other embodiments, themanipulator control system300 may comprise other components. For example, in an embodiment, themanipulator control system300 may not comprise thepressure estimator324, and instead the outputs of the pressure sensors312-a, bmay be directly input to theforce modulator322. In an embodiment, themanipulator control system300 may not comprise the traveling head position andvelocity observer328, and instead the output of the travelinghead position sensor314 and the output of the travelinghead velocity sensor316 may be directly input to the travelinghead controller326. Theflow regulator320 in combination with theforce modulator322 form an embodiment of thedrive modulator40 depicted inFIG. 1. Thepressure estimator324 is an embodiment of thedrive observer system50 depicted inFIG. 1. The traveling head position andvelocity observer328 is an embodiment of themanipulator observer system30 depicted inFIG. 1. The travelinghead controller326 is an embodiment of themanipulator controller20 depicted inFIG. 1.

In an embodiment, a slip bowl command (S_SB*) commands thefirst slip bowl154 to an open or a closed position or state. In a different embodiment, however, a slip bowl controller (not shown) may be employed to generate a slip bowl command based on a sensed or estimated position of thefirst slip bowl154 and based on a desired position of thefirst slip bowl154.

In an embodiment, theflow regulator320 receives a flow rate command (Q*) input from theforce modulator322 and a flow rate feedback (Q) from theflow sensor310 and automatically produces a current command (I*) to regulate the first hydraulic pump102-ato provide the desired hydraulic flow rate to the manipulator, for example to the hydraulic actuators110-a, b. In another embodiment, however, no flow rate feedback is provided to theflow regulator320 which operates in an open-loop control mode. An embodiment of theflow regulator320 is described further hereinafter.

In an embodiment, theforce modulator322 receives a traveling head force command (F*_TH) and a traveling head velocity command (V*_TH) from the travelinghead controller326. In an embodiment, theforce modulator322 also receives an estimated piston side pressure feedback (P̂_PS), an estimated rod side pressure feedback (P̂_RS), and an estimated flow rate disturbance feedback (Q̂_D) from thepressure estimator324. In combination with the present disclosure, one skilled in the art will readily appreciate that an estimated parameter value, while it is related to a sensed parameter value, may be different from the sensed parameter value. For example, an estimated parameter value may be a smoothed or filtered version of the sensed parameter value that attenuates noise produced by a sensor or by an environment. Additionally, an estimated parameter value may partially reduce or remove phase shifts and/or time lags of system response. In another embodiment, however, theforce modulator322 receives a sensed piston side pressure feedback (P_PS) and a sensed rod side pressure feedback (P_RS), for example in an embodiment which does not comprise thepressure estimator324. Theforce modulator322 automatically produces the flow rate command Q* and provides the flow rate command Q* to theflow regulator320. Theforce modulator322 also automatically produces a directional flow valve command (S_DV*) to control the state of thedirectional flow valve104. In an embodiment that comprises apressure estimator324, theforce modulator322 also automatically produces an observer feed forward flow rate command Q*_Oand provides the observer feed forward flow rate command Q*_Oto thepressure estimator324. In an embodiment, the directional flow valve command may be mapped by a digital-to-analog converter device (not shown) to produce an electrical current to energize the first or the second solenoid to actuate the diverting spool to control the first directional flow valve104-a. An embodiment of theforce modulator322 is described further hereinafter.

In an embodiment, thepressure estimator324 receives a sensed piston side pressure (P_PS) input, a sensed rod side pressure (P_RS) input, the directional flow valve command (S_DV*) input, the observer feed forward flow rate command (Q*_O) input, an estimated traveling head position (Ẑ_TH) input, an estimated traveling head velocity (V̂_TH) input and automatically produces the estimated piston side pressure, the estimated rod side pressure, and the estimated disturbance flow rate. In another embodiment, one or more of the estimated traveling head position (Ẑ_TH) and the estimated traveling head velocity (V̂_TH) may be sensed rather than estimated values. Each of the estimated pressures are a zero time lagged, filtered signal. An embodiment of thepressure estimator324 is discussed further hereinafter.

In an embodiment, the travelinghead controller326 receives a traveling headcommand input vector340 that comprises a traveling head position command (Z_TH*), the traveling head velocity command (V_TH*), a traveling head acceleration command (A_TH*), and a feed forward work string load on the traveling head (F*_WS/TH). The travelinghead controller326 also receives an estimated traveling head position (Ẑ_TH) input, an estimated traveling head velocity (V̂_TH) input and an estimated traveling head force disturbance (F̂_D) input from the position andvelocity observer328. In an embodiment, however, themanipulator control system300 does not comprise a traveling head position andvelocity observer328, and the travelinghead controller326 receives a sensed traveling head position (Z_TH) from the travelinghead position sensor314 and/or a sensed traveling head velocity (V_TH) from the travelinghead velocity sensor316. In general, a velocity sensor may provide a more accurate indication of manipulator velocity, for example traveling head velocity, than differentiating the output of a manipulator position sensor over time to calculate manipulator velocity, because the differentiation operation may produce unreliable results caused by noise in the signal generated by theposition sensor314. The travelinghead controller326 automatically produces the traveling head force command F*_THand provides the traveling head force command F*_THto theforce modulator322. The travelinghead controller326 automatically produces the observer feed forward traveling head force command F*_THOand provides the observer feed forward traveling head force command F*_THOto the traveling head position andvelocity observer328. An embodiment of the travelinghead controller326 is discussed further hereinafter.

The traveling head position andvelocity observer328 receives the observer feed forward traveling head force command F*_THOfrom the travelinghead controller326, the sensed traveling head position input from the travelinghead position sensor314, and the sensed traveling head velocity input from the travelinghead velocity sensor316. The traveling head position andvelocity observer328 automatically produces the estimated traveling head position, the estimated traveling head velocity, and the estimated traveling head force disturbance and provides the estimated traveling head position, the estimated traveling head velocity, and the estimated traveling head force disturbance to the travelinghead controller326. Each of the estimated traveling head position and estimated traveling head velocity is a zero time lagged, filtered signal. An embodiment of the traveling head position andvelocity observer328 is discussed further hereinafter. In combination with the present disclosure, one skilled in the art will readily appreciate that an estimated parameter value, while it is related to a sensed parameter value, may be different from the sensed parameter value. For example, an estimated parameter value may be a smoothed or filtered version of the sensed parameter value that attenuates noise produced by a sensor or by an environment. Additionally, an estimated parameter value may partially reduce or remove phase shifts and/or time lags of system response.

Turning now toFIG. 7, an embodiment of theflow regulator320 and an embodiment of thehydraulic pump102 are discussed. Theflow regulator320, in combination with theforce modulator322 discussed below, comprise an embodiment of thedrive modulator40 of thecontrol system architecture8 described with reference to FIG.1. Theflow regulator320 comprises a plurality of functional blocks including summation junctions, integrators, and gain units. Theflow regulator320 comprises a proportional-integral (PI) controller portion, a command feed-forward portion, and a pump control gain portion. The general purpose of the PI controller portion is to correct the error between a sensed flow rate and a flow rate command by producing a corrective signal that tends to drive the sensed flow rate to the flow rate command. The general purpose of the command feed-forward portion is to provide a command feed-forward signal that comprises a substantial component of the drive signal to control a motor that drives thehydraulic pump102 based on the flow rate command, which may promote the PI controller portion being more suitably tuned to the purpose of correcting for dynamic transients and disturbances, such as changes of the physical system, sensors, or errors. Generally, in control systems analysis, disturbance terms correspond to imperfections or errors, for example extraordinary conditions, imperfect measurements or sensing of system parameters, irregularities such as bubbles in the hydraulic fluid, etc.

A sensedflow rate406 is negatively summed with aflow rate command402 by afirst summation junction408 to determine a flow rate error term. The sensedflow rate406 provides a negative feedback term. The output of thefirst summation junction408 is amplified by a firstproportional gain410. The output of thefirst summation junction408 is integrated by afirst integrator412 and then amplified by a firstintegral gain414. The output of the firstproportional gain410 and the output of the firstintegral gain414 are summed by asecond summation junction416. The

components

408,410,412,414, and416 comprise a PI controller portion. The output of thesecond summation junction416 may be viewed as a corrective signal that tends to drive the sensedflow rate406 to the value of theflow rate command402. In an embodiment, the position of thefirst integrator412 and the firstintegral gain414 may be reversed, and the flow rate error term, the output of thefirst summation junction408, may be first amplified by the firstintegral gain414 and then may be integrated by thefirst integrator412. In combination with the present disclosure, the values of the firstintegral gain414 and of the firstproportional gain410 may be readily determined by one skilled in the control systems art. The process of determining proportional, integral, and derivative gains in control systems is discussed in further detail hereinafter.

The command feed-forward signal is produced by amplifying theflow rate command402 by a first feed-forward gain418. The first feed-forward gain418 is inversely proportional to an estimate of the angular velocity or rate of rotation of the motor driving thehydraulic pump102. In an embodiment, the first feed-forward gain418 may be a constant value, for example a constant value proportional to the reciprocal of the designed steady-state angular velocity of the motor. In another embodiment, however, the first feed-forward gain418 may be determined based on the actual value of the angular velocity ω_Pof the motor, for example the value of motor angular velocity determined by a sensor. The command feed-forward signal is summed with the output of thesecond summation junction416 by thethird summation junction420. The output of thethird summation junction420 is amplified by apump control gain422 to produce acurrent command404. In combination with the present disclosure, thepump control gain422 may be readily determined by one skilled in the control systems art based on design data provided by the manufacturer of thehydraulic pump102. In an embodiment, thepump control gain422 may not remain constant and may vary with the value of the output of thethird summation junction420. In this case the flow regulator may determine the pump control gain422 using a look-up table, gain scheduling, or other function definition. In some contexts, the first feed-forward gain418 and the pump control gain may be referred to as a model or a portion of a model of theflow regulator320 or of thedrive modulator40.

In an embodiment, thehydraulic pump102 is an axial piston pump system308 comprising a motor that turns a swash plate that drives an axial piston pump. The axial piston pump system308 may comprise acurrent regulator430, a pumpswash plate actuator432, a pumprotation multiplication junction434, apump chamber436, and aflow sensor310. Thecurrent regulator430 outputs a control current to actuate an angle of displacement of the pumpswash plate actuator432 based on thecurrent command404 provided by theflow regulator320. The angle of displacement of the pumpswash plate actuator432 determines the amount of piston displacement as the pistons reciprocate within thepump chamber436.

Theswash plate actuator432 is illustrated as coupled to thepump chamber436 through the pumprotation multiplication junction434 illustrated as having a rotational input designated by ω_P. The piston displacement, in combination with the rate of rotation of the pump swash plate, determines the flow output of thehydraulic pump102. The outputs of thepump chamber436 include thepump pressure output440 and thepump flow output442. The inputs of thepump chamber436 include a pump backpressure444, and a pump inlet pressure P. Theflow sensor310 provides the sensedflow rate406.

In an embodiment, the value of the firstproportional gain410 and the value of the firstintegral gain414 are both set to zero, the output of thesecond summation junction416 is substantially zero, and the output of the third summation junction is substantially determined by the output of the first feed-forward gain418. In another embodiment, the

components

408,410,412,414, and416—the PI controller portion—are not part of theflow regulator320. In both these embodiments the sensedflow rate406 is not used to generate thecurrent command404. These two embodiments may be employed when there is noflow sensor310 available, when the sensedflow rate406 output by theflow sensor310 is unreliable or time lags the actual flow rate excessively, or when the pumpswash plate actuator432 has a low frequency response.

Turning now toFIG. 8, an embodiment of theforce modulator322 is discussed. Theforce modulator322, in combination with theflow regulator320 discussed above, comprise an embodiment of thedrive modulator40 of thecontrol system architecture8 described with reference toFIG. 1. The combination of theforce modulator322 and theflow regulator320 may be said to transform or map a manipulator force command and a manipulator velocity command to an actuator control signal. Theforce modulator322 comprises a proportional controller portion, a command feed-forward portion, and adirectional valve modulator514. The general purpose of the proportional gain controller is to correct the error between a travelinghead force command504 and a calculated or estimated force by producing a corrective signal that tends to drive the estimated force to the value of the travelinghead force command504. The calculated force is determined based on sensed or estimated pressures in the rodside pressure line112 and in the pistonside pressure line114. In an embodiment, it is assumed that the pressure sensed in the rodside pressure line112 is substantially the same as the pressure in therod side chamber120 and that the pressure sensed in the pistonside pressure line114 is substantially the same as the pressure in thepiston side chamber122. In another embodiment, however, pressure sensors may be placed in therod side chamber120 and in thepiston chamber122. The estimated force is negatively summed with the travelinghead force command504 by afourth summation junction506 to produce a force error term. The estimated force provides a negative feedback term. The output of thefourth summation junction506 is amplified by a secondproportional gain508. The

components

506 and508 comprise a proportional controller. In combination with the present disclosure, the value of the secondproportional gain508 may be readily determined by one skilled in the control systems art.

Theforce modulator322 also comprises thedirectional valve modulator514. Thedirectional valve modulator514 determines a directionalflow valve command530 based on the output of thefifth summation junction512. The directionalflow valve command530 may control thedirectional flow valve104 to direct the fluid flow to therod side chamber120, to thepiston side chamber122, or to thehydraulic return line108. In an embodiment, the directionalflow valve command530 may be mapped by a digital-to-analog converter device (not shown) to produce an electrical current to energize the first or the second solenoid to actuate the diverting spool to control thedirectional flow valve104. In an embodiment, thedirectional valve modulator514 outputs an observer feed forward flowrate command403. In an embodiment, thedirectional valve modulator514 also outputs a flow command that is summed with an estimate of the hydraulicfluid flow disturbance650 to produce theflow rate command402. The hydraulicfluid flow disturbance650 is discussed further below with reference toFIG. 10. The estimate of the hydraulicfluid flow disturbance650 may take account of fluid leakage past seals within thedirectional flow valve104 and/or other hydraulic seals in the hydraulic system. In another embodiment, however, thedirectional value modulator514 directly outputs theflow rate command402 and no hydraulic fluid flow disturbance term is considered.

Theforce modulator322 also comprises a portion for calculating an estimated manipulator force as the force produced by thepiston side chamber122 subtracted from the force produced by therod side chamber120. The estimatedpiston side pressure532 is amplified by a piston side area gain546 to determine the force produced by thepiston side chamber122 of the hydraulic actuators110-a,b. The estimatedrod side pressure534 is amplified by a rod side area gain554 to determine the force produced by therod side chamber120 of the hydraulic actuators110-a, b. Because in the subject embodiment two actuators110-a, bare employed, the area gains546,554 are represented as multiplying their respective areas by a factor of two. In some contexts, the area gains546,554 may be referred to as a model or a model portion of theforce modulator322 or of thedrive modulator40. Aseventh summation junction556 sums the negative value of the piston side force with the positive value of the rod side force to determine the estimated traveling head force F̂_TH. In another embodiment, other control structures may be employed to estimate force. In another embodiment, sensed values of rod side pressure and piston side pressure may be used instead of the estimatedrod side pressure534 and the estimatedpiston side pressure532, respectively.

Turning now toFIG. 9, an embodiment of the travelinghead controller326 is described. The travelinghead controller326 is one embodiment of themanipulator controller20 of thecontrol system architecture8 described above with reference toFIG. 1. The travelinghead controller326 comprises a proportional-integral-derivative (PID) controller portion and a command feed-forward portion. The travelinghead controller326 also comprises constants to compensate or offset the weight and the damping gain of the traveling

head

156,160.

An estimated travelinghead position610 is negatively summed with a travelinghead position command602 by aneighth summation junction608 to determine a traveling head position error term. The estimated travelinghead position610 provides a negative feedback term. The output of theeighth summation junction608 is amplified by a thirdproportional gain616. The output of theeighth summation junction608 is integrated by asecond integrator618 and amplified by a secondintegral gain620. In another embodiment, the output of theeighth summation junction608 is first amplified by the secondintegral gain620 and then integrated by thesecond integrator618. An estimated travelinghead velocity606 is negatively summed with the travelinghead velocity command502 by aninth summation junction612 to determine a traveling head velocity error term. The estimated travelinghead velocity606 provides a negative feedback term. The output of theninth summation junction612 is amplified by a firstderivative gain622. The processing of the traveling head velocity error term is considered to be a derivative component with respect to traveling head position because generally velocity is the derivative of position. In another embodiment, the sensed traveling head position and sensed traveling head velocity are used in place of the estimated travelinghead position610 and the estimated travelinghead velocity606, respectively. Atenth summation junction624 sums the outputs of the thirdproportional gain616, the secondintegral gain620, and the firstderivative gain622. The

components

608,612,616,618,620,622, and624 comprise a PID controller portion. The output of thetenth summation junction624 may be viewed as a corrective signal that tends to drive the estimated travelinghead position610 and the estimated travelinghead velocity606 to the values of the travelinghead position command602 and the travelinghead velocity command502. In combination with the present disclosure, the values of the thirdproportional gain616, the secondintegral gain620, and the firstderivative gain622 may readily be determined by one skilled in the control systems art.

A travelinghead acceleration command634 is amplified by a second feed-forward gain636 to produce a feed-forward signal. The second feed-forward gain636 is proportional to the estimated mass of the traveling head. The feed-forward term corresponds to a force term, because the product of an acceleration multiplied by a mass is equivalent to a force. The estimated travelinghead velocity606 is amplified by an estimated travelinghead damping gain640. The output of thetenth summation junction624 is summed with the output from the second feed-forward gain636, the output from the dampinggain640, and the negative of the estimated travelinghead weight638 by aneleventh summation junction641 to produce an observer feed forward travelinghead force command643. The output of theeleventh summation junction641 is summed with negative values of a feed forward workstring load command632 corresponding to the predicted weight of thework string152 on the traveling

head

156,160 and an estimated traveling headdisturbance force term644 by atwelfth summation junction642 to produce the travelinghead force command504. In some contexts, the travelinghead force command504 may be referred to more generally as a manipulator force command. In another embodiment, the estimated traveling headdisturbance force term644 is not available and hence is not summed by thetwelfth summation junction642. In combination with the present disclosure, one skilled in the art may readily determine the second feed-forward gain636 and the travelinghead damping gain640 by experimentally collecting data, for example velocity, position, acceleration, and/or pressure data, and fitting these gain values to this data, a technique which may be commonly performed by one skilled in the control systems art. In some contexts, the second feed-forward gain636 and the estimated travelinghead damping gain640 may be referred to as a model of the traveling head and/or the manipulator.

Turning now toFIG. 10, a block diagram of thepressure estimator324 is described. Thepressure estimator324 is one embodiment of thedrive observer50 of thecontrol system architecture8 described with reference toFIG. 1. Thepressure estimator324 comprises a directional valve command feedforward component646, a rodside pressure observer647, a pistonside pressure observer648, and a directional valvedisturbance flow component649. Thepressure estimator324 receives inputs from the directionalflow valve command530, the observer feed forward flowrate command403, the sensedrod side pressure652, the sensedpiston side pressure676, the estimated travelinghead position610, and the estimated travelinghead velocity606. In another embodiment, sensed values for traveling head position and traveling head velocity are used in place of the estimated travelinghead position610 and the estimated travelinghead velocity606, respectively. Thepressure estimator324 outputs theflow disturbance650, the estimatedpiston side pressure532, and the estimatedrod side pressure534.

The directional valve command feedforward component646 receives the directionalflow valve command530 and the observer feed forward flowrate command403 as inputs and outputs a rodside flow command667 and a pistonside flow command690. When the directionalflow valve command530 has a piston side value, thedirectional flow valve104 is selected to direct hydraulic fluid under pressure from the hydraulicpressure supply line106 to the piston sidehydraulic line114 into thepiston side chamber122 and to return hydraulic fluid from therod side chamber120 to the rod sidehydraulic line112 to thehydraulic return line107 to thehydraulic fluid reservoir108. When the directionalflow valve command530 has a piston side value, the directional valve command feedforward component646 determines the pistonside flow command690 to be proportional to the value of the observer feed forward flowrate command403 and the rodside flow command667 to be proportional to the negative value of the observer feed forward flowrate command403 multiplied by the constant determined as the effective surface area of thepiston118 in therod side chamber120 divided by the effective surface area of thepiston118 in thepiston side chamber122.

When the directionalflow valve command530 has a rod side value, thedirectional flow valve104 is selected to direct hydraulic fluid under pressure from the hydraulicpressure supply line106 to the rod sidehydraulic line112 into therod side chamber120 and to return hydraulic fluid from thepiston side chamber122 to the piston sidehydraulic line114 to thehydraulic return line107 to thehydraulic fluid reservoir108. When the directionalflow valve command530 has a rod side value, the directional valve command feedforward component646 determines the rodside flow command667 to be proportional to the value of the observer feed forward flowrate command403 and the pistonside flow command690 to be proportional to the negative value of the observer feed forward flowrate command403 multiplied by the constant determined as the effective surface area of thepiston118 in thepiston side chamber122 divided by the effective surface area of thepiston118 in therod side chamber120. This may be expressed symbolically by:

S*_DV=Piston Side

- Q*_PSαQ*_O
- Q*_RSα(−Q*_O)(A_RS/A_PS)

S*_DV=Rod Side

- Q*_PSα(−Q*_O)(A_PS/A_RS)
- Q*_RSαQ*_O
  In an embodiment, the constant of proportionality is unity, but in other embodiments a non-unity constant of proportionality may be used. In some embodiments, when the directionalflow valve command530 has a piston side value, the pistonside flow command690 is proportional to the value of the observer feed forward flowrate command403 and the rodside flow command667 is set to about zero; when the directionalflow valve command530 has a rod side value, the rodside flow command667 is proportional to the observer feed forward flowrate command403 and the pistonside flow command690 is set to about zero. The directional valve command feedforward component646 may be implemented as a software component, a function call, or a portion of a software program that executes on a processor of a general purpose computer.

The rodside pressure observer647 receives the sensedrod side pressure652 the rodside flow command667, the estimated travelinghead position610, and the estimated travelinghead velocity606 and outputs the estimatedrod side pressure534 and an estimated rodside flow disturbance645. The pistonside pressure observer648 receives the sensedpiston side pressure676 the pistonside flow command690, the estimated travelinghead position610, and the estimated travelinghead velocity606 and outputs the estimatedpiston side pressure532 and an estimated pistonside flow disturbance651. Both the rodside pressure observer647 and the pistonside pressure observer648 are discussed in greater detail hereinafter.

The directional valvedisturbance flow component649 receives the directionalflow valve command530, the estimated rodside flow disturbance645, and the estimated pistonside flow disturbance651 as inputs and outputs theflow disturbance650. When the directionalflow valve command530 has a piston side value, the directional valvedisturbance flow component649 sets the value of theflow disturbance650 to the value of the estimated pistonside flow disturbance651. When the directionalflow valve command530 has a rod side value, the directional valvedisturbance flow component649 sets the value of theflow disturbance650 to the value of the estimated rodside flow disturbance645. The directionalvalve disturbance component649 may be implemented as a software component, a function call, or a portion of a software program that executes on a processor of a general purpose computer.

Turning now toFIG. 11, an embodiment of the rodside pressure observer647 and the pistonside pressure observer648 are discussed. Each of the

pressure observers

647,648 have a similar structure and are each directed to producing an estimated pressure output and an estimated flow disturbance term output. In combination with the present disclosure, one skilled in the art will readily appreciate that an estimated parameter value, while it is related to a sensed parameter value, may be different from the sensed parameter value. For example, an estimated parameter value may be a smoothed or filtered version of the sensed parameter value that attenuates noise produced by a sensor or by an environment. Additionally, an estimated parameter value may partially reduce or remove phase shifts and/or time lags of system response. Each of the

pressure observers

647,648 includes a proportional-integral (PI) controller portion, a flow feed-forward portion, and a model portion. While the discussion below is based on inputting an estimated traveling head position Ẑ_THand an estimated traveling head velocity V̂_THto each of the

pressure observers

647,648, in another embodiment a sensed traveling head position Z_THand a sensed traveling head velocity V_THmay be input to each

pressure observer

647,648 in the place of the estimated traveling head position Ẑ_THand the estimated traveling head velocity V̂_TH.

An estimatedrod side pressure673 is negatively summed with a sensedrod side pressure652 by athirteenth summation junction654 to produce a rod side pressure error term. The estimatedrod side pressure673 is a negative feedback term. The sensedrod side pressure652 may be provided by the second pressure sensor312-b. The output of thethirteenth summation junction654 is amplified by a fourthproportional gain656. The output of thethirteenth summation junction654 is first integrated by athird integrator658 and then amplified by a thirdintegral gain660. In another embodiment, the output of thethirteenth summation junction654 may by first amplified by the thirdintegral gain660 and then integrated by thethird integrator658. The outputs of the fourthproportional gain656 and the thirdintegral gain660 are summed by afourteenth summation junction662. The

components

654,656,658,660, and662 comprise a PI controller portion.

In combination with the present disclosure, the values of the fourthproportional gain656 and the thirdintegral gain660 may be readily determined by one skilled in the control systems art. In an embodiment, the values of the fourthproportional gain656 and the thirdintegral gain660 are not constant but may vary based on the estimate of the traveling head position Ẑ_TH, which may be referred to as gain scheduling. In an embodiment, the gain schedule may be looked up in a table. In another embodiment, the gain schedule may be defined by a mathematical function dependent on the estimate of the traveling head position Ẑ_TH. In another embodiment, the gain schedule may be defined by another method. Gain scheduling may be useful for achieving stable and fast response across the operating range of theactuators110. For example, the gains may be based on the volume of therod side chamber120 which may vary substantially over the full range of travel of thepiston118.

The

components

654,656,658,660, and662 which comprise the PI controller portion may be viewed as a filtering and/or smoothing mechanism to attenuate noise associated with the second pressure sensor312-band/or the operating environment. The output of thefourteenth summation junction662 may be viewed as a corrective signal that tends to drive the estimatedrod side pressure673 to the sensedrod side pressure652. The output of thefourteenth summation junction662 may also be viewed as the estimated rodside flow disturbance645. In some contexts, the estimated rodside flow disturbance645 may be referred to as an estimated flow disturbance value that is a smoothed value and based on sensor information.

The flow feed-forward portion produces a flow rate by amplifying the estimated travelinghead velocity606 of the manipulator by a rodside area gain666. This flow rate is associated with the flow of hydraulic fluid into and out of therod side chamber120 as the manipulator moves in the Z-axis of motion. The value of the rodside area gain666 is proportional to the effective surface area of the rod side of the piston, the area of thepiston118 compensated for by the area of therod116 as described above. When twohydraulic actuators110 are employed, for example the first and second hydraulic actuators110-a, b, then the area gain may be doubled. The output of the rodside area gain666 is negatively summed with the output of thefourteenth summation junction662 and with a rodside flow command667 by afifteenth summation junction668. The effect of the summation of the negative of therod side gain666, the rodside flow command667, and the output from thefourteenth summation junction662 is to produce through the model a filtered pressure related value that is zero time lagged or zero phase lagged with respect to the sensedrod side pressure652.

The output of thefifteenth summation junction668 is integrated by afourth integrator672 and then amplified by a rodside model gain670. In another embodiment, the output of thefifteenth summation junction668 may be first multiplied by the rodside model gain670 and then integrated by thefourth integrator672. The value of the rodside model gain670 is proportional to K̂_oil, the estimated bulk modulus of the hydraulic fluid, and is inversely proportional to {circumflex over (∇)}_RS, the estimated volume of therod side chamber120 for both the first and second hydraulic actuators110-a, band the hydraulic lines, for example the rod sidehydraulic line112 associated with the first and second hydraulic actuators110-a, b. Because the volume of therod side chamber120 varies with the estimated traveling head position Ẑ_TH, in an embodiment the value of the rodside model gain670 may vary and may be determined using gain scheduling in a manner similar to that discussed above.

The estimated bulk modulus of the hydraulic fluid, K̂_oil, is an indication of the compressibility of the hydraulic fluid. In an embodiment, the value of the estimated hydraulic bulk modulus K̂_oilmay be about 180,000 pounds per square inch (PSI), but in other embodiments and in using other hydraulic fluids a different estimated bulk modulus K̂_oilmay be used. The volume of therod side chamber120, in an embodiment, may be calculated from an about ten foot stroke and an about seven and one half inch bore or diameter. In other embodiments, different dimensions of therod side chamber120 may be appropriate. The calculation of the volume of therod side chamber120 should take account of the volume consumed by therod116. In combination with the present disclosure, the value of the rodside model gain670 may be readily determined by one skilled in the control systems art.

The integration of the flow term and multiplying through by the rodside model gain670 has the effect of transforming the flow term output by thefifteenth summation junction668 into a pressure term. The output of the rodside model gain670 is the estimatedrod side pressure673.

An estimatedpiston side pressure697 is negatively summed with a sensedpiston side pressure676 by asixteenth summation junction678 to produce a piston side pressure error term. The estimatedpiston side pressure697 is a negative feedback term. The sensedpiston side pressure676 may be provided by the first pressure sensor312-a. The output of thesixteenth summation junction678 is amplified by a fifthproportional gain680. The output of thesixteenth summation junction678 is integrated by afifth integrator682 and then amplified by a fourthintegral gain684. In another embodiment, the output of thesixteenth summation junction678 may be first amplified by the fourthintegral gain684 and then integrated by thefifth integrator682. The outputs of the fifthproportional gain680 and the fourthintegral gain684 are summed by aseventeenth summation junction686. The

components

678,680,682,684, and686 comprise a proportional-integral (PI) controller portion. In combination with the present disclosure, the values of the fifthproportional gain680 and the fourth integral gain584 may be readily determined by one skilled in the control systems art. In an embodiment, for the same reasons discussed above with respect to the rodside pressure observer647, the values of the fourthproportional gain680 and the fourthintegral gain684 may not be constant but may be determined by gain scheduling, by mathematical function dependent on the estimate of the traveling head position Ẑ_TH, or by some other method.

The

components

678,680,682,684, and686 which comprise the PI controller portion may be viewed as a filtering and/or smoothing mechanism to attenuate noise associated with the first pressure sensor312-aand/or the operating environment. The output of theseventeenth summation junction686 may be viewed as a corrective signal that tends to drive the estimatedpiston side pressure697 to the sensedpiston side pressure676. The output of theseventeenth summation junction686 may also be viewed as an estimated pistonside flow disturbance651. In some contexts, the estimated pistonside flow disturbance651 may be referred to as an estimated flow disturbance value that is a smoothed value and based on sensor information.

The command feed-forward portion produces a flow rate by amplifying the estimated travelinghead velocity606 of the manipulator by a pistonside area gain688. This flow is associated with the flow of hydraulic fluid into and out of thepiston side chamber122 as the manipulator moves in the Z-axis of motion. The value of the pistonside area gain688 is proportional to the area of thepiston118. When twohydraulic actuators110 are employed, for example the first and second hydraulic actuators110-a, b, then the pistonside area gain688 may be doubled. The output of the pistonside area gain688 is positively summed with the output of theseventeenth summation junction686 and with a pistonside flow command690 by aneighteenth summation junction692. The sense of summing of the traveling head velocity and area product is different for thepiston pressure observer648 versus the rodside pressure observer647 because the direction of traveling head motion has opposite flow effects on therod side chamber120 and thepiston side chamber122. The effect of the summation of the output of the pistonside area gain688, the pistonside flow command690, and the output from theseventeenth summation junction686 is to produce through the model a filtered pressure related value that is zero time lagged or zero phase lagged with respect to the sensedpiston side pressure676.

The output of theeighteenth summation junction692 is integrated by asixth integrator696 and then amplified by a pistonside model gain694. In another embodiment, the output of theeighteenth summation junction692 may be first amplified by the pistonside model gain694 and then integrated by thesixth integrator696. The value of the piston side model gain is proportional to K̂_oil, the estimated bulk modulus of the hydraulic fluid, and is inversely proportional to {circumflex over (∇)}_PS, the estimated volume of thepiston side chamber122 for both the first and second hydraulic actuators110-a, band the hydraulic lines, for example the piston sidehydraulic line114 associated with the first and second hydraulic actuators110-a, b. Because the volume of thepiston side chamber122 varies with the estimated traveling head position Ẑ_TH, in an embodiment the value of the pistonside model gain694 may be determined using gain scheduling in a manner similar to that discussed above.

The integration of the flow term and multiplying through by pistonside model gain694 has the effect of transforming the flow term output by theeighteenth summation junction692 into a pressure term. The output of the pistonside model gain694 is the estimatedpiston side pressure697.

In some contexts, the model portion of the rodside pressure observer647 may be considered to comprise the rodside area gain666 and the rodside model gain670, and these gains are based on the estimated effective rod side piston area, the estimated hydraulic fluid bulk modulus, and the estimated variable rod side chamber volume. In some contexts the model portion of the pistonside pressure observer648 may be considered to comprise the pistonside area gain688 and the pistonside model gain694, and these gains are based on the estimated effective piston side piston area, the estimated hydraulic fluid bulk modulus, and the estimated variable piston side chamber volume. In some contexts, the model portion of the rodside pressure observer647 and the model portion of the pistonside pressure observer648 may be referred to as a model of thehydraulic actuator110 with reference toFIG. 2A or of the hydraulic actuators110-a,110-bwith reference toFIG. 2B.

Turning now toFIG. 12, an embodiment of the traveling head position andvelocity observer328 is described. The traveling head position andvelocity observer328 is one embodiment of themanipulator observer30 of thecontrol system architecture8 described with reference toFIG. 1. The traveling head position andvelocity observer328 may be referred to in some contexts as a manipulator position and velocity observer. The traveling head position andvelocity observer328 provides the estimate of travelinghead position610, the estimate of travelinghead velocity606, and the estimate traveling headdisturbance force term644. In combination with the present disclosure, one skilled in the art will readily appreciate that an estimated parameter value, while it is related to a sensed parameter value, may be different from the sensed parameter value. For example, an estimated parameter value may be a smoothed or filtered version of the sensed parameter value that attenuates noise produced by a sensor or by an environment. Additionally, an estimated parameter value may partially reduce or remove phase shifts and/or time lags of system response. The traveling head position andvelocity observer328 comprises a proportional-integral-derivative (PID) controller portion, a feed-forward portion, and a model portion.

The estimated travelinghead position610 is negatively summed with a sensed travelinghead position730 by anineteenth summation junction734 to determine a sensed traveling head position error. The estimated travelinghead position610 provides a negative feedback term. The output of thenineteenth summation junction734 is amplified by a sixthproportional gain736. The output of thenineteenth summation junction734 is integrated by aseventh integrator738 and then amplified by a fifthintegral gain740. In another embodiment, the output of thenineteenth summation junction734 may be first amplified by the fifthintegral gain740 and then may be integrated by theseventh integrator738. The estimated travelinghead velocity606 is negatively summed with a sensed travelinghead velocity732 by atwentieth summation junction742 to determine a sensed traveling head velocity error. The estimatedvelocity606 provides a negative feedback term. The output of thetwentieth summation junction742 is amplified by a secondderivative gain744. The processing of the sensed traveling head velocity error term is considered to be a derivative component with respect to the sensed traveling head position because generally velocity is the derivative of position. A twenty-first summation junction746 sums the outputs of the sixthproportional gain736, the fifthintegral gain740, and the secondderivative gain744. The

components

734,736,738,740,742,744, and746 comprise a PID controller portion. In combination with the present disclosure, the values of the sixthproportional gain736, the fifthintegral gain740, and the secondderivative gain742 may be readily determined by one skilled in the control systems art.

The

components

734,736,738,740,742,744, and746 which comprise the PID controller portion may be viewed as a filtering and/or smoothing mechanism to attenuate noise associated with the sensed travelinghead position730, the sensed travelinghead velocity732, and/or the operating environment. The output of the twenty-first summation junction746 may be viewed as a corrective signal that tends to drive the estimated travelinghead position610 and estimated travelinghead velocity606 to the sensed travelinghead position730 and the sensed travelinghead velocity732 values. The output of the twenty-first summation junction746 may also be viewed as the estimated traveling headdisturbance force term644, an estimate of the un-modeled disturbances acting on the traveling head physical system. In some contexts, the estimated traveling headdisturbance force term644 may be referred to as an estimated disturbance force feedback value that is a smoothed and based on actuator position sensor information.

A twenty-second summation junction750 sums the output of the twenty-first summation junction746, the observer feed forward travelinghead force command643, an estimated travelinghead weight752, the negative of the output of the estimated travelinghead friction component760, and the negative of the output of the estimated travelinghead damping gain762. The effect of the summation of the output of the twenty-first summation junction746, the observer feed forward travelinghead force command643, the estimated travelinghead weight752, the negative of the output of the estimated travelinghead friction component760, and the negative of the output of the estimated travelinghead damping gain762 is to produce through the model a filtered velocity and position related value that is zero time lagged or zero phase lagged with respect to the sensed travelinghead position730 and the sensed travelinghead velocity732.

The estimated travelinghead friction component760 and the estimated travelinghead damping gain762 are provided to take into account friction and damping effects that may be present in the mechanical structure associated with the hydraulic actuators110-a, b, for example mechanical rails or guides which substantially constrain the hydraulic actuators110-a, bto motion in a single Z-axis. The estimated travelinghead friction component760 is a function of the polarity of the estimated traveling head velocity758. The damping effect is modeled as the estimated traveling head velocity758 amplified by the estimated travelinghead damping gain762. In combination with the present disclosure, the estimated travelinghead damping gain762 and the estimated travelinghead friction component760 may be determined and/or tuned by one skilled in the art by collecting data during experimental operation of the control system and fitting the traveling head model to the data. The value of the estimated travelinghead weight752 may be determined by weighing the traveling

head

156,160, by taking the net weight identified on a specification provided by a manufacturer of the traveling

head

156,160, or by some other known manner. The output of the twenty-second summation junction750 corresponds to a force term. The output of the twenty-second summation junction750 is amplified by an inverse estimated travelinghead mass gain766. The value of the inverse estimated travelinghead mass gain766 is inversely proportional to the mass of the traveling head, and the mass of the traveling

head

156,160 may be determined from the weight of the traveling

head

156,160.

The output of the inverse estimated travelinghead mass gain766 is an acceleration term. The output of the inverse travelinghead mass gain766 is integrated by an eighth integrator768. The output of the eighth integrator768 is a velocity term, the estimated travelinghead velocity606, because generally the integration of an acceleration produces a velocity. The output of the eighth integrator768 is integrated by aninth integrator770. The output of theninth integrator770 is a position term, the estimated travelinghead position610, because generally the integration of a velocity produces a position.

In some contexts, the estimated travelinghead weight752, the estimated travelinghead friction component760, the estimated traveling head damping gain, and inverse travelinghead mass gain766 may be referred to as a model of the traveling

head

156,160, as a model of the traveling

head

156,160 and associated

slip bowl

154,158, or as a model of a force coupling component. In some contexts, the traveling

head

156,160 may be referred to as a force coupling component, e.g., a force coupling component that couples the force output by theactuators110 to one of the

slip bowl

154,158. In some cases the model of the traveling

head

156,160 may take into account or include the

slip bowl

154,158. In some contexts, the estimated travelinghead friction component760 may be referred to as a force coupling friction component, the estimated travelinghead weight752 may be referred to as an estimated weight of the force coupling component, the estimated traveling head damping gain may be referred to as an estimated damping factor of the force coupling component, and the inverse travelinghead mass gain766 may be associated with an estimated mass of the force coupling component.

Turning now toFIG. 13, an embodiment of awork string controller800 is discussed. Thework string controller800 is one embodiment of thesystem controller10 of thecontrol system architecture8 described with reference toFIG. 1. Thework string controller800 is configured to control one or moremanipulator control systems300, for example a first manipulator control system300-a, a second manipulator control system300-b, and a third manipulator control system300-c, whereby to control thework string152. In an embodiment, thework string controller800 controls thework string152 through controlling the dual-jackingsystem150. In this embodiment, the first manipulator control system300-acontrols the first and second hydraulic actuators110-a, band thefirst slip bowl154, the second manipulator control system300-bcontrols the third and fourth hydraulic actuators110-c, dand thesecond slip bowl158, and the third manipulator control system300-bcontrols the third orstationary slip bowl162. In another embodiment, thework string controller800 may control thework string152 by another means, for example using more than two pairs ofhydraulic actuators110 or for example by using a different kind of actuator. In an embodiment, thework string controller800 comprises a simulatedforce feedback section802, amodel controller804, and a manipulator commandsgenerator806.

Thework string controller800 receives commanded work string values, for example a workstring position command818 and a workstring velocity command828, as control inputs. These commanded work string values818,828 may come from a user interface or an operator control station. Thework string controller800 also receives a plurality of sensed and commanded values from components that are part of thecontrol system architecture8. Thework string controller800 receives sensed or estimated values of manipulator position Z_TH1, Z_TH2and manipulator velocity V_TH1, V_TH2, for example from the traveling head position sensors314a, b, the traveling head velocity sensors316a, bor from the traveling head position andvelocity observer328. Thework string controller800 receives state value of slip bowls S_SB1, S_SB2, S_SB3from the slip bowls154,158,162. The sensed or estimated values of position and velocity and of slip bowl state may be considered to be feedback to thework string controller800 from the manipulator physical systems. Thework string controller800 outputs commanded values of manipulator position Z_TH1*, Z_TH2*, manipulator velocity V_TH1*, V_TH2*, manipulator acceleration A_TH1*, A_TH2*, workstring force F_WS/TH1*, F_WS/TH2*, and slip bowl state S_SB1*, S_SB2*, S_SB3*, for example to the travelinghead controller326 and to theslip bowl154. At a high level, thework string controller800 takes into account the command inputs and the feedback inputs to develop commands that are output to themanipulator controllers300.

Themodel controller804 comprises a PID controller section and a work string model section. The intention of themodel controller804 is to produce an estimatedwork string position840 and an estimatedwork string velocity842.

The estimatedwork string position840 is negatively summed with the workstring position command818 by a twenty-fourth summation junction820 to produce a position error term. The estimatedwork string position840 provides a negative feedback term. The output of the twenty-fourth summation junction820 is amplified by a seventhproportional gain821. The output of the twenty-fourth summation junction820 is integrated by atenth integrator822 and amplified by a sixthintegral gain823. In another embodiment, the output of the twenty-fourth summation junction820 may be first amplified by the sixthintegral gain823 and then integrated by thetenth integrator822. The estimatedwork string velocity842 is negatively summed with the workstring velocity command828 by a twenty-fifth summation junction830 to produce a velocity error term. The estimatedwork string velocity842 provides a negative feedback term. The output of the twenty-fifth summation junction830 is amplified by a thirdderivative gain831. The processing of the velocity term is considered to be a derivative component with respect to the work string position because generally velocity is the derivative of position. The output of the twenty-third summation junction814 is negatively summed with the output of the seventhproportional gain821, the output of the sixthintegral gain823, and the output of the thirdderivative gain831 by a twenty-sixth summation junction832. The output of the twenty-third summation junction814 effectively couples position and velocity feedback, in the form of simulated force feedback of the work string on themanipulators812, from the manipulators into the estimation of work string position and velocity. In an embodiment, the general intention is that if one of the manipulators is not achieving the targeted manipulator position, the simulated force feedback term grows large, and the position commands to the other non-lagging manipulators are adapted accordingly. This may promote better synchronization among manipulators when a heavy load or an operational anomaly occurs.

In another embodiment, sensed values of work string position and work string velocity may be fed directly to the manipulator commandsgenerator802. Alternatively, in another embodiment, a sensed value of the work string position and work string velocity may be available, the sensed value of work string position may be substituted for the estimatedwork string position840 input to the twenty-fourth summation junction820, and the sensed value of work string velocity may be substituted for the estimatedwork string velocity842 input to the twenty-fifth summation junction830.

The output of the twenty-sixth summation junction832 is amplified by amodel gain834. Themodel gain834 is inversely proportional to the estimated mass of the work string. During well bore servicing operations the mass of the work string may change, for example as joints of pipe are added to or removed from the work string, and in an embodiment, themodel gain834 may be a changing value rather than a static value. In another embodiment, however, themodel gain834 may be set to a static value associated with a static estimated mass of the work string. The output of themodel gain834 is integrated by aneleventh integrator836 to produce the estimatedwork string velocity842. The output of theeleventh integrator836 is integrated by atwelfth integrator838 to produce the estimatedwork string position840.

In an embodiment, the manipulator commandsgenerators806 may be selected to operate in a plurality of operation modes for controlling the first, second, and third manipulators cooperatively. A first mode may be a distributed load or high capacity sequential mode, where the first and second manipulators are controlled to concurrently grip thework string152 and to apply substantially equal force to thework string152 in the same direction at the same time. A second mode may be a high speed sequential mode, where the first and second manipulators trade off gripping thework string152 and applying force to thework string152 in a sequence of start and stop motions. The high speed sequential mode could also be extended to include all three manipulators. A third mode may be a high speed continuous mode, where the first and second manipulators trade off gripping thework string152 and applying force to thework string152 in a manner which constrains work string motions to substantially constant work string velocity or substantially constant work string acceleration or deceleration independent of manipulator trajectories. A fourth mode may be a high speed constrained mode, where the first and second manipulators trade off gripping thework string152 and applying force to thework string152 in a manner which is constrained by a maximum hand-off speed between the manipulators. In other embodiments with multiple manipulators and constraints, combinations of these modes are contemplated by the present disclosure. A user interface (not shown), for example a control panel, may be used to select the operation mode of the manipulator commandsgenerator806 and to input the workstring position command818 and the workstring velocity command828. In another embodiment, the mode of operation may be chosen automatically as the system conditions change, including but not restricted to the changing weight of thework string152. In response to the operation mode selection, the estimatedwork string position840, and the estimatedwork string velocity842, the manipulator commandsgenerators806 may employ an internal map or table or program to generate the operation mode specific manipulator controllercommand input vectors340. In an embodiment, apipe collar indication844, which may be provided by a collar locator such as that shown inFIG. 16, is also input to the manipulator commandsgenerator806. The manipulator commandsgenerator806 may employ thepipe collar indication844 to avoid commanding the first and second slip bowls154,158 and thestationary slip bowl162 to close on a pipe collar, a situation which would prevent the first and second slip bowls154,158 and thestationary slip bowl162 from closing and gripping thework string152 securely.

The manipulator commandsgenerator806 generates the firstslip bowl command214, the first traveling head velocity command502-a, the first traveling head position command602-a, the first traveling head acceleration command634-a, a first feed forward work string load command632-a, the secondslip bowl command220, the second traveling head velocity command502-b, the second traveling head position command602-b, the second traveling head acceleration command634-b, a second feed forward work string load command632-b, and the stationaryslip bowl command228.

Turning now toFIG. 14, an exemplary manipulatorforce feedback component810 is discussed. The manipulatorforce feedback component810 comprises a PID controller section, a slip bowl model section, and a slip bowl model PI controller section. The sensed travelinghead position730 is negatively summed with the commanded travelinghead position602 by a twenty-seventh summation junction860 to produce a position error term. The sensed travelinghead position730 provides a negative feedback term. The output of the twenty-seventh summation junction860 is amplified by an eighthproportional gain861. The output of the twenty-eighth summation junction860 is integrated by athirteenth integrator862 and amplified by a seventhintegral gain863. In another embodiment, the output of the twenty-eighth summation junction860 may be first amplified by the seventhintegral gain863 and then may be integrated by thethirteenth integrator862. The sensed travelinghead velocity732 is negatively summed with the commanded travelinghead velocity502 by a twenty-eighth summation junction864 to produce a velocity error term. The sensed travelinghead velocity732 provides a negative feedback term. The output of the twenty-eighth summation junction864 is amplified by a fourthderivative gain865. The processing of the traveling head velocity term is considered to be a derivative component with respect to the traveling head position because generally velocity is the derivative of position. The output of the eighthproportional gain861, the seventh integral gain, and the fourthderivative gain865 are summed by a twenty-ninth summation junction866 to produce an estimated work string force feedback on the travelinghead867. In combination with the present disclosure, the values of the eighthproportional gain861, the seventhintegral gain863, and the fourthderivative gain864 may be readily determined by one skilled in the control systems art.

The

slip bowl command

214,220,226 is processed by aslip bowl model870 to produce an estimatedslip bowl position871. In an embodiment, theslip bowl model870 may comprise a simple time delay to model the time it takes for the physical slip bowl system to follow the slip bowl command under normal situations. In another embodiment, the sensed slip bowl position and the

slip bowl command

214,220,226 may relate to a continuous slip bowl position from being fully disengaged to being fully engaged, where theslip bowl model870 may become unnecessary. The sensedslip bowl position216 is negatively summed with the estimatedslip bowl state871 by athirtieth summation junction872. The output of thethirtieth summation junction872 is amplified by a ninthproportional gain876. The output of thethirtieth summation junction872 is integrated by afourteenth integrator877 and amplified by an eighthintegral gain878. In another embodiment, the output of thethirtieth summation junction872 may first be amplified by the eighthintegral gain878 and then integrated by thefourteenth integrator877. The output of the ninthproportional gain876 and the output of the eighthintegral gain878 are summed by a thirty-first summation junction879 to produce an estimated work string force feedback on the slip bowl. In an embodiment, the value of the eighthintegral gain878 may be set to a substantially higher value than that of the seventhintegral gain863, so the estimated work string force feedback on the slip bowl may ramp up more rapidly than the estimated work string force feedback on the traveling head. In an embodiment, the eighthintegral gain878 may be about five times larger than the seventhintegral gain863. In another embodiment, the eighthintegral gain878 may be about ten times larger than the seventhintegral gain863. In yet another embodiment, the eighthintegral gain878 may be about fifty times larger than the seventhintegral gain863. For example, if the physical slip bowl system does not follow the estimated slip bowl state, for example if the

slip bowl

154,158,162 does not close because it has erroneously attempted to close on a pipe collar, it is desirable for the estimated work string force feedback on the slip bowl to rapidly increase in value to cause the several manipulators to come to a stop or even reverse direction momentarily. In combination with the present disclosure, the values of the ninthproportional gain876 and the eighthintegral gain878 may be readily determined by one skilled in the control systems art.

The output of the twenty-ninth summation junction866 and the thirty-first summation junction879 are summed by a thirty-second summation junction884 to determine a simulated force feedback of the work string on the manipulator.

The proportional, integral, and derivative gains in the several control components described above may be determined by an initial calculation based on a pole-zero system stability analysis and a frequency response analysis of the control system and later the gains may be further adjusted when deploying the control system to actual use. The general design approach is to design the control system based on a frequency domain mathematical model, for example expressed using Laplace transforms. This mathematical model includes both control elements and physical elements of the control system. Themanipulator control system300 and thework string controller800 controlling two or moremanipulator control systems300 may be implemented as hierarchical control systems, wherein the control feedback loops are closed and tuned in order from the lowest level or innermost loop to the highest level or outermost loop. The gains at each level may be tuned to the fastest frequency response which can be attained while maintaining a stable system with good disturbance rejection. In a work string controller system such as that disclosed herein, large disturbances are possible, for example large manipulator force transients associated with portions of the well bore collapsing in on thework string152 and increasing the friction experienced by thework string152 moving in the well bore. Therefore, it may be prudent to employ reasonable safety factors in setting the design values for disturbances that may be experienced when tuning the control gains. According to a generally accepted practice, when control loops are placed around other control loops, as is the case with thework string controller800 and themanipulator control system300, the outer loop gains may be designed to achieve about one quarter the frequency response of the next lower control loop. In some embodiments, however, this general practice may not be adhered to.

While the figures illustrate the control components in terms of the Laplace transform, which is a continuous time transform, one skilled in the control systems art could readily adapt the continuous time illustrations to discrete time illustrations. The principal modification would be replacing the continuous time Laplace transform of the integration operation, represented as 1/S in the several integrator blocks, by T_s/(Z−1), where Ts is a constant proportional to the control system data sampling interval and 1/(Z−1) is the discrete time Z-transform of the integration operation.

As would be appreciated by those of ordinary skill in the control systems art, some portions of the above described control system components may be combined or separated without materially altering the function of the control system. For example, other combinations of summation junctions may be employed to build theflow regulator320 without materially changing the control function of theflow regulator320. For example, in another embodiment, the summation provided by thesecond summation junction416 and thethird summation junction420 may be performed by a single summation junction without materially altering the control function of theflow regulator320. Additionally, the order of some portions of the above described control components may be changed without materially changing the function of the control system. For example, reversing the positions of thefirst integrator412 with the firstintegral gain414 in theflow regulator320 may not materially alter the function of theflow regulator320. All such trivial combinations and positional rearrangements that do not materially alter the control function of the controllers and control components described above are contemplated by this disclosure.

Turning now toFIG. 15, some kinematic parameters associated with a twelve stage traveling head trajectory are illustrated. The illustrated kinematic parameters provide one possible model of kinematic parameters associated with thecontrol system architecture8 depicted inFIG. 1, for example kinematic parameters associated with the dual-jackingsystem150 depicted inFIG. 3. The kinematic equations disclosed below, including equation (1) through equation (48), provide exemplary kinematic equations that may be employed by the manipulator commandsgenerators806 in part to generate the manipulator commands in one mode of operation, for example in a high speed continuous mode of operation. In other modes of operation, other kinematic parameters and other kinematic equations may apply. In combination with the present disclosure, one skilled in the art may make appropriate modifications and extensions of these kinematic parameters and associated kinematic equations to apply this information to other related control systems and/or manipulator systems. Table 1 identifies some trajectory stage lengths and describes the stages. Table 2 identifies some trajectory parameters.

TABLE 1

Trajectory Stage Lengths

L₁	0 in	Stage 1 - Maximum Acceleration to Match
		Pipe Speed
L₂	?	Stage 2 - Slip Bow Engagement (Hand Off)
		for Fully Loaded Single
L₃	?	Stage 3 - Accelerate for Hand Off
L₄	?	Stage 4 - Relative Jack Velocity for Slip
		Bowl Disengagement (Hand
L₅	?	Stage 5 - Accelerate Away from Pipe
L₆	?	Stage 6 - Potential Maximum Velocity Achieved
L₇	?	Stage 7 - Decelerate for Return Stroke
L₈	L_s	Stage 8 - Zero Velocity Dwell for Directional
		Valve Switch
L₉	L_s	Stage 9 - Maximum Acceleration on Return Stroke
L₁₀	?	Stage 10 - Potential Maximum Velocity Achieved
L₁₁	?	Stage 11 - Maximum Deceleration on Return Stroke
L₁₂	0 in	Zero Velocity Dwell for Directional Valve Switch

TABLE 2

Trajectory Parameters

D_HO	?	Hand Off Length
D_SJ	?	Fully Loaded Single Jack Portion of Stroke Length
D_SB	?	Distance Traveled during Slip Bowl Response Time t_SB
D_slip	1 in	Relative Distance for Slip Bowl Disengagement w.r.t. Pipe
F_s	6	Slip Insurance Safety Factor (Shutdown Conditional)
Q_max	60	Maximum Output Flow rate per Pump
	GPM
		Maximum Output Flow Rate of Change per Pump
V_dn	?	Maximum Achievable Velocity Downward
V_up	?	Maximum Achievable Velocity Upward
A_dn	?	Maximum Achievable Acceleration Downward
A_up	?	Maximum Achievable Acceleration Upward

v₁=0 (1)

L₁=0 (2)

\begin{matrix} L_{2} - L_{1} = \frac{1}{2} a_{1} t_{1}^{2} & (3) \end{matrix}

v₂=a₁t₁ (4)

v₂=v_P (5)

a₂=0 (6)

L₃−L₂=D_SJ+D_HO+D_SB (7)

L₃−L₂=v₂t₂ (8)

L₅−L₃=D_HO (9)

v₃=v₂ (10)

\begin{matrix} L_{4} - L_{3} = v_{3} t_{3} + \frac{1}{2} a_{3} t_{3}^{2} & (11) \end{matrix}

v₄=v₃+a₃t₃ (12)

a₃=a_DN_—_MAX (13)

D_slip=(L₅−L₃)−v_P(t₃+t₄) (14)

L₅−L₄=V₄t₄ (15)

a₄=0 (16)

v₅=v₄ (17)

\begin{matrix} L_{6} - L_{5} = v_{5} t_{5} + \frac{1}{2} a_{5} t_{5}^{2} & (18) \end{matrix}

L₆−L₅=D_slipF_s (19)

v₆=v₅+a₅t₅ (20)

t₆=0 (21)

a₆=0 (22)

L₇=L₆ (23)

L₂=L₈−L₅ (24)

v₇=a₇t₇ (25)

v₇=v₆ (26)

t₇=NT_s (27)

\begin{matrix} L_{8} - L_{7} = v_{7} t_{7} + \frac{1}{2} a_{7} t_{7}^{2} & (28) \end{matrix}

v₁₀=a₉t₉ (33)

v₁₀=−v_UP_—_MAX (34)

a₉=−a_UP_—_MAX (35)

L₉=L₈ (36)

\begin{matrix} L_{10} - L_{9} = \frac{1}{2} a_{9} t_{9}^{2} & (37) \end{matrix}

L₁₁−L₁₀=v₁₀t₁₀ (38)

L₁₁=L₈−L₁₀ (39)

a₁₀=0 (40)

v₁₁=v_UP_—_MAX (41)

t₁₁=t₉ (42)

a₁₁=a_UP_—_MAX (43)

v₁₂=0 (44)

t₁₂=T_switch (45)

a₁₂=0 (46)

L₁₂=0 (47)

t₂=t₁+t₃+t₄+t₅+t₆+t₇+t₈+t₉+t₁₀+t₁₁+t₁₂ (48)

Unknowns:

L₁, L₂, L₃, L₄, L₅, L₆, L₇, L₈, L₉, L₁₀, L₁₁, L₁₂

t₁, t₂, t₃, t₄, t₅, t₆, t₇, t₈, t₉, t₁₀, t₁₁t₁₂

v1, v2, v3, v4, v5, v6, v7, v8, v9, v10, v11, v12

a₁, a₂, a₃, a₄, a₅, a₆, a₇, a₈, a₉, a₁₀, a₁₁, a₁₂

Turning now toFIG. 16, an embodiment of acollar locater900 is described. In an embodiment, thecollar locator900 comprises a firstdigital camera908, a firstlight source914, a seconddigital camera918, a secondlight source920, and acollar detector922. Thework string152 includes afirst collar902 and asecond collar904. Collars in thework string152 may be formed where two ends of pipes are threaded into each other. The

collars

902 and904 have a larger diameter and a broader profile than the rest of the pipe sections when viewed by the first and second

digital cameras

908,918. The first and second

digital cameras

908,918 provide images of thework string152 to thecollar detector922 which processes these images to identify collars. Thecollar detector922 outputs acollar indication844. In an embodiment, thecollar indication844 may indicate when a collar is in view of the firstdigital camera908, when a collar is in view of the seconddigital camera918, and/or the estimated position of one or more collars.

In an embodiment, the digital image of a section of thework string152 may be thresholded by thecollar detector922 to distinguish thework string152 from the background. Thresholding is an image processing technique that designates pixels as object pixels or background pixels based on comparing the value of the pixels to a threshold value, for example a grayscale value. In different embodiments, object pixels may be either lighter or darker than the threshold value. After thresholding, an edge detection computer vision algorithm may be applied by thecollar detector922 to determine the approximate width of the profile of the portion of thework string152 within the image. If the width of thework string152 exceeds a limit, thecollar detector922 determines that a collar is in the image.

In an embodiment, the first and second

digital cameras

908, and918 may capture images of thework string152 at a sample rate of 30 times per second or greater. In this embodiment, thecollar locator900 may be able to determine a speed of thework string152 and to provide an estimate of the location of thefirst collar902 and thesecond collar904 after they have passed out of the view field of the first and second

digital cameras

908,918 based on the calculated work string velocity.

Turning now toFIG. 17, amethod950 of controlling thework string152 is described. Inblock951, the work string is placed in a well bore. Inblock952, a work string trajectory input is received. The work string trajectory characterizes a desired motion of thework string152 as one or more linked pairs of position and velocity, or work string target position and work string velocity target. These may be referred to as an ordered sequence of target pairs, wherein each target pair comprises a work string target position and a work string target velocity. In a first example, the work string trajectory input may include one linked pair of position and velocity having a position target of 5000 feet and a velocity target of 5 feet per second. In a second example, the work string trajectory input may include six linked pairs of position and velocity—a first position target of 5010 feet and a first velocity target of 5 feet per second, a second position target of 4990 feet and a second velocity target of −20 feet per second, a third position target of 5010 feet and a third velocity target of 20 feet per second, a fourth position target of 4990 feet and a fourth velocity target of −20 feet per second, a fifth position target of 5010 feet and a fifth velocity target of 20 feet per second, and a sixth position target of 0 feet and a sixth velocity target of −5 feet per second. This second example corresponds to lowering thework string152 into the well bore to a depth of about 5000 feet and oscillating thework string152 partially in and out of the well bore and then retracting thework string152 out of the well bore. It will readily be appreciated that a wide variety of work string manipulation regimes may be described and/or defined by such a series of linked pairs of position and velocity, all of which are contemplated by the present disclosure.

Inblock954, the simulated force feedback of the work string on the manipulators is determined. The simulated force feedback of the work string on the manipulators is the sum of the simulated force feedback of the work string on each of the manipulators of the control system. In an embodiment, three manipulators are deployed, for example the dual-jackingsystem150, where the first travelinghead156 and thefirst slip bowl154 comprise a first manipulator, thesecond traveling head160 and thesecond slip bowl158 comprise a second manipulator, and thestationary slip bowl162 comprises a third manipulator. In this embodiment, the simulated force feedback of the work string on the first manipulator812-a, the simulated force feedback of the work string on the second manipulator812-b, and the simulated force feedback of the work string on the third manipulator812-care determined and summed.

Inblock956, the estimatedwork string velocity842 and the estimatedwork string position840 are determined. In an embodiment, a model force is determined based on the estimated feedback force of the manipulators on thework string812, on the workstring position command818, and on the workstring velocity command828. The model force is multiplied by the reciprocal of the estimated work string mass to transform the model force to a model work string acceleration. The estimatedwork string velocity842 is determined by integrating the model work string acceleration. The estimatedwork string position840 is determined by integrating the estimatedwork string velocity842.

Inblock958, a first manipulator controller command input vector340-ais determined based on the estimatedwork string velocity842, the estimatedwork string position840, thepipe collar indication844, and the mode of operation. The first manipulator controller command input vector340-ais transmitted to the first manipulator control system300-a.

Inblock960, a second manipulator controller command input vector340-bis determined based on the estimatedwork string velocity842, the estimatedwork string position840, thepipe collar indication844, and the mode of operation. The second manipulator controller command input vector340-bis transmitted to the second manipulator control system300-b.

Inblock962, a first manipulator position command and a first manipulator velocity command are determined by using the first manipulator controller command input vector340-ato drive the first manipulator physical system components to desired values.

Inblock964, a second manipulator position command and a second manipulator velocity command are determined by using the second manipulator controller command input vector340-bto drive the second manipulator physical system components to desired values. In block966, the first and second manipulators are automatically controlled based on the first and second manipulator position and velocity commands. Themethod950 then ends or may be repeated at a periodic rate effective to control thework string152.

The system described above may be implemented on any general-purpose computer with sufficient processing power, memory resources, and network throughput capability to handle the necessary workload placed upon it.FIG. 18 illustrates a typical, general-purpose computer system suitable for implementing one or more embodiments disclosed herein. Thecomputer system1080 includes a processor1082 (which may be referred to as a central processor unit or CPU) that is in communication with memory devices includingsecondary storage1084, read only memory (ROM)1086, random access memory (RAM)1088, input/output (I/O)devices1090, andnetwork connectivity devices1092. The processor may be implemented as one or more CPU chips.

Thesecondary storage1084 is typically comprised of one or more disk drives or tape drives and is used for non-volatile storage of data and as an over-flow data storage device ifRAM1088 is not large enough to hold all working data.Secondary storage1084 may be used to store programs which are loaded intoRAM1088 when such programs are selected for execution. TheROM1086 is used to store instructions and perhaps data which are read during program execution.ROM1086 is a non-volatile memory device which typically has a small memory capacity relative to the larger memory capacity of secondary storage. TheRAM1088 is used to store volatile data and perhaps to store instructions. Access to bothROM1086 andRAM1088 is typically faster than tosecondary storage1084.

I/O devices1090 may include printers, video monitors, liquid crystal displays (LCDs), touch screen displays, keyboards, keypads, switches, dials, mice, track balls, voice recognizers, card readers, paper tape readers, or other well-known input devices.

Thenetwork connectivity devices1092 may take the form of modems, modem banks, ethernet cards, universal serial bus (USB) interface cards, serial interfaces, token ring cards, fiber distributed data interface (FDDI) cards, wireless local area network (WLAN) cards, radio transceiver cards such as code division multiple access (CDMA) and/or global system for mobile communications (GSM) radio transceiver cards, and other well-known network devices. Thesenetwork connectivity devices1092 may enable theprocessor1082 to communicate with an Internet or one or more intranets. With such a network connection, it is contemplated that theprocessor1082 might receive information from the network, or might output information to the network in the course of performing the above-described method steps. Such information, which is often represented as a sequence of instructions to be executed usingprocessor1082, may be received from and outputted to the network, for example, in the form of a computer data signal embodied in a carrier wave. In an embodiment, portions of thecontrol system8, thecontroller202, themanipulator control system300, and/or thework string controller800 may be reconfigured and/or gains be tuned from a remote computer via thenetwork connectivity devices1092.

Such information, which may include data or instructions to be executed usingprocessor1082 for example, may be received from and outputted to the network, for example, in the form of a computer data baseband signal or signal embodied in a carrier wave. The baseband signal or signal embodied in the carrier wave generated by thenetwork connectivity devices1092 may propagate in or on the surface of electrical conductors, in coaxial cables, in waveguides, in optical media, for example optical fiber, or in the air or free space. The information contained in the baseband signal or signal embedded in the carrier wave may be ordered according to different sequences, as may be desirable for either processing or generating the information or transmitting or receiving the information. The baseband signal or signal embedded in the carrier wave, or other types of signals currently used or hereafter developed, referred to herein as the transmission medium, may be generated according to several methods well known to one skilled in the control systems art.

Theprocessor1082 executes instructions, codes, computer programs, scripts which it accesses from hard disk, floppy disk, optical disk (these various disk based systems may all be considered secondary storage1084),ROM1086,RAM1088, or thenetwork connectivity devices1092. While only oneprocessor1092 is shown, multiple processors may be present. Thus, while instructions may be discussed as executed by a processor, the instructions may be executed simultaneously, serially, or otherwise executed by one or multiple processors.

While several embodiments have been provided in the present disclosure, it should be understood that the disclosed systems and methods may be embodied in many other specific forms without departing from the spirit or scope of the present disclosure. The present examples are to be considered as illustrative and not restrictive, and the intention is not to be limited to the details given herein. For example, the various elements or components may be combined or integrated in another system or certain features may be omitted or not implemented.

Also, techniques, systems, subsystems, and methods described and illustrated in the various embodiments as discrete or separate may be combined or integrated with other systems, modules, techniques, or methods without departing from the scope of the present disclosure. Other items shown or discussed as directly coupled or communicating with each other may be indirectly coupled or communicating through some interface, device, or intermediate component, whether electrically, mechanically, or otherwise. Other examples of changes, substitutions, and alterations are ascertainable by one skilled in the art and could be made without departing from the spirit and scope disclosed herein.