The content of the invention
In order to overcome defect present in above-mentioned prior art, it is an object of the invention to provide a kind of sewage disposal process prisonControl intelligent early-warning cloud system and sewage disposal monitoring and pre-alarming method.
To achieve these goals, the invention provides a kind of sewage disposal process monitoring intelligent early warning cloud system, includingDistributed data acquisition terminal module and/or live photographing unit, in addition to process monitoring embedded data acquisition instrument and cloudCalculation server, the distributed data acquisition terminal module, live photographing unit respectively with the embedded number of the process monitoringCommunicated to connect according to Acquisition Instrument, the process monitoring embedded data acquisition instrument communicates to connect with the cloud computing server.
The system is simple in construction, each stage in sewage disposal process can be monitored, dirt is provided for unit of operationA variety of real time datas of water biochemical process, and be aided with the means taken pictures at scene and intuitively picture prison is carried out to sewage disposal sceneControl, reflects the treatment effect of each biochemical treatment link, contributes to lifting sewage biochemical treatment efficiency in real time.
Further, the distributed data acquisition terminal module includes distributed in-line meter, distributed in-line meterOn-site data gathering device and distributed dynamoelectric plant facility data acquisition unit, the distributed in-line meter on-site data gathering deviceInput connect the distributed in-line meter signal output part, the distributed dynamoelectric plant facility data acquisition unit inputEnd is connected on the terminal plate of electromechanical facility control signal；The distributed in-line meter on-site data gathering device and distributed machineElectric plant facility data acquisition unit is bi-directionally connected with the process monitoring embedded data acquisition instrument respectively, is in communication with each other.
Distributed in-line meter on-site data gathering device collection water quality parameter and pond liquid level, distributed dynamoelectric plant facilityThe data acquisition unit collection waste water control furnished equipments collection water inlet elevator pump in pond, the sludge reflux pump of sedimentation basin, aeration drumThe running state data of blower fan, nonmetallic drug feeding pipeline water sensor control signal and electromechanical facility equipment energy consumption ginsengNumber.
Further, in addition to discrete type PLC, the discrete type PLC are embedded in the process monitoringFormula data collecting instrument is bi-directionally connected, and is in communication with each other；The discrete type PLC signal output part connection is described distributed onlineThe control terminal of meter locale data acquisition unit, distributed dynamoelectric plant facility data acquisition unit and/or electromechanical facility.Discrete typePLC receives the control signal sent from cloud computing server, control by process monitoring embedded data acquisition instrumentThe running status of electromechanical facility involved by the switching value and project of electromechanical facility, will to adapt to the technology of Energy Saving ControlAsk.
Further, the live photographing unit includes image unit and image transmitting subsystem, and the image unit is defeatedGo out end connection described image transmission subsystem input, described image transmission subsystem is adopted with the process monitoring embedded dataCollect instrument communication connection.
Further, in addition to wireless short-distance signal transmission unit, the distributed in-line meter and electromechanical facility divideDo not communicated to connect with the process monitoring embedded data acquisition instrument by the wireless distances signal transmission unit.It is short that this is wirelessDistance signal transmission unit is used for distributed in-line meter, the standardized digital signal of electromechanical facility operating condition and operational factorState is acquired and sent, and solves the distributed in-line meter probe in part or electromechanical facility operating condition collection position installation is tiredThe situation that difficult, power supply can not supply in time with network.
Further, the distributed dynamoelectric plant facility data acquisition unit gathers the lifting in sewage disposal pond in real timeMedicament dosing pump, water needed for pump, the sludge reflux pump of sedimentation basin, the running status of aeration blower and biochemical wastewater treatmentPit level measuring apparatus, nonmetallic drug feeding pipeline water sensor control signal.
Further, in addition to intelligent terminal, the cloud computing server are communicated to connect with intelligent terminal, will surveyed in real timeData and final analysis result and the image photograph at waste water control scene are measured, are shown by intelligent terminal.
Further, in addition to the internet cloud computing server system based on SaaS cloud computing technology frameworks, distributionGathered data receives and non-relational database application service layer, the adjustable high-performance computer resource of virtualization, internet netNetwork, user's intelligent terminal, and by the cloud computing server group system of distributed parallel computing combination, it is acquired data operationThe resource allocation of required adjustment cloud computing server and the combination of each node configure.
Further, distributed type assemblies data acquisition resource layer, computer resource cluster management layer, application service are passed throughLayer and the client terminal display unit of sewage unit of operation show sewage unit of operation.Distributed type assemblies data acquisition resource layerThe data of process monitoring data Acquisition Instrument are integrated and sent by Chinese Ministry of Environmental Protection's national standard (HJT212), computing resource layer pipeReason layer carries out computational resource allocation automatically by data volume is uploaded, to ensure the calculating demand for services of water quality data, and by cloud computingServer realizes that details is encapsulated in same platform, by it is virtual it is single it is overall in the form of be presented to sewage unit of operation.
The invention also provides a kind of sewage disposal prison based on above-mentioned sewage disposal process monitoring intelligent early warning cloud systemMethod for early warning is surveyed, is comprised the following steps：
S1, distributed data acquisition terminal module is to the water quality data in sewage disposal process, each waste water control facilityOperational factor be acquired；Live photographing unit is taken pictures, collection site photo；
S2, water quality data, the operational factor of each waste water control facility, scene photograph are sent to process by network and supervisedEmbedded data acquisition instrument is controlled, the process monitoring embedded data acquisition instrument is again by the water quality data received, each sewageThe operational factor for the treatment of facility, scene photograph are sent to cloud computing server；
S3, cloud computing server is to the water quality data, the operational factor of each waste water control facility and the scene that receivePhoto carry out data access, by project specific sewage treatment process requirements carry out data prediction and data normalization calculating,Data are visualized in intelligent terminal chart and early warning is handled；
S4, acquired results in step S3 are passed through into mobile client terminal display.
Further, the in-line meter is carried out with molten to the reaction tank of sewage bacteria metabolism biochemical process in real timeThe on-line monitoring based on oxygen is solved, the supplemental characteristic of sewage is gathered, is gone with the microorganism of intensively monitoring biochemical processing procedure of sewageExcept the ability of sewage organic pollution, to the ability of dephosphorization denitrification in the biochemistry pool in sewage disposal, the sludge reflux of sedimentation basinRatio, sewage water liquid level data are monitored in real time；
Distributed dynamoelectric plant facility data acquisition unit is acquired to the operational factor of each waste water control facility in real time.
Further, the step S3 comprises the following steps：
S3-1, cloud computing server is to the water quality data, the operational factor of each waste water control facility and the scene that receivePhoto as data type be divided into water monitoring data, equipment operational factor digital switch quantity, scene captured by photo dividedClass stores；
S3-2, through core pivot element analysis algorithm, independent composition analysis algorithm, cluster algorithm, algorithm of support vector machine withThe multi-data processing technology that Fuzzy Artificial Neural Networks algorithm is mutually combined, establish the process data normalizing of sewage disposal processStandardize dimension-reduction treatment and principal component analysis processing operational equation；What the data in data queue of distributed in-line meter was formedMore vector matrixs carry out Data Dimensionality Reduction, the characteristic vector to matrix is handled；The process monitoring data of sewage disposal is subjected to featureVector calculates, and the learning algorithm of fuzzy neural network is the further evolution of BP neural network algorithm, and fuzzy neural network is divided intoInput layer, blurring layer, fuzzy reasoning layer composition, the data learning sewage by fuzzy neural network algorithm from characteristic vectorBiochemical moving law is managed, dissolved oxygen setting value is combined in terms of sewage disposal biochemical aeration and carries out in-line meter control optimizing, mouldPaste neutral net is predicted using the non-linear rule learnt to unknown effluent quality data, establishes prediction effluent qualityOptimization artificial intelligence neural networks model, feature learning is carried out to the characteristic vector sample of sewage disposal, with reference to what is providedSample data is with carrying out contrast test；
S3-3, after process monitoring data information is handled by S3-2 in the process monitoring content of sewage disposal fault detect,Extract and portray from process monitoring data information by supporting vector grader, fuzzy neural network, radial base neural netThe process feature signal of process operation characteristic, neural network learning classification is then carried out to process feature data message to determine dirtWater treatment procedure run shape gesture, gathered data according to different sewage treatment process using a variety of neural network ensembles study,After prediction, the sewage disposal sludge biochemical process runtime failure caused by important data exception is investigated in time, is avoidedSludge accident in sewage disposal process by activated sludge process and cause the accident of sewage disposal to occur.Clearly reflect at biochemical sewageManage the dynamic relationship and sewage disposal operation situation between leading variable and the auxiliary variable based on dissolved oxygen；
S3-4, in terms of effluent quality model, the data of Kernel-based methods monitoring data collection instrument and the mathematics system that deploysMeter recurrence, pivot analysis and blind source algorithm, support vector machine classifier, fuzzy neural network, radial base neural net mutually groupThe data that biochemical wastewater treatment is gathered through distributed in-line meter are entered into line number in the multivariate data water quality model of conjunctionDimension-reduction treatment and the processing of pivot analysis dimensionality reduction mathematical statistics are standardized according to normalizing, clear process operation is extracted from process dataThe process feature signal of characteristic, with mainly to the microorganism based on the aerobic bacteria of sewage disposal biochemical reaction, by aerobic bacteriaGrowth course and aeration tank in dissolved oxygen in-line meter relevance come to aerator carry out intelligent control, with radiallyBase neural net, SVMs, fuzzy neural network combination are classified to Process character information to determine at sewageProcess operation shape gesture is managed, the gathered data progress effluent quality and process monitoring that provided COD, BOD are monitored with reference to end are adoptedCollect the prediction of data Nonlinear Classification；
Fuzzy neural network parameter learning control based on the dissolved oxygen in-line meter parameter of sewage disposal biochemical reactionSimulation, become in biochemical reaction amount of inlet water and liquid level, influent COD, water inlet total phosphorus and total nitrogen, water inlet PH, water inlet sludge suspension thingDuring change, gathered data new data sample is added in original training sample, the online updating of sample is realized, then passes through S3-2Described Processing with Neural Network combination carries out computing and optimizing in cloud computing server to freshly harvested training sample, by computing knotFruit provides the electric discrete type PLC of discrete type and is controlled different dosing regimen and aeration time, to reach in water outletThe operational factor that water quality reduces fluctuation is stable.
According to different sewage treatment process, it is combined by pivot analysis, independent component analysis, fuzzy neural network,The optimization artificial intelligence neural networks model of prediction effluent quality is established, solves the in-line meter number in biochemical processing procedure of sewageAccording to non-linear relation existing between effluent quality, data basis is provided for the fitting of nonlinear data system.
The present invention is given birth to by the real-time collection to in-line meter, electromechanical facility operating condition, live timing photo to sewageChange processing procedure overall process in-line meter and carry out overall monitor with electromechanical facility operating condition, have data source comprehensively excellentPoint, gather and carry out data and photo carries out carrying out computing by cloud computing server by biochemical wastewater treatment process specifications,The operation result of cloud computing server is effectively assessed the links of biochemical sewage system, ensures biochemical wastewater treatmentEffluent quality is fluctuated in rational scope, while also provides data foundation for biochemical wastewater treatment, also by process optimizationRealize that energy-saving and emission-reduction provide reference data.
Based on SaaS cloud computing server technologies, integrated use cloud computing server technology, big data NOSQL dataStorehouse technology, Internet computing technique, in-line meter acquisition technique, make gathered data resource layer, calculation resources layer, using clothesBusiness layer and customer service layer cross-correlation, the core pivot element analysis for being acquired data in cloud computing server with gathered data are calculatedWhat method, independent composition analysis algorithm, cluster algorithm, algorithm of support vector machine and BP artificial neural network algorithms be combined with each otherMulti-data processing technology and with effluent quality COD, BOD5, total phosphorus, total nitrogen nonlinear model relation in cloud computing serverCarrying out data operation is associated, and operation result is fed back into sewage disposal system electric automatic control unit provides control signal.
In terms of sewage disposal process monitoring, exist for sludge biochemical profile, dissolved oxygen, PH of biochemical waste water treatment pool etc.Line instrument carries out procedure parameter, and monitoring is simultaneously to sewage treatment facility equipment running monitoring in real time, to solve biochemical wastewater treatment workThe sludge bio-chemical accidents of Frequent Troubles in skill flow, and sewage treatment process process and sewage treatment facility are become extremelyChange, in-line meter sensing data are monitored in real time extremely, and operation of sewage disposal system operating mode is carried out into real-time fault detection,And the failure being born to sewage disposal system carries out quantitative analysis, distinguishes fault type, according to different sewage treatment processMultivariable application pivot analysis algorithm, cluster algorithm, corresponding with the algorithm proposition that artificial neural network be combined with each otherEmergency management, sewage disposal system maintenance prevention fault detection and diagnosis measure；
In terms of wastewater treatment effluent quality early warning measurement, establish the data of Kernel-based methods monitoring data collection instrument and deployMathematical statistics return, pivot analysis and blind source algorithm, support vector machine classifier, fuzzy neural network, radial direction base nerve netMultivariate data computing mode that network is mutually combined and with effluent quality COD non-linear interrelated system, establish at sewageNon-linear relation model between reason and artificial intelligence model and the electric discrete automatic control system of sewage disposal operation, to electricDiscrete automatic control system provides real-time control signal, while is supplied to sewage unit of operation sewage by Internet interfaceProcessing system operation conditions and warning information.
The additional aspect and advantage of the present invention will be set forth in part in the description, and will partly become from the following descriptionObtain substantially, or recognized by the practice of the present invention.
Embodiment
Embodiments of the invention are described below in detail, the example of the embodiment is shown in the drawings, wherein from beginning to endSame or similar label represents same or similar element or the element with same or like function.Below with reference to attachedThe embodiment of figure description is exemplary, is only used for explaining the present invention, and is not considered as limiting the invention.
In the description of the invention, unless otherwise prescribed with limit, it is necessary to explanation, term " installation ", " connected ", "Connection " should be interpreted broadly, for example, it may be mechanically connect or electrical connection or the connection of two element internals, can be withIt is to be joined directly together, can also be indirectly connected by intermediary, for the ordinary skill in the art, can be according to toolBody situation understands the concrete meaning of above-mentioned term.
A kind of sewage disposal process monitoring intelligent early warning cloud system as Figure 1-3, including distributed data acquisition are wholeEnd module and/or live photographing unit, in addition to process monitoring embedded data acquisition instrument and cloud computing server, described pointCloth data collection station module, live photographing unit communicate to connect with the process monitoring embedded data acquisition instrument respectively,The process monitoring embedded data acquisition instrument communicates to connect with the cloud computing server, here distributed data acquisition terminalModule can be multiple.
The distributed data acquisition terminal module includes distributed in-line meter, distributed in-line meter field data is adoptedStorage and distributed dynamoelectric plant facility data acquisition unit, the input of the distributed in-line meter on-site data gathering device connectThe distributed in-line meter signal output part is connect, the distributed dynamoelectric plant facility data acquisition unit input is connected to machineOn the terminal plate of electric facility control signal；The distributed in-line meter on-site data gathering device and distributed dynamoelectric plant facilityData acquisition unit is bi-directionally connected with the process monitoring embedded data acquisition instrument respectively, specifically can be mutual by ethernet local area networkIt is bi-directionally connected, is in communication with each other.
Distributed in-line meter on-site data gathering device, distributed dynamoelectric plant facility data acquisition unit, scene are taken pictures listMember sends the data each gathered, picture to cloud computing server, cloud computing by process monitoring embedded data acquisition instrumentServer carries out data parsing, data classification to the data that receive, aggregation of data computing, data loading, entered by procedural modelRow classification etc., and shown by intelligent terminal, involved data algorithm uses existing this area in the present inventionGeneral-purpose algorithm.
In order to more intuitively show data that the monitoring system is gathered and is calculated, the monitoring system also includes intelligenceTerminal, the cloud computing server communicates to connect with intelligent terminal, by real-time measurement data and final analysis result and dirtThe image photograph at water harnessing scene, is shown by intelligent terminal.
Cloud computing server is included the data transmission unit collected, cloud processing server data based on the Internet transmissionReceive and communication front end unit, NOSQL big data application data library units and application service software and back-stage management processing unit.
Process monitoring embedded data acquisition instrument passes through the data gathered packing as the client in LAN mutualNetworking uploads to cloud computing server, and cloud computing server is simultaneously and concurrently asked by multithread pool big data at the data of responseReason is required, the data and photo of upload are received, and is carried out data parsing by the form of communications protocol and is saved in NOSQL numbersAccording in storehouse, it is necessary to transmission by being controlled order to process monitoring data Acquisition Instrument；By intelligent terminal by cloud meterCalculate the chart after the computing of server and referential data carries out real-time exhibition.
The system also includes discrete type PLC, the discrete type PLC and the embedded number of the process monitoringIt is bi-directionally connected according to Acquisition Instrument, specifically, discrete type PLC passes through process monitoring embedded data acquisition instrument open protocol endMouth is bi-directionally connected with process monitoring embedded data acquisition instrument by ethernet local area network, is in communication with each other；The discrete type PLC controlsDevice control output end connects the distributed in-line meter on-site data gathering device, distributed dynamoelectric plant facility data acquisition unitAnd/or the control terminal of electromechanical facility.Discrete type PLC is received by process monitoring embedded data acquisition instrument and comes from cloud meterThe digital output modul signal that server is sent is calculated, controls distributed in-line meter on-site data gathering device, distributed dynamoelectric facilityThe switching value of on-site data gathering device and/or electromechanical facility.
The in-line meter of in-line meter monitoring unit be arranged on each pond reaction tank, regulating reservoir, adjust back pond, concentration basin,In the specified location of Aerobic Pond, anoxic pond, anaerobic pond or floss hole, by outputting standard signal (4- caused by in-line meter20mA、DC：0-10V, RS-485) these signals are acquired by in-line meter distributable field data acquisition unit, and lead toCross ethernet local area network the data collected are sent on process monitoring embedded data acquisition instrument.In-line meter is distributed existingField data collector can be that one or more is acquired to in-line meter signal.In-line meter monitoring unit is given birth to each sewageChange the PH, ORP, DO, conductance of processing pond (reaction tank, regulating reservoir, adjusting back pond, concentration basin, Aerobic Pond, anoxic pond, anaerobic pond)The parameter such as rate, water temperature, TOC, COD is acquired.
For solve part distributed in-line meter probe or electromechanical facility operating condition gather position installation difficulty, power supply withThe situation that network can not be supplied in time, the system also include wireless short-distance signal transmission unit, the distributed in-line meterCommunicated respectively with the process monitoring embedded data acquisition instrument by the wireless distances signal transmission unit with electromechanical facilityConnection.The wireless short-distance signal transmission unit is used for the standard digital of distributed in-line meter, electromechanical facility operating conditionSignal is acquired and sent with operational factor state.Here wireless short-distance signal transmission unit uses but is not limited to LORAWireless short-distance communication module.
In-line meter is led to when installing inconvenience caused by installation site or power supply and network reason by LORA wireless short-distancesModule is interrogated, the Monitoring Data of in-line meter is acquired and sent, solar energy can be used in the case where power supply does not possessIndependently-powered mode is powered, to ensure LORA wireless short-distance communication module energy normal power supplies.
In addition, process monitoring embedded data acquisition instrument gathers different types of in-line meter field data, and complete numberAccording to preservation with transmitting gathered data bag to cloud computing server.Cloud computing server is by HJT212 standards to installing at the sceneProcess monitoring embedded data acquisition instrument send inquiry or control instruction, cloud computing server is at data receiver and dataReason, data statistics, the system of data prediction.Data processing and computing referred to herein are using existing computational methodsCan.Distributed in-line meter on-site data gathering device can use flexible interface mode with distributed in-line meter：Pass through numberWord interface signal is connected with in-line meter, or is connected using industrial connector terminals with in-line meter.Wherein, digital interface bagInclude：RS232、RS485、RS422、4-20mA、DC(0-10V).Interface type：Meet ISA, IEC, IFC standard.RS232 serial ports：Meet EIA RS-232 interface standards；RS485 interfaces：Meet EIA RS-485 interface standards.
Distributed dynamoelectric plant facility data acquisition unit is arranged in Central Control Room or field control electrical control cubicles, to main machineThe running status of electric facility carries out the collection of digital switch quantity, using industrial connector terminals to passive/active signal collection sideFormula is monitored to the state of electromechanical facility, and records start-stop time of electromechanical facility, run time, the energy consumption of electromechanical equipmentComputing.
Particularly, distributed dynamoelectric plant facility data acquisition unit gathers the elevator pump in biochemical pond, sedimentation basin in real timeMedicament dosing pump, pond liquid level measurement needed for sludge reflux pump, the running status of aeration blower and biochemical wastewater treatmentEquipment, nonmetallic drug feeding pipeline water sensor control signal.Pass through distributed dynamoelectric plant facility data acquisition unit pairDosing pump and pond liquid level measuring apparatus are monitored with the control signal of the water sensor of nonmetallic drug feeding pipeline, effectivelyRationally whether dosing and drug feeding pipeline timely when wherein a side is faulty in corresponding control in management and control biochemical wastewater treatment linkFault-signal is sent by distributed data acquisition module and paints process monitoring embedded data acquisition instrument, is embedded in by process monitoringFormula data collecting instrument is transmitted cloud computing server platform by HJ212 communications protocol requirements, is carried out interrelated logic and is sentencedDisconnected and processing.
In the present embodiment, distributed dynamoelectric plant facility data acquisition unit is accompanied with multiple standards outputting communication agreementOn-line monitoring instrument.And electromechanical facility operating condition data are stored in the database of process monitoring embedded data acquisition instrumentIn.
Live photographing unit includes image unit and image transmitting subsystem, and the image unit output end connects the figureAs transmission subsystem input, described image transmission subsystem communicates to connect with the process monitoring embedded data acquisition instrument.
Live photographing unit is timed to scene according to set time takes pictures, then according to set time will be capturedScene photograph is sent to process monitoring embedded data acquisition instrument, process monitoring embedded data by Ethernet transmission subsystemTaken photo is automatically forwarded to be retained in cloud computing server by Acquisition Instrument by internet, is finding there is photo pictureFind to alarm by intelligent terminal during invasion during the exception of face.
The system also includes internet cloud computing server system, distributed capture based on SaaS cloud computing technology frameworksData receiver and non-relational database application service layer, the adjustable high-performance computer resource of virtualization, Internet, useFamily intelligent terminal, and by the cloud computing server group system of distributed parallel computing combination, be acquired needed for data operationThe resource allocation of cloud computing server is adjusted to configure with the combination of each node.By distributed type assemblies data acquisition resource layer,Computer resource cluster management layer, application service layer and the client terminal display unit of sewage unit of operation show sewage operationUnit.The data of process monitoring data Acquisition Instrument are pressed Chinese Ministry of Environmental Protection's national standard by distributed type assemblies data acquisition resource layer(HJT212) integrated and sent, computing resource layer-management layer carries out computational resource allocation automatically by data volume is uploaded, to protectThe calculating demand for services of water quality data is demonstrate,proved, and cloud computing server is realized that details is encapsulated in same platform, with virtualSingle overall form is presented to sewage unit of operation.
The invention also provides a kind of sewage disposal prison based on above-mentioned sewage disposal process monitoring intelligent early warning cloud systemMethod for early warning is surveyed, is comprised the following steps：
S1, distributed data acquisition terminal module is to the water quality data in sewage disposal process, each waste water control facilityOperational factor be acquired；Live photographing unit is taken pictures to sewage disposal scene, collection site photo；
In-line meter is arranged in the specified location of each reaction tank biochemical wastewater treatment course of reaction carried out in real timeOn-line real time monitoring, the data of in-line meter are carried out by one or more distributed in-line meter on-site data gathering device real-timeCollection, while the data of the in-line meter collected are sent to the process monitoring embedded data with net in ether localAcquisition Instrument.
Distributed dynamoelectric plant facility data acquisition unit gathers, monitored the fortune of each biochemical wastewater treatment electromechanical facility in real timeRow state and operational factor are acquired, and electromechanical facility is transported by one or more distributed dynamoelectric plant facility data acquisition unitThe control signal of row operating mode is gathered in real time, generates biochemical wastewater treatment electromechanical facility service data, while be sent to processMonitor embedded data acquisition instrument.
The distributed dynamoelectric plant facility data acquisition unit gathers water inlet pond, ties up elevator pump, the sedimentation basin in pond in real timeSludge reflux pump, the running status and dosing pump, pond liquid level measuring apparatus, nonmetallic drug feeding pipeline of aeration blowerThe control signal of water sensor.
S2, water quality data, the operational factor of each waste water control facility, scene photograph are sent to process by network and supervisedEmbedded data acquisition instrument is controlled, the process monitoring embedded data acquisition instrument is again by the water quality data received, each sewageThe operational factor for the treatment of facility, scene photograph are sent to cloud computing server；
Live photographing unit is by the control instruction of process monitoring embedded data collection instrument in the self-defined tune of man-machine dialog interfaceTake pictures when adjusting interval time, self-timing can temporally show captured after setting timing photo opporunity interval is completedField photo is sent to process monitoring embedded data acquisition instrument.
S3, cloud computing server is to the water quality data, the operational factor of each waste water control facility and the scene that receivePhoto carry out data access, by project specific sewage treatment process requirements carry out data prediction and data normalization calculating,Data are visualized in intelligent terminal chart and early warning is handled.Here cloud computing server is to the water quality data, each that receivesThe operational factor and scene photograph of waste water control facility can be uploaded with JT212-2005 communication protocol standards and data parse.
S4, acquired results in step S3 are passed through into mobile client terminal display.
The in-line meter is carried out using dissolved oxygen as base to the reaction tank of sewage bacteria metabolism biochemical process in real timeThe on-line monitoring of plinth, the data such as SS, COD of sewage, ORP, PH, electrical conductivity, DO, water temperature are gathered, with intensively monitoring biochemical sewageThe microorganism of processing procedure removes the ability of sewage organic pollution, to the energy of dephosphorization denitrification in the biochemistry pool in sewage disposalPower, the sludge reflux ratio of sedimentation basin, sewage water liquid level data are monitored in real time；To pass through corresponding in-line meter dataThrough the hard measurement mathematical modelling algorithms based on artificial neural network, draw the parameter comparison of corresponding effluent quality and carry out artificialIntelligent early-warning and machine learning, nitrification/denitrification capacity of biochemical wastewater treatment is assessed, at the same for it is follow-up electrically fromDynamic control provides data basis with medicine system.Distributed dynamoelectric plant facility data acquisition unit is set to each waste water control in real timeThe operational factor applied is acquired.
Cloud computing server will receive the data for the real-time online instrument sent by process monitoring embedded data acquisition instrumentWith the operational parameter data being calculated by data analysis, and the becoming of being showed of in-line meter Monitoring Data according to uploadThe contents such as gesture form.Judge whether corresponding biochemical wastewater treatment technique closes by the technical matters flow of biochemical wastewater treatment simultaneouslyThe relevance data analyses such as reason carry out Data Integration and classified with data, and the relevant parameter collected is by cloud computing server by phaseThe statistics and the data operation rule of artificial intelligence answered carry out computing and classification learning to the data collected, calculate correspondingDuring numerical value, the system provides corresponding early warning, alarm content is shown and prompted by client wechat platform.
The step S3 comprises the following steps：
S3-1, cloud computing server is to the water quality data, the operational factor of each waste water control facility and the scene that receivePhoto as data type be divided into water monitoring data, equipment operational factor digital switch quantity, scene captured by photo dividedClass stores；
S3-2, through core pivot element analysis algorithm, independent composition analysis algorithm, cluster algorithm, algorithm of support vector machine withThe multi-data processing technology that Fuzzy Artificial Neural Networks algorithm is mutually combined, establish the process data normalizing of sewage disposal processStandardize dimension-reduction treatment and principal component analysis processing operational equation；What the data in data queue of distributed in-line meter was formedMore vector matrixs carry out Data Dimensionality Reduction, the characteristic vector to matrix is handled；The process monitoring data of sewage disposal is subjected to featureVector calculates, and the learning algorithm of fuzzy neural network is the further evolution of BP neural network algorithm, and fuzzy neural network is divided intoInput layer, blurring layer, fuzzy reasoning layer composition, the data learning sewage by fuzzy neural network algorithm from characteristic vectorBiochemical moving law is managed, dissolved oxygen setting value is combined in terms of sewage disposal biochemical aeration and carries out in-line meter control optimizing, mouldPaste neutral net is predicted using the non-linear rule learnt to unknown effluent quality data, establishes prediction effluent qualityOptimization artificial intelligence neural networks model, feature learning is carried out to the characteristic vector sample of sewage disposal, with reference to what is providedSample data is with carrying out contrast test；
S3-3, in the process monitoring content of sewage disposal fault detect, after process monitoring data information is handled by S3-2,Extract and portray from process monitoring data information by supporting vector grader, fuzzy neural network, radial base neural netThe process feature signal of process operation characteristic, neural network learning classification is then carried out to process feature data message to determine dirtWater treatment procedure run shape gesture, gathered data according to different sewage treatment process using a variety of neural network ensembles study,After prediction, the sewage disposal sludge biochemical process runtime failure caused by important data exception is investigated in time, is avoidedSludge accident in sewage disposal process by activated sludge process and cause the accident of sewage disposal to occur.Clearly reflect at biochemical sewageManage the dynamic relationship and sewage disposal operation situation between leading variable and the auxiliary variable based on dissolved oxygen；
S3-4, in terms of effluent quality model, the data of Kernel-based methods monitoring data collection instrument and the mathematics system that deploysMeter recurrence, pivot analysis and blind source algorithm, support vector machine classifier, fuzzy neural network, radial base neural net mutually groupThe data that biochemical wastewater treatment is gathered through distributed in-line meter are entered into line number in the multivariate data water quality model of conjunctionDimension-reduction treatment and the processing of pivot analysis dimensionality reduction mathematical statistics are standardized according to normalizing, clear process operation is extracted from process dataThe process feature signal of characteristic, with mainly to the microorganism based on the aerobic bacteria of sewage disposal biochemical reaction, by aerobic bacteriaGrowth course and aeration tank in dissolved oxygen in-line meter relevance come to aerator carry out intelligent control, with radiallyBase neural net, SVMs, fuzzy neural network combination are classified to Process character information to determine at sewageProcess operation shape gesture is managed, the gathered data progress effluent quality and process monitoring that provided COD, BOD are monitored with reference to end are adoptedCollect the prediction of data Nonlinear Classification；
Fuzzy neural network parameter learning control based on the dissolved oxygen in-line meter parameter of sewage disposal biochemical reactionSimulation, become in biochemical reaction amount of inlet water and liquid level, influent COD, water inlet total phosphorus and total nitrogen, water inlet PH, water inlet sludge suspension thingDuring change, gathered data new data sample is added in original training sample, the online updating of sample is realized, then passes through S3-2Described Processing with Neural Network combination carries out computing and optimizing in cloud computing server to freshly harvested training sample, by computing knotFruit provides the electric discrete type PLC of discrete type and is controlled different dosing regimen and aeration time, to reach in water outletThe operational factor that water quality reduces fluctuation is stable.
In the description of this specification, reference term " one embodiment ", " some embodiments ", " example ", " specifically showThe description that example " or " some examples " waits means specific features, structure, material or the spy with reference to the embodiment or example descriptionPoint is contained at least one embodiment or example of the present invention.In this manual, to the schematic representation of above-mentioned term notNecessarily refer to identical embodiment or example.Moreover, specific features, structure, material or the feature of description can be anyOne or more embodiments or example in combine in an appropriate manner.
Although an embodiment of the present invention has been shown and described, it will be understood by those skilled in the art that：NotIn the case of departing from the principle and objective of the present invention a variety of change, modification, replacement and modification can be carried out to these embodiments, thisThe scope of invention is limited by claim and its equivalent.