US20120200155A1 - Electric power distribution methods and apparatus - Google Patents

Abstract

A plurality of end-user locations are served by a commercial utility grid. More than one and less than all of the end-user locations are themselves interconnected by a feeder, the feeder not metallically connected to the utility grid. The end-user locations each have a local AC bus that is not metallically connected to the utility grid or to the feeder, but that is linked by a coupler to both the utility grid and to the feeder. None of the local AC buses or the feeder is required to have the same phase or frequency as the utility grid. Locally generated electric power may be passed by means of the feeder to other end-user locations that are on the feeder. Each local AC bus has two or more inverters powering the bus.

Description

CROSS-REFERENCE TO RELATED APPLICATIONS
• This patent application is a divisional of U.S. application Ser. No. 12/519,507, filed Dec. 12, 2008, which claims the benefit of U.S. application No. 60/996,954 filed Dec. 12, 2007, which applications are hereby incorporated herein by reference for all purposes.
BACKGROUND
• Much attention has been given in recent times to energy policy and energy conservation generally, and to electric power generation and transmission in particular. Traditional models for power generation and distribution may once have been the only workable ways to deliver power to end users. But a variety of factors including political events, rising energy costs, technological progress, and concern for the environment have drawn attention to the need for new paradigms and approaches.
• It is apparent that one important goal is to make it possible to draw upon distributed energy resources as a source of electric power for end users. There are, however, a number of forces seemingly conspiring to limit or frustrate the use of distributed energy resources. One problem is that the operators of commercial power grids refuse to permit large amounts of power to be fed into the grid from end-user locations. Typically the operator of a commercial power grid will permit feeding of such power only up to a very small percentage of the distribution capacity of the grid. This means that if one end-user location happens to have a large amount of locally generated power available, it is likely to be difficult or impossible to use the commercial power grid as a way to distribute that power to other end user locations.
• It would thus be very desirable if a workable approach could be found for passing large amounts of power from one end-user location to another, despite the lack of cooperation on the part of the operator of the commercial power grid.
• As will be discussed in more detail below in connection with the invention, experience reveals that moving electric power from one part of an end-user location to another, or from one end-user location to another, in an efficient and reliable way, is not easy. Traditional ways of passing power from one grid to another have many drawbacks. In more recent times, inverters have been developed that convert DC to AC in an efficient fashion and with improved quality of alternating current. But the inverters, taken singly, do not serve the end users as well as might be desired.
• It would be very helpful if an approach could be found for coordinating the frequency and phase of generated AC so as to avoid conflicts among the two or more inverters that might be connected to a given AC power bus.
• Patents of possible background interest include U.S. Pat. No. 7,145,266 to Lynch, et alia entitled Parallel-connected inverters with separate controllers having impedance current regulators, U.S. Pat. No. 7,116,010 to Lasseter et alia, entitled Control of small distributed energy resources, and U.S. Pat. No. 6,693,409 to Lynch, et alia entitled Control system for a power converter and method of controlling operation of a power converter.
SUMMARY OF THE INVENTION
• A plurality of end-user locations are served by a commercial utility grid. More than one and less than all of the end-user locations are themselves interconnected by a feeder, the feeder not metallically connected to the utility grid. The end-user locations each have a local AC bus that is not metallically connected to the utility grid or to the feeder, but that is linked by a coupler to both the utility grid and to the feeder. None of the local AC buses or the feeder is required to have the same phase or frequency as the utility grid. Locally generated electric power may be passed by means of the feeder to other end-user locations that are on the feeder. Each local AC bus has two or more inverters powering the bus. The inverters for a particular local AC bus are linked by out-of-band signaling with a latency far shorter than the period of the AC power, and at any given moment one of the inverters is a master and the others are slaves, so far as voltage, frequency, and phase are concerned. The DC buses powering the inverters are themselves isolated from each other. The linkage from each inverter to its AC bus is free from any transformer.
DESCRIPTION OF THE DRAWING
• The invention will be described with respect to a drawing in several figures, of which:
• FIG. 1 shows a small-town distribution system;
• FIG. 2 shows a plurality of inverters powering a local AC bus;
• FIG. 3 shows a detail of a bank of inverters, with particular attention to out-of-band control links;
• FIG. 4 shows detail of an inverter;
• FIG. 5 shows an internal power circuit for a building;
• FIG. 6 shows a prior-art standard interconnection;
• FIG. 7 shows an installation with an independent grid for a single building; and
• FIG. 8 shows an installation with multiple independent grids, as well as an unsynchronized feeder.
• Where possible, like reference numerals have been used among the figures to denote like elements.
DETAILED DESCRIPTION
• FIG. 1 shows a small-town distribution system. In an exemplary embodiment, AC power is transmitted to the area by a utility company omitted for clarity inFIG. 1. The power is transmitted to the area byline11 which may be 115 kilovolts (kV) three-phase AC power. This power is reduced in voltage at asubstation24 operated by the utility company for distribution on a distribution line orgrid12 which may carry 13 kV. This distribution grid is likewise three-phase AC power.Residences13 are served by thedistribution grid12, as is a medium industrial plant16,small industries17,18, and19, ashopping mall25A.
• It may be that the utility company will place a strict limit on the amount of electrical power that any one customer or end user is permitted to generate locally under circumstances that might lead to power being fed back into the utility grid. Such a strict limit may in part be justified by legitimate engineering concerns, and may be in part caused by mindsets dating from earlier decades when all aspects of the utility grid were centrally planned and controlled.
• Thus for example in a prior-art system, if theshopping mall25A were to possess a source of on-site power21, and if the connection of the on-site power throughline15 were such that power could be fed back into theutility grid12, there might be strict limits on the permitted generation capacity of the on-site power21. The limit may be only a small percentage of the power distribution capacity of thelocal grid12. In such a prior-art environment, there are many social, environmental, and economic benefits that might have been available if only the on-site power21 could have been larger in its power generation capacity than the capacity permitted by the utility or regulatory environment. Such social, environmental, and economic benefits are, however, completely forgone in a prior-art system.
• In oneembodiment41 of the invention,feeders14 are provided.Independent area42 is defined by arespective feeder14 which connects end-user locations16 and17. Note that the end-user locations16 and17 represent more than one but less than all of the end-user locations served by the utility, and represent more than one but less than all of the end-user locations served by thedistribution grid12. As will be described in more detail below, the presence of thefeeder14 permits a local distribution of power from anindependent power plant22,23. Thefeeder14 is not metallically linked to thedistribution grid12, and need not be operating at the same phase, or at the same frequency, as the power provided by the utility viagrid12.
• Eachend user16,17 has a respective local AC grid omitted for clarity inFIG. 1. At each local AC grid is a coupler coupling thegrid12 thereto, and a coupler coupling thefeeder14 thereto; the couplers are likewise omitted for clarity inFIG. 1.
• Independent area43 is also defined by arespective feeder14 which connects end-user locations25A,19, and18. Note that the end-user locations25A,19 and18 represent more than one but less than all of the end-user locations served by the utility, and represent more than one but less than all of the end-user locations served by thedistribution grid12. As will be described in more detail below, the presence of thefeeder14 permits a local distribution of power from on-site power generators20,21. Just as was described in connection withindependent area42, thefeeder14 ofindependent area43 is not metallically linked to thedistribution grid12, and need not be operating at the same phase, or at the same frequency, as the power provided by the utility viagrid12.
• Eachend user25A,19,18 has a respective local AC grid omitted for clarity inFIG. 1. At each local AC grid is a coupler coupling thegrid12 thereto, and a coupler coupling thefeeder14 thereto; the couplers are likewise omitted for clarity inFIG. 1.
• As will be described in more detail below, theembodiment41 offers many benefits when compared with prior-art power distribution systems. Alocal power source22,23,21,20 can be of much greater power generation capacity than would be feasible in a prior-art system. Particular end users are able to enjoy greater reliability (e.g. up-time) with respect to the entirety of their energy needs, as compared with prior-art systems. The benefits that flow from generation of power locally to where it is needed, as compared with the legacy approach of generation power at great distances from where it is needed, may be fully enjoyed. Theembodiment41 may, under some circumstances, also permit an end-user to save money as compared with purchasing most of its power from the utility.
• It will be appreciated that while many embodiments of a system such as that discussed in connection withFIG. 1 will be embodiments where thedistribution grid12 is operated and powered by a commercial electrical power utility, the teachings of the invention and its benefits are not limited thereto. For example, thegrid12 might be a grid operated by a military unit, perhaps up to the size of a division, in an isolated area such as a desert area.
• Thesystem41 ofFIG. 1 will now be described from a different perspective, namely that of an expanded power distribution system with severalindependent grids14.
• The architecture and control features of theindependent grid14 allow for an almost unlimited ability to expand and integrate with existinggrids12 as future growth desires.
• The figure shows a small-town distribution system41 that has been added to with small on-site power21,20 and largerindependent power units22,23. The existingdistribution system12 could not support interconnection of these generators onto the distribution side of thetransformer24.
• By adding the power converters and control scheme, the power can be integrated into the existing grid and supplied to the end users without the need for upgrades to the existinggrid12.
• Further expansion could include larger independent grids or new ones connected through more power converters to the existing grid or to the other independent grids or both.
• FIG. 2 shows an end-user system40. The end-user system may represent an entire building, or may represent a load-panel area within a large building. In some cases the end-user system40 may represent a plurality of buildings nearby to each other.
• Thesystem40 comprises a plurality ofinverters29,32,33,37 powering alocal AC bus26. Each inverter has aline28 connecting to thebus26. Thebus26 haslocal loads27.
• Thesystem40 receives utility power fromline12, which may be stepped down locally by means oftransformer30. For example the utility may provide 13 kV on theline12, stepped down to three-phase 480 V AC by means of the transformer.
• In day-to-day operation, the 480 VAC power from the utility is coupled by means ofcoupler29 tolocal AC bus26, and thence tolocal loads27.
• Importantly, however, there may be one or more local power generation or storage devices forming part ofsystem40. As one example there may be an engine and generator oralternator34, generating AC power which is coupled bycoupler33 to thelocal bus26. Importantly thispower source34 can be very large, much larger than what would be permitted in a prior-art system. A portion of the power (the portion permitted by the utility) can be fed back into thegrid12 by means ofcoupler29.
• What will also be appreciated is that spare power in thelocal bus26 can also be fed intofeeder14 to be made available to other end users. This takes place throughcoupler32 andoptional transformer31, which may for example step up locally generated power at 480VC to a 13 kV level which is more suitable to distribution to the other end users, who might be a kilometer distant. (The higher voltage reduces resistive losses in the feeder.) The amount of power that can be shared from one end-user location to the next (using feeder14) is far, far greater than the amount of power that could be shared if the only sharing mechanism were that of theutility grid12. It is difficult to overstate the benefit of what has just been said about thefeeder14 when compared with a prior-art system where the only sharing opportunity is thelegacy grid12 operated by a legacy utility that may be suspicious of end users who wish to generate large amounts of power locally.
• InFIG. 2 we see examples of other equipment that might be interconnected with thelocal bus26. For example aDC storage system38,39 permits storing DC energy received from thelocal bus26, or storing DC energy received fromgeneration facility35.Generation facility35 may be a cogeneration facility providing waste heat to some other process while generating power to be delivered to the local bus26 (and/or tostorage38,39).
• Stated differently,FIG. 2 shows a proposed buildingpower distribution system40, which has been made into anindependent grid26, unsynchronized with the main existinggrid12. There is power coming from two different, unsynchronizedexternal feeds12,14, as well as twoonsite generators34,35 and incorporatingenergy storage38,39 in one of them.
• Theinverters29,32,33,37 are controlled in such as way as to feed the 480 VAC main (local)bus26. In this way isolated, independent DC sources are made to behave like a single, larger voltage source feeding thebus26.
• Separate system controls allow for the net energy supply to be varied between theisolated units29,32,33,37 to allow for more economic control of energy sources and to improve redundancy.
• In the above arrangement, the DC link voltages of thevarious inverters29,32,33,37 are not connected together. This increases redundancy since the failure of a single DC link will not cause the failure of any of the others.
• It will be appreciated that while many embodiments of a system such as that discussed in connection withFIG. 2 will be embodiments where the local AC bus is a three-phase “wye” grid, the teachings of the invention and its benefits are not limited thereto. For example, the localC bus grid12 might be a single-phase AC bus, or might be a “delta” three-phase system.
• In the event of failure of thesystem40, it is possible to restore service to thelocal bus26 by openingbreakers28 and closing a bypass or transferswitch25. In an exemplary embodiment, these steps are carried out manually, and it is anticipated that these steps would be required only very infrequently.
• In a system of whichFIG. 2 is an example, there is no limit to the number of different power sources. WhileFIG. 2 shows four possible power sources, there could be more. Typically there would be N+1 or N+2 redundancy. The internal building loads27 are the same as they would be for a standard utility power distribution layout.
• The key to controlling thelayout41 ofFIG. 2 is that all theinverters29,32,33,37 that are tied to themain AC bus26 must share theload27.
• One embodiment being described herein is a power system which includes a utility grid providing AC electric power connectivity and extending geographically to a plurality of first end-user locations and to a plurality of second end-user locations. The first end-user locations comprise more than one and less than all of the end-user locations of the utility grid. Each of the first end-user locations comprises a local bus providing AC electric power connectivity. Each local bus has associated with it at least one AC load metallically connected thereto and drawing power therefrom. Each local bus is coupled by means of a respective first at least one coupler to the utility grid.
• The at least one coupler comprises a first power-supply-inverter and a second power-supply-inverter, each power-supply-inverter having an AC interface and a DC interface and a controller. Each power-supply-inverter is disposed in a first mode to receive DC power received at the DC interface and to generate AC power delivered at the AC interface, and is disposed in a second mode to receive AC power received at the AC interface and to generate DC power delivered at the DC interface. The DC interface of the first power-supply-inverter is electrically connected to the DC interface of the second power-supply-inverter.
• The at least one coupler has a first AC interface defined as the AC interface of the first power-supply-inverter and a second AC interface defined as the AC interface of the second power-supply-inverter. The controller of the first power-supply-inverter and the controller of the second power-supply-inverter coupled so as to prevent the first power-supply-inverter and the second power-supply-inverter from being in the first mode simultaneously for extended intervals, and so as to prevent the first power-supply-inverter and the second power-supply-inverter from being in the second mode simultaneously for extended intervals. By “extended intervals” we may mean in excess of a few milliseconds or in excess of half a second.
• Each power-supply-inverter is disposed when in its first mode to generate its AC power delivered at its AC interface consistent in voltage and phase and frequency with any AC power present external to said AC interface.
• FIG. 3 shows a detail of a bank ofinverters44,46,48,50, with particular attention to out-of-band control links52,53.
• FIG. 3 is intended to portray a basic power layout, ignoring for the moment the source of DC power to the inverters. In this embodiment the AC grid (local bus)26 is a four-wire system withNeutral line57, fed directly by theinverters44,46,48,50 without a transformer downstream of them (between the inverter and the local bus26).
• It will be appreciated that a control scheme is required such that all of the inverters that are sourcing both kW and kVAR current into the same voltage node (here, the local bus26) will work together and not fight each other for control. The control scheme used in this system to solve this problem works by having one unit, called the master, maintain the voltage of the four-wire 480/277 VAC bus. Its individual phase currents are measured and the values passed along a highspeed communication link52 to the slave units to be copied. For example at a particular moment, the master might beinverter46 havingrespective controller47, while theother controllers45,49,51 are acting as slaves, each controlling itsrespective inverter44,48,50.
• Eachindividual inverter44,46,48,50 still maintains over-current and other protection features, and amain breaker56 can be remotely tripped by the system or inverter controller under certain conditions.
• A standard prior-art way of paralleling a group of synchronous rotating generators gives oscillations because they are trying to share power and speed but can only be controlled by a mechanical fuel throttle that has a slow reaction time between throttle change and output power change. There are no power electronics in such a system. The power electronic converters of the present embodiments can change power at least 500 to 1000 times faster.
• The method for providing maximum fault current as fast as possible involves the use of measured voltage limits by each individual inverter, such that if a voltage is out of range due to a fault, the maximum amount of current is sourced by each unit for as long as possible or until the fault is cleared.
• The fault current must be sourced to open a distribution breaker. This is a difficult inverter challenge while being controlled in a current-sharing manner. SeeFIG. 5 which shows aninternal power circuit73 for a building. Within the building is a 480 VACmain bus71.Branch feeders72,74,76,78 deliver AC power to circuits throughout the building. Independent 800 VDC supplies81,82,83, and84 are shown, each able to deliver AC power to themain bus71.
• Consider what happens if there is afault85 giving rise to a fault current86. Theinverters81,82,83,84 must be able collectively to source enough fault current to open the protection on the faulted feeder without taking down the whole independent grid.
• One of the biggest limitations of inverter based microgrids is their inability to be retro-fitted to existing buildings without completely re-doing the breaker scheme because the inverters cannot source enough fault current to trip a breaker. Thus the inverters will trip off (instead of a breaker being tripped) and the whole building goes dark, because the inverters have tripped off. This contrasts with a simpler prior-art system in which a fault current would have blasted open the breaker of one of the branch feeders and the rest of the building would have stayed on.
• With a single, low impedance voltage source it is a simple calculation to determine how much fault current can be sourced versus how much is needed for certain breakers. If the breaker is too big for the inverter (that is, if the tripping current for the breaker exceeds the current-sourcing capability of the inverter), then the breaker needs to have intelligent fault sensing controls added, or else the problem of the whole building going dark comes back.
• With parallel inverters, what is needed is for the inverters to act more quickly than they could in a simple current-following mode. In an exemplary embodiment, the inverters are configured to sense a fault through voltage sensing, and then each inverter is released to feed maximum current until the voltage comes back, which happens after the breaker feeding the fault opens.
• FIG. 5 also showsoptional power sources90,91 which may provide AC power to theAC bus71. For example aninverter91 may receive power from (say) a photovoltaic array. Induction generator90 may receive power from a source of rotary energy such as a turbine. As will be discussed further below, each of these sources is necessarily synchronous to whatever voltage source defines the AC voltage on theAC bus71. Here, the voltage source is the one ormore inverters81,82,83,84. What may happen, and what is in fact not uncommon, is that the power from one or more of these sources may be of poor quality. The source may only deliver power to one phase. Even if the source delivers power to all three phases, it may not be in perfect phase relationship. Loads that are assymmetric (as between the three phases) or that introduce power-factor loads may be present on theAC bus71, and if they do, this may be beyond the ability of thesources90,91 to correct or compensate for. As will be discussed below, however, with suitable configuration theinverters81,82,83,84 can sample the voltage waveforms present on the three phases, and can nearly instantaneously deliver voltage to the phases in such a way as to overcome nearly all such problems. In this way, an end user of asystem73 can make use of commercial, off-the-shelf power sources even if they produce power of poor non-utility-grade quality. In an exemplary embodiment, the combined power generation capacity of the non-utility-grade sources90,91 might be up to fifty percent of the combined power generation capacity of theinverters81,82,83,84. In another embodiment the percentage might be sixty-five percent.
• To recap, in the layout ofFIG. 4 above, the DC link voltages of the various inverters are not connected together. This helps reduce fault current levels and increases redundancy. The output AC inverters are still able to work together as a single unit voltage source by having the transistor PWM (pulse-width modulation) pattern sent from a master controller (one of45,47,49,51) and passed through to the slave controller (the others of45,47,49,51) of each individual inverter control.
• The Master PWM Controller acts as it would act if it were driving a single inverter, adjusting the PWM pattern to maintain a fixed voltage and frequency no matter what the load. This can include adjusting the PWM pattern to compensate for non-linear load characteristics that would otherwise cause voltage distortion, thus actively filtering harmonic currents.
• To take advantage of price differentials at different times for different sources of energy, it is necessary to control how much of each energy source supplies the load. In the exemplary control scheme the power flow can be controlled accordingly.
• Corrected elsewhere, the DC voltage will now stay stable and the power flow change will be done by each slave inverter varying its PWM pattern based on the current slave signal multiplied by (x) a proportioning signal coming from the CanBUS. Note that this may include a proportion greater than 100%, which will cause the slave to supply more power than the master.
• It is instructive to return briefly toFIG. 2 for a description of the distributed generation control strategies.
• The energy sources for the on-site power34,35 can come from many different sources, but the most common are the burning of natural gas or syngas, the use of variable speed engine generators or turbines, the use of fuel cells, and the use of solar power. Many of these sources require DC/AC conversion.
• In addition,energy storage technologies38,39 can help balance power in the grid, smooth the peak generation requirements, and provide short-term power during loss of other generation sources.
• The power electronics and control scheme can take advantage of the ability to run rotating machinery at variable speeds and convert the electric power. Importantly, in a typical prior-art system, any rotating-machinery power source such as a generator or alternator is required to run at some fixed fraction or multiple of the frequency of the local bus, and is required to maintain a fixed phase relationship with the local bus. But in the approach ofFIG. 2, a power source drawing upon rotating machinery is able to serve its purpose even if the rotation is at some frequency and phase that is not linked to anything about the local bus. Saying this in a different way, there is no requirement that the rotating machinery be rotating at any particular frequency, to be able to generate power to be supplied to thelocal bus26.
• The approach according to the invention is thus a much more efficient approach than the standard prior-art way of generating fixed-frequency electric power from rotating generators, especially over wide power ranges which are more necessary with smaller grids.
• Similarly if a gas or steam turbine generator is used with a system according to the invention, it becomes possible to eliminate a gearbox. This decreases size and allows variable speeds, which increases efficiency. Size and weight issues can be particularly important in dense areas where construction space is limited.
• Returning toFIG. 3, it is noted that the DC/AC or DC/DC inverters shown each have aCanBUS connection53 and a dedicated high-speed link52 as shown.
• The main redundant feature of the inverter controls is to be able to keep running if one unit fails. If the master unit fails, the next slave unit down the line becomes the master and continues on. So for example, ifcontroller47 is the master and if the other controllers are slaves, a provision must be made for the possibility that the controller47 (or its inverter46) may fail. In that case, an arbitration mechanism is employed to promote one of theother controllers45,49,51 to “master” status.
• A series of alarm and warning messages can be sent viaETHERNET55 over the internet to remote monitoring facilities. This enables remote diagnostic capabilities and the ability to more quickly dispatch necessary maintenance support.
• Under this arrangement according to the invention, since any single inverter or energy source can fail without causing complete grid failure, then the faster the failure can be fixed, the less the chance of a grid failure due to a second equipment failure.
• Depending upon the particular grid and design and cost considerations, it may be required that the grid run at a reduced load until repairs can be completed. This may be carried out using demand response signals from thesystem level controller54 to an intelligent load shedding control, but done in such a way that the AC voltage on the grid always stays within specification.
• In the event of a failure of thesystem level controller54, theinverters44,46,48,50 are programmed to go into a default mode and keep supplying the grid with balanced or pre-set proportions from the various energy sources. An alarm will be sent viaETHERNET55 or simply the absence of the required signal will trigger an alarm upstream. Again, this is done seamlessly.
• These features cover all the significant faults that could stop theindependent grid26 from providing voltage within the specified limits under any single point of failure.
• The final redundant feature is the ability to switch the whole system off and go to a bypass switch (25 inFIG. 2) to feed the load just as a traditional distribution system does. This would normally be done manually for safety reasons.
• The bypass feature is easy to integrate because the independent grid is designed to integrate into existing grids with very few changes required.
• A system control scheme for a single independent grid will now be described.
• To both balance thermal energy requirements and take advantage of price differentials at different times for different sources of energy, it is necessary to control how much of each energy source supplies the load at a given moment.
• In the control scheme according to an embodiment of the invention, this is done by sending a mathematical multiplier to eachinverter44,46,48,50 viaCanBUS53, such that it can be multiplied inside the slave units to the proportioning signals coming from whichever inverter is serving as the master inverter at that time.
• Two-way communication viaCanBUS53 is used to adjust the amount of energy from different sources in the event of an inverter failure. The inverters will automatically re-assign master control to the next slave unit, but the total amount of energy fed to the independent grid must be maintained.
• The system may also include intelligent switchgear on certain loads to enable fast load shedding of less critical loads in the event of disruptions from an energy source causing total load capacity problems.
• The communication to a central station for optimizing energy use and monitoring system conditions is done via ETHERNET link55 to the internet.
• As may be appreciated from the above discussion, what is described includes the application of a control scheme to control multiple inverters in order to drive an independent electrical grid fed by multiple energy sources, controlling the energy flow from the various sources, along with redundant back-up capabilities.
• A main purpose behind creating such a control scheme is to facilitate the efficient integration of distributed energy resources (DER) into the existing power grid, without being limited to existing penetration level limits imposed by operators of the existing grid. This ability to integrate unlimited amounts of various sources of electrical energy, without regard to the present state of the existing grid, is at the heart of the need for this technology application.
• The main way in which this task is accomplished is by having the Independent Grid (feeder14) be seen by themain grid12 as a load-reduction type device, as opposed to a parallel interconnected generator.
• For projects requiring capacity larger than a single grid can support, theindependent grids14 are capable of being interconnected with other independent grids, for unlimited expandability.
• The system provides for the independent grid and the efficient use of distributed generation assets, while overcoming the main challenges of integration with respect to control and stability of the existing grid.
• Additional control features to improve redundant capability. The Multiple Inverter Control Scheme can include a redundant, always on-line, back up PWM generator controller. If the Master PWM healthy signal is lost, the voltage source inverters can switch to the backup without interruption or delay.
• There are other software features to command the voltage source inverters and energy source converters to behave in certain ways under certain conditions to prevent complete system trips. The system may also include intelligent switchgear on certain loads to enable fast load shedding of less critical loads in the event of disruptions from an energy source causing total load capacity problems.
• Upstream grid fault current limiting. In a Distributed Generation application with the inverter tied in parallel to a larger grid, the path of power flow during an upstream grid fault is difficult to predict.
• The main grid operator will want to shut down all connected distributed generation sources as quickly as possible so that the existing protection breakers do not see an increase in the amount of fault current that they must interrupt.
• By simply measuring current, the inverter controller cannot tell that such an upstream fault has occurred due to the various loads (such as induction motors) and other generation sources which may create resonant circulating paths for the current. The only way to sense an upstream fault is through voltage measurement and comparison between the phases.
• The inverter controller can be programmed to measure such anomalies and shut down on the next 200-microsecond transistor switch. The controller must have the capability to filter noise from the measurements such that it does not trip on nuisance events.
• High Speed current control/power flow control. As noted above, with the inverter switching40+times per half cycle, the ability to control current is much faster than any other means of standard electrical controls.
• The inverter controller monitors the AC voltage and switches ON and OFF the transistors to create current flow. The controller monitors the feedback of the current sensor and can adjust the switching of the transistors for both amplitude and phase angle of the current relative to the voltage.
• The control bandwidth is a function of the transistor switching frequency, with the feedback sampling frequency at a higher rate for improved accuracy.
• When the inverter is tied to an infinite grid, it is always in current control mode. If the grid is unstable, the inverter can help stabilize it by monitoring the frequency and voltage and adjusting real or reactive current, or both, as required.
• The inverter will have some inherent harmonic compensation for the main grid due to the low source impedance of the current source. Active harmonic filtering through current control of AC voltages measured at multiple frequencies of the fundamental can be incorporated, but this is a significant extra feature and will require switching frequencies higher than 5 kHz for harmonics beyond 5th and 7th.
• Output AC voltage source operating mode. If the inverter is NOT connected to an existing infinite grid, it can act as the voltage source. In this mode it will turn the transistors ON and OFF and measure the output voltage, adjusting to keep an accurate 60 Hz 480V sinusoidal waveform. The current draw will be dependent upon the load, with the current sensors acting as protection devices. The inverter may be programmed to lower the output voltage under heavy current draw if desired.
• The inverters can be placed in parallel to increase the capacity of the overall output. With special controls, multiple inverters can be grouped together to act as one, even though the DC link sources are not connected together.
• FIG. 4 shows detail of an inverter.
• The diagram shows a simple circuit of 6 transistors including61,62 connected in3 series pairs between aDC supply63 through an inductor to anAC line58. The output AC current and voltage are measured onlines59 and fed back to theinverter controller60. Thus there is current and voltage feedback as to all 3 phases of the AC line.
• The basic concept of the power flow control is that either thetop transistor61 orbottom one62 is switched ON (depending upon the AC waveform polarity at the time) to create a path from the AC voltage through the inductor to the + or − DC link, causing current to flow and energy to be stored in the magnetic field of the inductor.
• When this transistor turns OFF, the energy stored will be dumped through the freewheel diode of the opposite transistor into the DC link capacitors.
• The timing of the transistor firing can be manipulated such that the amplitude of the current and phase angle of the voltage and current can be controlled. This allows for separate real and reactive current control, current limits and with 180 degree phase shifts, the direction of the net power flow.
• Fast Response. Due to the switching frequency of the transistors (5 kHz), changes can be made every 200 uSeconds, or 41.5 times in each half cycle (at 60 Hz).
• FIG. 6 shows a prior-art standard interconnection. Power is generated by a utility company atgenerators101,102, each linked by atransformer103,104 to a 230 kilovolt (for example)transmission line106. This is stepped down bytransformers107,108 to a 115kilovolt transmission line109. This power may be stepped down by atransformer110 to a 13.8kilovolt distribution line111. The power may also be stepped down by atransformer114 to a 13.8kilovolt distribution line115. The transmission lines may be any of several voltages including 230 kV or 345 kV or 500 kV.
• Distribution line115 may servebuildings116,117, and118, each having arespective transformer119,120,121. Within each building is a respectiveAC power bus139,140,141.
• Asingle building118 may have agenerator123 and an inverter-based local source of generatedpower124. AC power from these sources is carried to thebuilding118 by means of abus125. As will be appreciated from the above discussion, the regulatory environment will typically place a strict and low limit on the amount of power that may be passed frombus125 back into the utility grid at115,109, and so on.
• Importantly, the transformer connections are each, by definition, synchronous as between the primary and secondary windings thereof. This means that thegenerator123 is strictly required to be synchronous with the utility grid and with theutility generators101,102. Likewise theinverter125 is strictly required to be synchronous with the utility grid and with theutility generators101,102.
• FIG. 7 shows an installation with an independent grid for asingle building126. The115kilovolt transmission line109 is seen, just as inFIG. 6.Distribution line115 is seen, as inFIG. 6. Building129 functions as inFIG. 6. Importantly, however, inFIG. 7 it may be seen that building126 is quite different. Utility power is passed throughtransformer121 toline127, where it is rectified at128 and provided toDC bus129. In this example apower source130 provides power to a DC/DC converter131 to theDC bus129. Also in this example a bidirectional DC/DC converter133 connects a DCenergy storage device132 with theDC bus129. Theenergy storage device132 might, for example, be a battery.
• What is important here is that the direction of the power flow is FROM the grid not TO the grid, thus the independent grid is not seen by the existing grid as a generator but as a load, and in this case, a reduced load, due to the on-site generators providing much of the energy required for the building.
• Agenerator134 provides power torectifier135, and from there to theDC bus129. The DC power at theDC bus129 is inverted at126 and is provided vialine137 tointernal AC bus138 to power loads in thebuilding126.
• Importantly, it will be appreciated that there is no requirement that thegenerator134 be synchronous with anything else.
• FIG. 8 shows aninstallation160 with multiple independent grids, as well as anunsychronized feeder153.FIG. 8 shows many of the same functional blocks asFIG. 7. New functional blocks inFIG. 8 includecoupler155 which couples theDC bus129 with the feeder153 (perhaps through atransformer154 as shown). Building144,line building126, has aninternal bus146 that is not tied metallically to the building feed at142. In this way building144 is similar to building126, which likewise has aninternal bus138 that is not tied metallically to the building feed at121.
• Note that in an exemplary embodiment, the feeder is a distribution level voltage, 13.8 kV. The power conversions take place at 480V, but any distance will be required to have 13.8 kV lines. Note that this is also the highest voltage that can be used with the generator, in this case a 10 MW turbine.
• Associated with building144 is alocal DC bus150, much as building126 has associated with it alocal DC bus129. This permits building144 to draw power from thefeeder153 as desired, thereby perhaps making use of the power fromsources130,132, or134.
• In this example there is awaste treatment plant162, with agas turbine161. Theturbine161 turnsgenerator163, which supplies AC power (in this example, at 13.8 kilovolts) to thefeeder153. In this way either or both ofbuildings144,126 is able to draw upon the power from thegenerator163.
• To the extent thatcoupler155 serves as an inverter (delivering power fromDC bus129 to feeder153), thecoupler155 is required to be synchronous withgenerator163. But neithercoupler155 norgenerator163 is required to be synchronous with the utility grid at113 and115.
• Note thatInverter155 is a three-wire current source, and it synchronizes to the voltage source of the 10 MW generator. It can cause current to flow, and control the phase angle of this current to produce +/− kW or VAR depending on the phase angle. A zero degree shift is kW, a 90 degree shift is VAR, and in between is a vector sum of the two.Inverter128 is simply programmed not to allow −kW phase angles of current to flow.
• Any feeder such asfeeder153 necessarily has exactly one power source that defines the voltages on the feeder (thus called a “voltage source”). Any other power sources that also deliver power to thefeeder153 are necessarily not voltage sources but are mere current sources. Thus in a typical arrangement the inverter155 (e.g. inFIG. 8) serves as a current source and not a voltage source.
• Alternatively, however, it might develop that thegenerator163 might go out of service. If this were to happen, then some other power source such asinverter155 could be reconfigured to serve as the voltage source for thebus153. In this way, power could be delivered (seeFIG. 8) from the microsite containingpower sources121,130,134 to a different microsite such as that of building144 even ifgenerator163 were out of service.
• Inverter136 is a four-wire voltage source. It outputs three single phase voltages between phases A, B, C and N. It does this no matter what the load, up to the silicon limit of the transistors so long as the DC can be maintained within limits by whatever energy sources are feeding or taking from it.
• With further reference toFIG. 8, many advantages and benefits of the invention may be appreciated.
• Consider, for example, thewaste treatment plant162. In a prior-art system, such awaste treatment plant162 might be nearly unusable because of regulatory and engineering constraints imposed by the utility company providing power at113. In the arrangement shown inFIG. 8, however, power from thewaste treatment plant162 can pass throughtransformer152 andelectronics151 to alocal DC bus150 associated withbuilding144. Similarly, power from thewaste treatment plant162 can pass throughtransformer154 andelectronics155 to alocal DC bus129 associated withbuilding126. Nothing about this system requires that the power from thewaste treatment plant162 be “utility grade”. It could be inaccurate in its frequency or phasing, for example, as compared with the utility grid at113. If the power from theplant162 is three-phase power (as it is very likely to be in most embodiments), it is no problem at all if there are unbalanced loads somewhere on thefeeder153. Even if there are unbalanced loads, the rectified power passed toDC bus150 or129 is capable of being put fully to use.
• InFIG. 8, theelectronics128 may be configured so that the power flow is solely from left to right inFIG. 8. If so, then on a practical level the system that is to the right ofelectronics128 will appear to the existing utility grid as a load reduction and not as an interconnected generator.
• As mentioned above, in an exemplary embodiment theplant162 with itsgenerator163 might well not be generating utility grade power. For example the power generated by thegenerator163 may sometimes be of slightly higher frequency relative to specifications, and may at other times be of slightly lower frequency. There might also be unbalanced loads somewhere on thefeeder153 as between the three phases. There could also be poor-power-factor loads somewhere on thefeeder153, giving rise to a condition of the voltage and current on thefeeder153 being pulled out of phase with each other. In the face of all of these possible degradations in the quality of the power supplied on thefeeder153, it is possible to configure inverter155 (together withlocal DC bus129 andenergy storage133,132) to bring about substantial improvements in the power quality on thefeeder153, as will now be discussed.
• In an exemplary power-quality-enhancement approach,inverter155 draws power fromfeeder153 from time to time, some of which is stored inenergy storage133,132. The power drawn need not be power of high quality since it will be rectified anyway on its way tostorage133,132 (passing over DC bus129).Inverter155, however, also monitors the instantaneous voltages present on the three (typical) phases of thefeeder153. Ifinverter155 notices an imperfect voltage waveform on one of the phases, including timing issues of the waveform that diverge from the desired frequency regulation, then it nearly instantaneously pumps some power into that phase (or draws power from that phase) so as to bring the waveform closer to the ideal. The availability of power on the desired time scale—far faster than the reaction time of the mechanical throttle ongenerator134—is achieved through theenergy storage device132. This activity by theinverter155 is carried out upon each of the three phases and can result in the effective waveforms on the phases of thefeeder153 being utility-grade quality.
• In this arrangement, thegenerator163 serves as a voltage source, whileinverter155 serves as a three-wire power source (or sink) for the purpose of dynamic frequency stabilization.
• The power quality issues that are voltage related can be solved by various forms of VAR control or harmonic filtering, but that will not fix a frequency problem, which can only be fixed by changing net kW flow from an energy source. It is not correct to say that theinverter155 sinks or sources current (which a D-VAR does as well); it sinks or sources power (kW), which it gets or sends from the energy storage source.
• On a practical level this could mean, among other things, that the power provided to other end users on thefeeder153 can be sold at a price that is appropriate for utility-grade power (typically a retail price). In contrast, had thegenerator163 been connected in a prior-art fashion directly to a utility grid such as at115, the power would only be able to be sold at much lower wholesale prices. Said differently, the availability of thehardware155,129,133,132 together with appropriate configuration permits providing frequency stability so that non-utility-grade power is rendered utility-grade, offering economic benefits.
• Returning now toFIG. 8, another benefit may be seen. Consider the effect upon building144 or126 if theutility grid113,114,115 is lost to the end users atbuildings144 and126. In a prior-art arrangement, any local backup power generation is permitted to be connected to the building only by means of a “transfer switch”. The transfer switch connects either theutility grid115 to thebuilding bus146, or the local backup power such as fromgenerator134 orgenerator163. With such an arrangement, any switching of the transfer switch will result in a disruption of power.
• Some transfer switches require tens or hundreds of milliseconds to switch. But another issue is that some utilities require that the circuit “go dark” for at least hundreds of milliseconds, before the local backup power comes on line. As such, this is often a regulatory, not technical, limitation. But as mentioned above, the systems according to the invention do not suffer from such interruptions.
• In contrast, however, with anarrangement160 as shown inFIG. 8, a loss of theutility grid115 need not result in any disruption at all of the power to the enduser AC buses146,138. The loss of utility power merely means thatelectronics148,128 are then unable to provide DC power to thelocal DC buses150,129 respectively. But other sources of DC power permit continued and uninterrupted AC power throughinverters149,136 tolocal AC buses146,138.
• Returning again toFIG. 8, it will be appreciated thatelectronics155,128,151,148 can be configured to respond to out-of-band signaling that permits taking action based upon price changes among the various sources of electrical power. If a particular power source becomes more expensive, the electronics can draw less power from that source. If a particular power source becomes less expensive, then the electronics can draw more power from that source.
• Any electrical power distribution system will have fault protection devices which are intended to open in the event of a fault, so as to isolate a failed portion of the system from other non-failed portions of the system. What is undesirable, however, is if in the event of a fault, the fault currents fail to rise to a level that suffices to trigger the relevant fault protection device. With appropriate configuration, however, the inverters feeding a given feeder or bus or grid can respond to internal control logic to sense voltage anomalies on the feeder or bus or grid; when this happens the inverters can supply extra current so as to help clear the relevant fault protection devices as quickly as possible.
CONCLUSION
• The advantages of this control scheme, over other ways to perform the sharing of multiple power sources, include: [0144] A lower output source impedance requirement for the inverters since impedance is not required to be added to aid active power sharing. This reduces cost and efficiency loss, and increases the amount of available fault current that can be sourced from the inverters. [0145] A simple control scheme may be employed, with no need for active Micro-grid type controls. [0146] It is possible to retain redundant operation with no single-component failure able to cause a complete shutdown.
• While the invention has been described with respect to particular embodiments, the invention is not limited thereto. Those skilled in the relevant arts will have no difficulty devising myriad obvious improvements and variations, all of which are intended to be within the scope of the claims which follow, when properly construed.

Claims (22)

1. A system for use with AC power at a frequency, the frequency defining a period, the system comprising:

an AC feeder;

a control path, the control path being separate from the AC feeder, the control path communicating control information with a latency no greater than one-tenth the period of the AC power;

at least two inverters, each inverter receiving DC power from a respective DC power source, each inverter having an AC output metallically connected to the AC feeder, the metallic connection free of any electrical transformer, the respective DC power sources isolated from each other;

each inverter disposed in a first state to generate AC power unsynchronized with any other AC phase, and in said first state to measure its AC output phase and to communicate that phase external to the inverter by means of the control path;

each inverter disposed in a second state to receive an AC phase signal from external to the inverter by means of the control path, and to generate AC power synchronized with the received AC phase signal, said AC power provided to the AC feeder;

control means responsive to a predetermined event causing the first one of the at least two inverters to take said first state and causing any other inverters to take said second state;

said means after said predetermined event responding to failure of the first one of the at least two inverters by causing an inverter other than the first one of the at least two inverters to change to said first state;

said changes of state occurring within an interval less than one-half the period of the AC power;

the AC feeder having loads drawing AC power therefrom.

2. The system ofclaim 1 wherein the number of inverters is at least three.

3. The system ofclaim 1 wherein each inverter generates wye three-phase electrical power.

4. A method for use with a power system, the power system comprising a utility grid providing AC electric power connectivity and extending geographically to a plurality of first end-user locations and to a plurality of second end-user locations, the first end-user locations comprising more than one and less than all of the end-user locations of the utility grid, each of the first end-user locations comprising a local bus providing AC electric power connectivity, each local bus having associated with it at least one AC load metallically connected thereto and drawing power therefrom, each local bus coupled by means of a respective first at least one coupler to the utility grid, the at least one coupler comprising a first power-supply-inverter and a second power-supply-inverter, each power-supply-inverter having an AC interface and a DC interface and a controller, each power-supply-inverter disposed in a first mode to receive DC power received at the DC interface and to generate AC power delivered at the AC interface, and disposed in a second mode to receive AC power received at the AC interface and to generate DC power delivered at the DC interface, the DC interface of the first power-supply-inverter electrically connected to the DC interface of the second power-supply-inverter, whereby the at least one coupler has a first AC interface defined as the AC interface of the first power-supply-inverter and a second AC interface defined as the AC interface of the second power-supply-inverter, the controller of the first power-supply-inverter and the controller of the second power-supply-inverter coupled so as to prevent the first power-supply-inverter and the second power-supply-inverter from being in the first mode simultaneously for extended intervals, and so as to prevent the first power-supply-inverter and the second power-supply-inverter from being in the second mode simultaneously for extended intervals, each power-supply-inverter disposed when in its first mode to generate its AC power delivered at its AC interface consistent in voltage and phase and frequency with any AC power present external to said AC interface, the power system further comprising a feeder providing AC electric power connectivity and extending geographically to the plurality of first end-user locations, each local bus coupled by means of a respective second at least one coupler to the feeder, the feeder being unsynchronized with the utility grid, the method comprising the steps of:

at a first time, powering each of the local buses by means of a respective coupler drawing power from the utility grid;

at a second time, generating locally generated electrical power at one of the first end-user locations, and passing some of said locally generated electrical power to others of the first end-user locations by means of the feeder.

5. A method for use with a system, the system for use with AC power at a frequency, the frequency defining a period, the system comprising an AC feeder;

a control path, the control path being separate from the AC feeder, the control path communicating control information with a latency no greater than one-tenth the period of the AC power;

at least two inverters, each inverter receiving DC power from a respective DC power source, each inverter having an AC output metallically connected to the AC feeder, the metallic connection free of any electrical transformer, the respective DC power sources isolated from each other;

each inverter disposed in a first state to generate AC power unsynchronized with any other AC phase, and in said first state to measure its AC output phase and to communicate that phase external to the inverter by means of the control path;

each inverter disposed in a second state to receive an AC phase signal from external to the inverter by means of the control path, and to generate AC power synchronized with the received AC phase signal, said AC power provided to the AC feeder;

the AC feeder having loads drawing AC power therefrom;

the method comprising:

at a first time, causing the first one of the at least two inverters to take said first state and causing any other inverters to take said second state;

in response to failure of the first one of the at least two inverters, causing an inverter other than the first one of the at least two inverters to change to said first state, said changes of state occurring within an interval less than one-half the period of the AC power.

6. A system comprising:

a three-wire AC power grid;

at least one four-wire wye inverter connected to the AC power grid and serving as a voltage source therefor, the at least one four-wire wye inverter having a respective first power output, any power on the AC power grid having a power quality, the power on the AC power grid defining a voltage and frequency for each phase thereof;

a three-wire inverter connected to the AC power grid and serving as a current source therefor, the three-wire inverter having a respective power output;

the four-wire inverter disposed to sample the voltage on each phase of the AC power grid, and to source or sink power on each phase of the AC power grid so as to dynamically maintain frequency and voltage regulation and improve the power quality of the power on the AC power grid.

7. A system comprising:

a three-wire AC power grid;

at least one four-wire wye inverter connected to the AC power grid and serving as a voltage source therefor, the at last one four-wire wye inverter having a respective first power output, any power on the AC power grid having a power quality, the power on the AC power grid defining a voltage for each phase thereof;

an induction generator connected to the AC power grid and serving as a current source therefor, the induction generator having a respective power output;

the four-wire inverter disposed to sample the voltage on each phase of the AC power grid, and to source or sink power on each phase of the AC power grid so as to dynamically maintain frequency and voltage regulation and improve the power quality of the power on the AC power grid.

8. A method for use with a system comprising a three-wire AC power grid, at least one four-wire wye inverter connected to the AC power grid and serving as a voltage source therefor, the at least one four-wire wye inverter having a respective first power output, any power on the AC power grid having a power quality, the power on the AC power grid defining a voltage for each phase thereof, an additional AC current source connected to the AC power grid and serving as a current source therefor, the additional AC current source having a respective power output, the method comprising the steps of:

at the four-wire inverter, sampling the voltage on each phase of the AC power grid, and

sourcing or sinking power on each phase of the AC power grid so as to dynamically maintain frequency and voltage regulation and improve the power quality of the power on the AC power grid.

9. A system comprising:

a three-wire AC power grid;

a generator connected to the AC power grid and serving as a voltage source therefor, any power on the AC power grid having a power quality, the power on the AC power grid defining a voltage for each phase thereof;

a four-wire wye inverter connected to the AC power grid, the four-wire wye inverter also connected to a DC bus, the DC bus having DC power storage connected thereto;

the four-wire inverter disposed to sample the voltage on each phase of the AC power grid, and to source or sink power on each phase of the AC power grid so as to dynamically maintain frequency and voltage regulation and improve the power quality of the power on the AC power grid.

10. A method for use with a system comprising a three-wire AC power grid, a first generator connected to the AC power grid and serving as a voltage source therefor, any power on the AC power grid having a power quality, the power on the AC power grid defining a voltage for each phase thereof, a four-wire wye inverter connected to the AC power grid, the four-wire wye inverter also connected to a DC bus, the DC bus having DC power storage connected thereto, the method comprising the steps of:

at the four-wire inverter, sampling the voltage on each phase of the AC power grid, and

sourcing or sinking power on each phase of the AC power grid so as to dynamically maintain frequency and voltage regulation and improve the power quality of the power on the AC power grid.

11. A method for use with a first bus and a second bus, the first bus and the second bus connected by a breaker, and with a plurality of inverters, each inverter providing AC power to the first bus at a first voltage, each inverter defining some maximum current associated therewith, the method comprising the steps of:

sensing a fault on the second bus by means of voltage sensing;

in the event of a sensed fault, releasing each inverter to feed its maximum possible current to the first bus.

12. The method ofclaim 11 further comprising the step, performed after the releasing step, of detecting a return of the first voltage, and thereafter, ceasing the feeding of maximum possible current to the first bus.

13. A system comprising a first bus and a second bus, the first bus and the second bus connected by a breaker, and a plurality of inverters, each inverter providing AC power to the first bus at a first voltage, each inverter defining some maximum current associated therewith, each inverter responsive to a sensed fault sensed by means of voltage sensing by delivering its maximum possible current to the first bus.

14. The system ofclaim 13 wherein each inverter is further responsive to detection of a return of the first voltage by ceasing the delivery of the maximum possible current to the first bus.

15. An inverter, the inverter providing AC power at a first voltage, each inverter defining some maximum current associated therewith, each inverter responsive to a sensed fault sensed by means of voltage sensing by delivering its maximum possible current.

16. The inverter ofclaim 15 wherein the inverter is further responsive to detection of a return of the first voltage by ceasing the delivery of the maximum possible current.

17. A system comprising a first DC bus, a second DC bus, a feeder, a first AC bus, a second AC bus, and first and second transformers each connected with the feeder;

the system further comprising a first inverter coupling the first AC bus with the first DC bus, a second inverter coupling the second AC bus with the second DC bus;

a third inverter coupling the first DC bus with the first transformer;

a fourth inverter coupling the second DC bus with the second transformer;

a first load connected with the first AC bus;

a first generator connected with the feeder, the first generator having a power output level;

a second load connected with the feeder;

a second generator connected with the second DC bus and having a power output level;

the second load consuming power amounting to at least the power output of the first generator;

the first load consuming power amounting to at least a portion of the power output of the second generator.

18. A method for use with system comprising a first DC bus, a second DC bus, a feeder, a first AC bus, a second AC bus, and first and second transformers each connected with the feeder;

the system further comprising a first inverter coupling the first AC bus with the first DC bus, a second inverter coupling the second AC bus with the second DC bus;

a third inverter coupling the first DC bus with the first transformer;

a fourth inverter coupling the second DC bus with the second transformer;

a first load connected with the first AC bus;

a first generator connected with the feeder, the first generator having a power output level;

a second load connected with the feeder;

a second generator connected with the second DC bus and having a power output level, the method comprising the steps of:

generating power at the first generator at a respective power output level;

generating power at the second generator at a respective power output level;

at the second load, consuming power amounting to at least the power output of the first generator;

and at the first load, consuming power amounting to at least a portion of the power output of the second generator.

19. A system comprising a first DC bus, a second DC bus, a feeder, a first AC bus, a second AC bus, and first and second transformers each connected with the feeder;

the system further comprising a first inverter coupling the first AC bus with the first DC bus, a second inverter coupling the second AC bus with the second DC bus;

a third inverter coupling the first DC bus with the first transformer;

a fourth inverter coupling the second DC bus with the second transformer;

a first load connected with the first AC bus;

a first generator connected with the feeder, the first generator having a power output level;

a second load connected with the feeder;

a second generator connected with the second DC bus and having a power output level;

the second load consuming power amounting to at least the power output of the first generator;

the first load consuming power amounting to at least a portion of the power output of the second generator;

the first and second AC buses not synchronized with each other;

the feeder not synchronized with either of the first and second AC buses.

20. A method for use with system comprising a first DC bus, a second DC bus, a feeder, a first AC bus, a second AC bus, and first and second transformers each connected with the feeder;

the system further comprising a first inverter coupling the first AC bus with the first DC bus, a second inverter coupling the second AC bus with the second DC bus;

a third inverter coupling the first DC bus with the first transformer;

a fourth inverter coupling the second DC bus with the second transformer;

a first load connected with the first AC bus;

a first generator connected with the feeder, the first generator having a power output level;

a second load connected with the feeder;

a second generator connected with the second DC bus and having a power output level, the method comprising the steps of:

generating power at the first generator at a respective power output level;

generating power at the second generator at a respective power output level, the power generated at the second generator not synchronized with the power generated at the first generator;

at the second load, consuming power amounting to at least the power output of the first generator, the power consumed at the second load not synchronized with the power generated at the first generator and not synchronized with the power generated at the second generator;

and at the first load, consuming power amounting to at least a portion of the power output of the second generator.

21. A system comprising a first DC bus, a second DC bus, a feeder, a first AC bus, a second AC bus, and first and second transformers each connected with the feeder;

the system further comprising a first inverter coupling the first AC bus with the first DC bus, a second inverter coupling the second AC bus with the second DC bus;

a third inverter coupling the first DC bus with the first transformer;

a fourth inverter coupling the second DC bus with the second transformer;

a first load connected with the first AC bus;

a second generator connected with the second DC bus and having a power output level;

the first load consuming power amounting to at least a portion of the power output of the second generator.

22. A method for use with system comprising a first DC bus, a second DC bus, a feeder, a first AC bus, a second AC bus, and first and second transformers each connected with the feeder;

the system further comprising a first inverter coupling the first AC bus with the first DC bus, a second inverter coupling the second AC bus with the second DC bus;

a third inverter coupling the first DC bus with the first transformer;

a fourth inverter coupling the second DC bus with the second transformer;

a load connected with the first AC bus;

a generator connected with the second DC bus and having a power output level, the method comprising the steps of:

generating power at the generator at a respective power output level;

at the load, consuming power amounting to at least a portion of the power output of the generator.

Images