CN117348400A - Water network multichannel parallel linkage control system - Google Patents
Water network multichannel parallel linkage control system Download PDFInfo
- Publication number
- CN117348400A CN117348400A CN202311393936.0A CN202311393936A CN117348400A CN 117348400 A CN117348400 A CN 117348400A CN 202311393936 A CN202311393936 A CN 202311393936A CN 117348400 A CN117348400 A CN 117348400A
- Authority
- CN
- China
- Prior art keywords
- water
- water level
- control
- gate opening
- gate
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 title claims abstract description 399
- 238000013528 artificial neural network Methods 0.000 claims abstract description 61
- 238000012384 transportation and delivery Methods 0.000 claims abstract description 61
- 238000010248 power generation Methods 0.000 claims abstract description 56
- 238000000034 method Methods 0.000 claims abstract description 46
- 238000004088 simulation Methods 0.000 claims abstract description 30
- 230000006870 function Effects 0.000 claims description 31
- 230000008859 change Effects 0.000 claims description 26
- 230000001276 controlling effect Effects 0.000 claims description 17
- 239000011159 matrix material Substances 0.000 claims description 16
- 238000005457 optimization Methods 0.000 claims description 13
- 230000001105 regulatory effect Effects 0.000 claims description 13
- 230000003044 adaptive effect Effects 0.000 claims description 11
- 230000009471 action Effects 0.000 claims description 10
- 238000011217 control strategy Methods 0.000 claims description 7
- 238000004364 calculation method Methods 0.000 claims description 5
- 238000010521 absorption reaction Methods 0.000 claims description 4
- 230000008569 process Effects 0.000 abstract description 14
- 238000012544 monitoring process Methods 0.000 description 18
- 238000010586 diagram Methods 0.000 description 10
- 238000005381 potential energy Methods 0.000 description 7
- 239000013598 vector Substances 0.000 description 7
- 230000006399 behavior Effects 0.000 description 6
- 230000006872 improvement Effects 0.000 description 6
- 238000004422 calculation algorithm Methods 0.000 description 5
- 238000004590 computer program Methods 0.000 description 5
- 238000010276 construction Methods 0.000 description 5
- 230000010485 coping Effects 0.000 description 5
- 238000009826 distribution Methods 0.000 description 5
- 230000005611 electricity Effects 0.000 description 5
- 238000012545 processing Methods 0.000 description 5
- 238000004891 communication Methods 0.000 description 4
- 238000005259 measurement Methods 0.000 description 4
- 238000011144 upstream manufacturing Methods 0.000 description 4
- 230000008901 benefit Effects 0.000 description 3
- 230000006378 damage Effects 0.000 description 3
- 238000013461 design Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 230000005540 biological transmission Effects 0.000 description 2
- 230000007547 defect Effects 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 238000003973 irrigation Methods 0.000 description 2
- 230000002262 irrigation Effects 0.000 description 2
- 238000013178 mathematical model Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000008447 perception Effects 0.000 description 2
- 238000013468 resource allocation Methods 0.000 description 2
- 238000003860 storage Methods 0.000 description 2
- 230000006978 adaptation Effects 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000019771 cognition Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 239000012530 fluid Substances 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 239000000178 monomer Substances 0.000 description 1
- 230000006855 networking Effects 0.000 description 1
- 238000013439 planning Methods 0.000 description 1
- 230000036632 reaction speed Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 230000002194 synthesizing effect Effects 0.000 description 1
- 230000007704 transition Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B13/00—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion
- G05B13/02—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric
- G05B13/04—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric involving the use of models or simulators
- G05B13/042—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric involving the use of models or simulators in which a parameter or coefficient is automatically adjusted to optimise the performance
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02E—REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
- Y02E10/00—Energy generation through renewable energy sources
- Y02E10/20—Hydro energy
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Computation (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- Feedback Control In General (AREA)
Abstract
The invention discloses a water network multichannel parallel linkage control system, which comprises a multichannel parallel water delivery subsystem, wherein the multichannel parallel water delivery subsystem is used for delivering water in a multichannel parallel water delivery mode; the parallel linkage simulation subsystem for water supply and power generation is used for simulating the channel water level by utilizing a hydrodynamic model and combining a water level error model prediction control method to obtain gate opening information required by controlling the multi-channel parallel water delivery subsystem to reach a preset water level; the water supply and power generation parallel linkage control subsystem is used for calculating a control rule value by using a PID control method of the self-adaptive Fourier series neural network, controlling the channel gate according to the control rule value and feeding back the water delivery state information of the multi-channel parallel water delivery subsystem to the water supply and power generation parallel linkage simulation subsystem. The invention can realize the processes of automatic control, feedback and readjustment of the gate, greatly reduces the labor cost and improves the control efficiency of the gate.
Description
Technical Field
The invention relates to the technical field of diversion engineering, in particular to a water network multichannel parallel linkage control system.
Background
The water network is connected by each channel, the channels in the water network have various structures and types, wherein the parallel connection of water supply and power generation is one of the parallel connection types of multiple channels, and the parallel connection type of water supply and power generation is also one with higher requirements on a control system. The complete control system is needed for realizing the parallel linkage control of water supply and power generation, and the control system needs to have clear perception on the whole condition of the diversion project. This requires the construction of a suitably controlled engineering structure, and in addition, sensors are deployed in the engineering for acquiring data and monitoring the operating conditions, and the control system controls the gate after acquiring water level data.
In the prior art, a manual control mode is mainly adopted for gate control, and although some researches on automatic control and online control of the gate are carried out, the technology only realizes that machinery replaces manual direct operation of the gate, and the operation of the gate still depends on empirical control in terms of the nature or manual control.
The water network multi-channel combined automatic control is a multi-target nonlinear system from the mathematical modeling perspective, and from the aspects of artificial intelligence and real-time control, the system needs to be regarded as a whole, the stability of the whole is ensured, a plurality of nodes in the system are accurately regulated, and the nodes are in linkage control to play a role together. Also, engineering systems and natural systems need to be coordinated with each other, and only then can not only meet the requirements of engineering per se but also develop continuously, and the environment around the engineering systems and the natural systems are not damaged in a disastrous way.
Along with the development of the internet of things, the application of the internet of things in channel gate control is gradually developed. Adopt thing networking technology design gate intelligent control system can realize the remote control of irrigated area shunting, real-time supervision, improved the management level that irrigates the district shunting irrigation, improve the district water resource utilization ratio. However, the system still has some defects, on one hand, the issuing of the operation instruction is performed manually; on the other hand, the control of the gate depends on judgment of manual knowledge experience, and the gate control is still lacking in scientific and reasonable aspects.
The measurement and control integrated gate integrates the gate, the opening and closing equipment, the flow measuring equipment, the control equipment and the power supply equipment, integrates the gate opening and closing, the flow calculation, the remote control and the communication functions, combines the calculation of the gate opening, the channel water level, the instantaneous flow and the time period water quantity, remotely performs the measurement and control of the canal gate through a computer and a communication network system, or realizes the automatic adjustment of the water distribution quantity of the gate under the given flow water level or opening, and realizes the automation of the channel water measurement section or the direct opening measurement and control flow. However, the problem of automatic control of the gate monomers solved by the technology relates to the combined application of two or more gates, and in practical application, the situations of excessive water quantity, insufficient water supply or water level fluctuation are possibly caused, so that a combined automatic control method is needed.
Disclosure of Invention
Aiming at the defects in the prior art, the invention provides a water network multichannel parallel linkage control system.
In order to achieve the aim of the invention, the invention adopts the following technical scheme:
a water network multichannel parallel linkage control system, comprising:
the multichannel parallel water delivery subsystem is used for delivering water in a multichannel parallel water delivery mode;
the parallel linkage simulation subsystem for water supply and power generation is used for acquiring water delivery state information of the multi-channel parallel water delivery subsystem, simulating channel water level by utilizing a hydrodynamic model and combining a water level error model prediction control method, and acquiring gate opening information required by controlling the multi-channel parallel water delivery subsystem to reach a preset water level;
the water supply and power generation parallel linkage control subsystem is used for acquiring gate opening information output by the water supply and power generation parallel linkage simulation subsystem, calculating a control rule value by using a PID control method of the self-adaptive Fourier series neural network, controlling the channel gate according to the control rule value, and feeding back water delivery state information of the multi-channel parallel water delivery subsystem to the water supply and power generation parallel linkage simulation subsystem.
Optionally, the multi-channel parallel water delivery subsystem specifically includes:
The water inlet of the first main channel and the water inlet of the second main channel are respectively provided with a water diversion gate, and a generator set is arranged in the first main channel; the water outlets of the first main channel and the second main channel are communicated with the stilling pool, and the water outlet of the stilling pool is communicated with the third main channel.
Optionally, the water delivery state information of the multichannel parallel water delivery subsystem acquired by the water supply and power generation parallel linkage simulation subsystem specifically includes:
gate opening information of the water diversion gate, water level information of the first main channel, the second main channel and the third main channel, and flow information of the generator set in the first main channel.
Optionally, the water supply and power generation parallel linkage simulation subsystem simulates the channel water level by using a hydrodynamic model in combination with a water level error model prediction control method, and the gate opening information required by controlling the multi-channel parallel water delivery subsystem to reach a preset water level specifically comprises:
the water level information of each main channel and the gate opening information of each water diversion gate are read by utilizing a hydrodynamic model, the gate opening information of each water diversion gate is used as an input variable, the water level information of each main channel is used as an output variable, and the change of the channel water level is regulated by controlling the change of the gate opening;
Meanwhile, the water level information of each main channel at the current moment is obtained by utilizing a water level error model prediction control method, Q, R parameter values are determined according to the obtained water level information, a time step to be predicted, a prediction interval and an expected value of an error are set, a gate opening degree information is used as a control variable to construct an optimization objective function of the water level error, an integrator delay model is used for establishing a differential error equation of the stilling pool, a corresponding control strategy is generated according to the range of the prediction interval, the water level error value of the prediction interval is obtained, and iterative optimization is carried out according to the comparison result of the water level error value of the prediction interval and the actual water level error value.
Optionally, the optimization objective function for constructing the water level error by using the gate opening information as the control variable is specifically:
wherein U is * For the control sequence of the gate opening actual application, U is the control sequence of the predicted gate opening, x 0 For initial state of system evolution according to applied action sequence, J (U, x 0 ) As a cost function, k is a time step, N h For the predicted water level value, x (k) is Q, R is a constant weighting matrix for the quadratic deviation penalty, Q l Constant weighting matrix for linear penalty, T is transposed sign, u (k) is gate opening, x (N) h ) Is the final error between the control result and the target in the prediction interval.
Optionally, the differential error equation for establishing the stilling pool by adopting the integrator delay model is specifically:
wherein D is j E is the differential error of the stilling pool j For the water level error at the j position of the stilling pool, n is the total number of stilling pools, e i Is the water level error at the other stilling pool except the stilling pool j.
Optionally, the water supply and power generation parallel linkage control subsystem calculates the control rule value by using a PID control method of the adaptive Fourier series neural network specifically comprises the following steps:
constructing a gate linkage controller by using a PID controller, a multi-input multi-output Fourier series neural network, a multi-input single-output Fourier series neural network and a system controller;
the PID controller establishes an optimal objective function according to the difference value between the input current gate opening value and the target gate opening value, and combines the gain parameters output by the multi-input multi-output Fourier series neural network to output the gate opening value subjected to PID regulation to the system controller;
the system controller controls the water diversion gate according to the gate opening value regulated by the PID, and outputs the current gate opening value of the system;
the multi-input multi-output Fourier series neural network calculates an approximate value of the jacobian matrix system according to the neural network connection weight, then calculates the self-adaptive neural network bias and the neural network connection weight according to the neural network self-adaptive equation, and finally calculates gain parameters of the PID controller according to the self-adaptive neural network bias and the neural network connection weight, and outputs the gain parameters to the PID controller.
Optionally, the optimal objective function established by the PID controller is specifically:
e(k)＝R(k)-y(k)
wherein E (k) is an optimal objective function, E (k) is a difference between a current gate opening value and a target gate opening value, R (k) is the current gate opening value, and y (k) is the target gate opening value.
Optionally, the neural network adaptive equation is specifically:
wherein,bias for adaptive tuned neural network, +.>For the original neural network bias, η is the learning rate and e (k) is the current gateThe difference between the gate opening value and the target gate opening value, y (k) is the target gate opening value, u (k) is the gate opening value after PID adjustment, O h (k) For the gain parameter of the PID controller, +.>Connecting weights for the self-adaptive neural network, < ->Connecting weights for the original neural network, H j Is an intermediate variable.
Optionally, the calculating manner of calculating the gain parameter of the PID controller according to the adaptively adjusted network bias and the network connection weight is as follows:
wherein O is h Is a gain parameter of the PID controller,for self-adaptive regulated neural network bias, N 1 For the length of the sequence, N mc For the sequence length, l is the number of product nodes, < ->Connecting weights for the self-adaptive adjusted neural network, H j Is an intermediate variable.
The invention has the following beneficial effects:
the invention has three levels of improvement. Firstly, the improvement on engineering design is that a generator set is added in a simple water delivery channel, and in order to ensure the safety of engineering, the traditional single-channel water delivery is changed into multi-channel parallel water delivery, the capacity and the adjustment space of a water delivery system are improved, the requirement on water quantity is smaller, the application range is wider than that of a simple power generation engineering, the power generation system can be suitable for medium and small rivers, and the water quantity and the water energy can be utilized in the utilization of water resources. The generated electric power is used for starting the control equipment of the gate, and meanwhile, the control equipment is matched with the generation of light energy and wind energy, so that the automatic operation can be realized all the year round. The control of the gate is only needed when water is used for generating electricity, the gate does not need to be controlled when water is not used, and the unit power generation is not influenced. Then, a control mode of multiple gates is adopted for the parallel channels, gates and monitoring equipment are arranged at key nodes of the channels, so that professionals can see the running state of the channels in real time, meanwhile, a control system is utilized to collect water level flow data of the channels, analyze and output the opening of the gates, and adjust the water quantity of the channels through adjusting the gates. And the arrangement of the multiple gates can effectively improve the water distribution flexibility and risk coping capability of channels, and the power generation system is combined with the control system to realize unattended and autonomous operation of engineering. Finally, the control device is improved, which is different from manual control and on-line control, and the control mode adopted by the invention can realize the processes of automatic control, feedback and readjustment of the gate, thereby greatly reducing the labor cost and improving the control efficiency of the gate.
Drawings
FIG. 1 is a schematic diagram of a water network multichannel parallel linkage control system according to an embodiment of the present invention;
FIG. 2 is a schematic diagram of a water network multichannel parallel linkage control system according to an embodiment of the present invention;
FIG. 3 is a schematic diagram of a gate linkage controller according to an embodiment of the present invention;
fig. 4 is a schematic diagram of a channel water level adaptive adjustment flow in an embodiment of the present invention.
Detailed Description
The following description of the embodiments of the present invention is provided to facilitate understanding of the present invention by those skilled in the art, but it should be understood that the present invention is not limited to the scope of the embodiments, and all the inventions which make use of the inventive concept are protected by the spirit and scope of the present invention as defined and defined in the appended claims to those skilled in the art.
The conventional diversion engineering often neglects the utilization of water energy aiming at the utilization of water quantity, and along with the construction of river and lake water system communication engineering and intelligent water network, the utilization efficiency of water resources needs to be improved. The water diversion mode of many newly-built channels mostly adopts a transmission control mode, and the control mode is more suitable for a simple water diversion channel, and the water diversion mode is a mode of utilizing water level difference to conduct water diversion, and water flow has water falling potential energy, so that labor cost is reduced to a certain extent, and the water falling potential energy is not well utilized. The water falling potential energy can be used for generating electricity in a mode of additionally installing the hydroelectric generating set, so that the upstream water can generate electricity and can also meet the downstream water supply requirement, the water resource utilization rate is improved, meanwhile, economic benefits can be brought, and the water quantity of the channel needs to be adjusted more when the electricity is generated and added into the water channel. The automatic control mode also fails to well utilize the water falling potential energy, and the control mode is more flexible than the transmission control mode, but the automatic control of water supply power generation is not involved in the current application environment, and the control system is also required to be improved for the complex engineering environment.
The mode of directly conducting water diversion from the reservoir can accurately control the downward drainage flow through controlling the water diversion gate, the downward drainage flow can be obtained through calculating the opening degree and the flow velocity of the gate, and the water diversion gate is convenient to adjust according to the downstream water demand. After parallel power generation by water supply, the change of the water demand of the generator set is possibly unstable, the regulation frequency of the water quantity of the channel is increased, the condition that the water level of the channel is suddenly changed is likely to occur only by regulating the reservoir gate, the sudden change of the water level can harm the canal lining, the lining structure is damaged, and engineering safety is threatened.
The improvement of the adjustment capability needs clear cognition on the adjustment mode of the water supply and power generation channels, and the previous gate adjustment is often linkage adjustment between the same projects, such as linkage between power stations or linkage between water supply channels. At present, the power generation needs to be considered in the channel for water supply, and in this case, the requirements of the two needs to be considered by how different projects are linked. After the power generation system is added, the whole water delivery system becomes more complex, the adjustment is more difficult, and the condition that the whole body is pulled and moved easily occurs. Therefore, it is very important to have a clear perception of the overall operating state of the water delivery system, and after that, the system is reasonably controlled.
The control of the channel water volume is basically realized by controlling the gate opening, and therefore, the gate opening is first determined when the water volume is to be adjusted. The opening degree of the gate influences the water discharge amount, the change of the downstream water amount can bring about the change of the water level, the change of the water level can be obtained through monitoring equipment, so that monitoring stations are required to be arranged at key positions of water delivery engineering, the water level is uploaded to a control system, and the control system controls the gate. The improved reservoir-channel water delivery system can improve the utilization efficiency of water resources, strengthen the water resource allocation capacity, solve the problem of mismatching of water resource allocation and productivity and population, and improve the water safety. Meanwhile, the improvement of the water delivery system meets the development requirement of river and lake water system communication engineering in China, and assistance is provided for realizing intelligent water conservancy modernization.
As shown in fig. 1 to 4, a water network multichannel parallel linkage control system provided by an embodiment of the present invention includes:
the multichannel parallel water delivery subsystem is used for delivering water in a multichannel parallel water delivery mode;
the parallel linkage simulation subsystem for water supply and power generation is used for acquiring water delivery state information of the multi-channel parallel water delivery subsystem, simulating channel water level by utilizing a hydrodynamic model and combining a water level error model prediction control method, and acquiring gate opening information required by controlling the multi-channel parallel water delivery subsystem to reach a preset water level;
The water supply and power generation parallel linkage control subsystem is used for acquiring gate opening information output by the water supply and power generation parallel linkage simulation subsystem, calculating a control rule value by using a PID control method of the self-adaptive Fourier series neural network, controlling the channel gate according to the control rule value, and feeding back water delivery state information of the multi-channel parallel water delivery subsystem to the water supply and power generation parallel linkage simulation subsystem.
The water supply-power generation parallel linkage control system provided by the embodiment has three levels of improvement. Firstly, the improvement on engineering design is that a generator set is added in a simple water delivery channel, and in order to ensure the safety of engineering, the traditional single-channel water delivery is changed into multi-channel parallel water delivery, the capacity and the adjustment space of a water delivery system are improved, the requirement on water quantity is smaller, the application range is wider than that of a simple power generation engineering, the power generation system can be suitable for medium and small rivers, and the water quantity and the water energy can be utilized in the utilization of water resources. The generated electric power is used for starting the control equipment of the gate, and meanwhile, the control equipment is matched with the generation of light energy and wind energy, so that the automatic operation can be realized all the year round. The control of the gate is only needed when water is used for generating electricity, the gate does not need to be controlled when water is not used, and the unit power generation is not influenced. Then, a control mode of multiple gates is adopted for the parallel channels, gates and monitoring equipment are arranged at key nodes of the channels, so that professionals can see the running state of the channels in real time, meanwhile, a control system is utilized to collect water level flow data of the channels, analyze and output the opening of the gates, and adjust the water quantity of the channels through adjusting the gates. And the arrangement of the multiple gates can effectively improve the water distribution flexibility and risk coping capability of channels, and the power generation system is combined with the control system to realize unattended and autonomous operation of engineering. Finally, the control device is improved, which is different from manual control and on-line control, and the control mode adopted by the invention can realize the processes of automatic control, feedback and readjustment of the gate, thereby greatly reducing the labor cost and improving the control efficiency of the gate.
In order to fully utilize the water head difference, a generator set is added in the water delivery process, so that the potential energy of falling water is reduced, and meanwhile, a part of economic benefits can be brought. However, there is still a certain risk that the power generation and water delivery tasks are carried out by only one channel. Because the generator set is influenced by the power grid, the power generation load fluctuates, and the corresponding power generation flow also fluctuates. In addition, when the generator set fails or even stops, the power generation flow is reduced to zero in a short time, the total main canal water supply flow needs to be kept relatively stable in a certain period, and the short-time and large-scale fluctuation of the flow causes abrupt change of the water level of the total main canal, thereby threatening the safe operation of the channel. If a sudden accident occurs in the generator set, after the emergency closing of the outlet gate of the generator set, the upstream water is not poured, and the risk is easily caused. In addition, after stopping the water supply, the downstream water demand cannot be satisfied, and thus the linkage effect is generated.
A main canal is additionally built on the basis of the original main canal, so that the problem that water flow does not leak when a single canal is blocked can be solved, and meanwhile, the downstream water supply requirement can be guaranteed. The scale of the channel is consistent with that of the original main channel, the standby regulation function is realized when the generator set works normally, water can be conveyed from the channel when the generator set breaks down, and the requirement of downstream water can be met while further damage to the generator set is avoided. Thus, the water delivery problem at the downstream and the drainage flow guiding problem at the unit fault can be solved. However, when the standby main canal is used for water delivery, sudden increase of water flow still causes harm to canal lining, and the problem of sudden change of the canal water level cannot be effectively solved. In view of the control requirement of channel water level change, the input flow needs to be smoothly controlled, and the operation of other water conservancy facilities is not influenced while the water level is ensured to be slowly changed. Therefore, a reliable and efficient water supply and power generation linkage control system is needed to be established, and balance between power generation and water supply flow is coordinated. Through the construction of the system, the opening degree of the gate of the main canal water inlet gate is timely adjusted, and the power generation of the power station and the water delivery safety of the channel are ensured. On the basis of the system, automatic elements are added, so that the linkage control of water supply and power generation is realized, the opening degree of the gate is automatically regulated, the reaction speed is improved, part of labor cost can be saved, the labor intensity of pipe transporting personnel is reduced, and a powerful technical means is provided for scientific scheduling and daily operation.
The canal system control has the characteristics of high nonlinearity and time-varying property, and the integrated monitoring control of the canal needs to be clear in controlling the hydrologic condition of the whole canal, and also needs to be provided with a high-efficiency stable linkage control system, so that the canal gate is regarded as an organic whole and has the multi-target regulation and control capability. The invention provides a hydrological model prediction system added with a water level error model prediction control method, which combines water level error control with a hydrodynamic model to accurately control, predict and adjust the channel water level. Aiming at a multi-input multi-output mode of gate linkage control, a fixed PID parameter cannot be adaptively adjusted according to the working condition of a canal system, and the parallel linkage control mode of water supply and power generation can realize multi-channel joint automatic real-time control of a water network, and can reasonably utilize water falling potential energy to generate power while solving the water diversion problem, so that the utilization efficiency of water resources is improved.
In an optional embodiment of the present invention, the multi-channel parallel water delivery subsystem constructed in this embodiment specifically includes:
the water inlet of the first main channel and the water inlet of the second main channel are respectively provided with a water diversion gate, and a generator set is arranged in the first main channel; the water outlets of the first main channel and the second main channel are communicated with the stilling pool, and the water outlet of the stilling pool is communicated with the third main channel.
The water delivery state information of the multichannel parallel water delivery subsystem acquired by the water supply and power generation parallel linkage simulation subsystem specifically comprises the following components:
gate opening information of the water diversion gate, water level information of the first main channel, the second main channel and the third main channel, and flow information of the generator set in the first main channel.
Specifically, the generator set is added on the basis of traditional single-channel water delivery, and the potential energy of falling water is effectively utilized. After the generator set is added, the water delivery capacity of the channel is affected by the generator set to a certain extent, and the condition that the generator set fails and water cannot be delivered is possible to happen. Therefore, one side of the original main channel is additionally paved with a main channel, the two main channels are matched with water delivery, and under normal conditions, the main channel of the generator set is mainly responsible for water supply, and the side main channel is used for auxiliary adjustment. In special cases, the side main canal is responsible for supplying water, and the construction standards of the two channels are consistent, so that the two channels have the capability of meeting the downstream water demand. After the generator set and the main channel are added, the number of gates is correspondingly increased, and the control mode of the gates is correspondingly adjusted. The method comprises the steps of arranging monitoring equipment at a reservoir position to monitor the opening degree of a gate, arranging monitoring equipment at a generator set to monitor the flow change of the generator set, arranging water level monitoring equipment along a downstream main channel, feeding water level information back to a control system in real time, synthesizing three data by a control algorithm, searching an optimal gate control strategy, sending a result to an executing mechanism, controlling the gate to adjust, and completing the water distribution requirement.
In an optional embodiment of the present invention, the parallel linkage simulation subsystem for water supply and power generation simulates the channel water level by using a hydrodynamic model in combination with a water level error model prediction control method, and the gate opening information required for controlling the multi-channel parallel water delivery subsystem to reach the preset water level specifically includes:
the water level information of each main channel and the gate opening information of each water diversion gate are read by utilizing a hydrodynamic model, the gate opening information of each water diversion gate is used as an input variable, the water level information of each main channel is used as an output variable, and the change of the channel water level is regulated by controlling the change of the gate opening;
meanwhile, the water level information of each main channel at the current moment is obtained by utilizing a water level error model prediction control method, Q, R parameter values are determined according to the obtained water level information, a time step to be predicted, a prediction interval and an expected value of an error are set, a gate opening degree information is used as a control variable to construct an optimization objective function of the water level error, an integrator delay model is used for establishing a differential error equation of the stilling pool, a corresponding control strategy is generated according to the range of the prediction interval, the water level error value of the prediction interval is obtained, and iterative optimization is carried out according to the comparison result of the water level error value of the prediction interval and the actual water level error value.
The optimization objective function for constructing the water level error by taking the gate opening information as a control variable comprises the following specific steps:
wherein U is * For the control sequence of the gate opening actual application, U is the control sequence of the predicted gate opening, x 0 For initial state of system evolution according to applied action sequence, J (U, x 0 ) As a cost function, k is a time step, N h For the predicted water level value, x (k) is Q, R is a constant weighting matrix for the quadratic deviation penalty, Q l Constant weighting matrix for linear penalty, T is transposed sign, u (k) is gate opening, x (N) h ) Is the final error between the control result and the target in the prediction interval.
The differential error equation for establishing the stilling pool by adopting the integrator delay model is specifically as follows:
wherein D is j E is the differential error of the stilling pool j For the water level error at the j position of the stilling pool, n is the total number of stilling pools, e i Is the water level error at the other stilling pool except the stilling pool j.
Specifically, after the power generation system is added into the water supply channel, the whole engineering becomes more complex, and the power generation is also considered while the water supply requirement is met, and the requirements of the two engineering on the water quantity are different, so that the two engineering are required to be controlled in a linkage way. The water quantity control mode is to control gates, and the number of the gates of the whole engineering is increased after the generator set is added. Moreover, because the engineering structure becomes more complex, the mode of manually adjusting the gates which is generally used is difficult to apply, and the gates are difficult to be linked through manual control, a control system is improved, the water level and flow change conditions of a main channel and a unit are collected by using a control algorithm, the opening of the gates is calculated and output by the control algorithm, and the water diversion gates are adjusted in a linked mode. The collection of water level information mainly sets up the monitoring station at fifty meters of gate low reaches and monitors, because the channel all is the artificial channel, and the water level condition is comparatively stable under the circumstances that no water volume flows in and out, can not set up to the monitoring station. And monitoring the opening degree of the gate, namely setting monitoring equipment at the position of the gate, and feeding back the opening degree of the gate to a control subsystem to ensure that the control requirement is met.
The parallel linkage control of water supply and power generation mainly comprises two aspects of contents, an analog subsystem and a control subsystem. The simulation subsystem is mainly used for simulating the hydrologic condition of the channel in a mathematical model and physical model mode, and mainly used for knowing the water level condition of each section in the channel. In addition, the method is also responsible for analyzing and predicting the water level when the channel water quantity needs to be adjusted, and finding out the flow which ensures that the water level changes within a safety threshold. The control subsystem mainly carries out coordinated control on gates related in the channels, rapidly and safely adjusts water level, ensures channel safety, and meets water supply requirements.
The simulation subsystem comprises acquisition of data required by model simulation, and the model simulation is two parts. Data acquisition includes two aspects: on the one hand, monitoring stations are arranged at proper positions of channels and units according to standard specifications, monitoring data are obtained, and the monitoring data mainly comprise water level flow data and gate opening data. On the other hand, the water information uploaded in the region downstream of the channel is received. And finding a water level fluctuation safety threshold corresponding to the channel according to the construction standard of the channel, and adding the water level fluctuation safety threshold serving as the safety standard of channel water delivery into the hydrodynamic model. The model simulation is to simulate the channel water flow state by utilizing a hydrodynamic model to obtain the change condition of the channel water level, observe the water level fluctuation state by changing the input water quantity and the output water quantity of the channel, find the input water quantity and the output water quantity which meet the premise of safe operation of the channel, further determine the opening degree of a gate and upload the result to a control subsystem.
In order to timely observe the change condition of the channel water level, the water level error model predictive control (DE-MPC) method is utilized to observe the change condition of the channel water level while the hydrodynamics model is utilized to simulate.
MPC is a control strategy based on a continuous re-planning of a sequence of control operations that must be implemented within a certain range. To this end, an MPC controller solves an optimization problem in each time step. In this problem, a mathematical model of the system is used to predict its behavior over a prediction horizon as a function of the input sequence applied. The behavior of large systems, such as irrigation channels with multiple ponds, can be represented with sufficient accuracy by the following linear time invariant state space model:
x(k+1)＝Ax(k)+B u u(k)+B d d(k)
where x (k) is the error value at time step k; u (k) is the gate opening value at time step k; d (k) is a vector of known measurable disturbances at time step k, A is a state transition matrix, B u For input to the state matrix, B d For interference with the state matrix, x (k+1) is the error value at the next time step k+1.
In order to optimize the behaviour of the system, it is necessary to define a cost function that scales its performance according to the control objective. It is assumed that there is a problem of balanced distribution of the water quantity in the regulating control channel, i.e. the controller has to divert the system state to a given reference value. For simplicity, it may be assumed that the source is a state reference without loss of generality. Thus, the control objective can be mathematically defined as the following function:
Wherein N is h To predict water level values, Q and R are constant weighting matrices, penalizing the quadratic deviation with respect to state and manipulated variable vector, Q l Also a constant weighting matrix that penalizes the state bias linearly.
The function depends on the current state x 0 It is the initial state of the system evolving according to the applied action sequence, consisting of u= (U (k), U (k+1), U (k+n) h -1)), and the expected value d= (D (k), D (k+1),.. h ) A) representation. Otherwise, the evolution of the system state will not be predicted.
The behavior of the controller varies according to the relationship between Q and R. If R is relatively greater than Q, the controller will focus on minimizing the use of manipulated variables at the cost of greater deviation in the state vector. Vice versa, i.e. if R is relatively lower than Q, the optimization will result in an important change of the control actions to reduce the deviation of the state vector.
The sequence of control actions applied in the system may be calculated to minimize the objective function. Thus, at each time step, the MPC controller solves the following optimization problem:
subject to the system model and to consideration constraints on state and manipulated variables. Using only the first action calculated, i.e. the actual application of the control action sequence U * ＝(u * (k)，u * (k+1)，...，u * (k+N h -1)) u * (0). The rest of the sequence provides information about the expected evolution of the control sequence, but it is not implemented. In the next time step, the MPC controller again solves the same optimization problem according to the latest information and applies the corresponding control actions. This process is repeated in every time step in a so-called back-off level.
The control of the channel flow may also be equivalent to the control of the water level, with the aim of adjusting the water level difference error and keeping the water level of the basin at a given reference level. As previously mentioned, MPC requires a model of the control system to predict its behavior over a prediction horizon. For control purposes, an Integrator Delay (ID) model is used for the canal pool. The ID model divides the canal pool into a uniform fluid with an attribute delay time and a return water segment with an attribute storage area. The water level h at the downstream end of the basin is the control outflow (q out (k) Taking into account k) d Inflow of backwater segment of water flow delay time step (q in (k-k d ) (q) and discharge outflow (q) off-take (k) Is provided) is provided. The disturbance flow is derived from the water intake plan of the consumer. The discrete time invariant canal pool model applied in the present invention is defined as:
where h (k+1) is the water level at time step k+1, h (k) is the water level at time step k, A s For average storage area T c To control the time step.
From a control point of view, adjustment errors are often of concern instead of water level. This enables us to penalize the deviation from zero. For this reason, it is necessary to introduce a change in the variable and rewrite the schema as:
in addition to the error related to the target water level, in the invention, the processing mode of the differential error of the water level is further improved, and the differential error of the water level is obtained as follows:
D j ＝e j -e j+1 ＝(y j -SP j )-(y j+1 -SP j+1 )
wherein D is j For differential error at the absorption cell j, e j For the water level error at the j position of the stilling pool, e j+1 For the water level error of the stilling pool j+1, SP j For a predefined target water level, y, at the absorption basin j j+1 To the downstream water level of the stilling pool j+1, SP j+1 For a predefined target water level, y, at the absorption basin j+1 j For the downstream water level at the sink j, SP is a predefined target water level and e is a water level error. When an error occurs in one pool, adjacent upstream and downstream pools are first disturbed, so that all pools gradually participate in the control process.
In the present invention, to speed up the process of sharing errors between all pools, the differential error variable is determined by:
wherein e i For the water level error at the other stilling ponds except the stilling pond j, n is the total number of stilling ponds.
Utilizing the above equation enables the controller to react faster in terms of sharing errors because all of the canal pools are participating in the sharing process at the same time.
When the DE-MPC method is used for carrying out predictive regulation on the channel water level, firstly, water level data of the channel at the current moment is measured or water level data of the channel at the control starting moment is estimated to obtain the current or a certain moment system state. And selecting proper Q, R parameter values according to faster control actions or smoother control processes as required after the water level information is obtained. And selecting the time step, the prediction interval and the expected value of the error to be predicted, and inputting the gate opening as a control variable. After control starts, the system will make corresponding control plans, i.e., u (k), u (k+1), according to the range of the prediction interval h -1) and obtaining the control result of the prediction interval, i.e. the water level error value d (k), d (k+1), -d (k+n h ) And selecting u (k) for control, and repeating the steps until the optimal value is reached by taking the moment k+1 as a control starting point according to the actual water level error value after the control is finished.
The process is completed in a hydrodynamic model, the hydrodynamic model firstly reads the actual water level and the gate opening of each gate, the gate opening of each section of the channel is used as an input variable, the water level is used as an output variable, the change of the channel water level is influenced by controlling the change of the gate opening, and the change condition of the channel water level is simulated. The combined DE-MPC control method is that the gate of the channel is controlled according to the control strategy formulated by the DE-MPC control method, the opening of the gate is changed, the hydrodynamic model is used for simulation, and the water level value obtained in the simulation result is input into the DE-MPC for subsequent control. After the whole predictive control process is completed, the gate opening is uploaded to a control module, and the control system controls the actual gate.
The hydrodynamic model does not have to be defined forcefully, and can be selected according to the actual conditions of the channel, such as SMS, WMS and the like, so long as the water level flow simulation can be realized.
The method can quickly obtain the control opening of the channel gate, and can effectively judge whether the channel water level change has a problem or not according to the size of the error value by taking the water level error as an expected value, thereby influencing the channel safety and the normal water supply of upstream and downstream users. On the one hand, the judgment result is to adjust the auxiliary control system to maintain the stable water level, and on the other hand, the judgment result can be used for professional reference, whether to make other water level adjustment plans or to perform water level early warning on a user, and the like.
In an optional embodiment of the present invention, the water supply power generation parallel linkage control subsystem calculates a control rule value by using an adaptive fourier series neural network PID control method specifically includes:
constructing a gate linkage controller by using a PID controller, a multi-input multi-output Fourier series neural network, a multi-input single-output Fourier series neural network and a system controller;
the PID controller establishes an optimal objective function according to the difference value between the input current gate opening value and the target gate opening value, and combines the gain parameters output by the multi-input multi-output Fourier series neural network to output the gate opening value subjected to PID regulation to the system controller;
The system controller controls the water diversion gate according to the gate opening value regulated by the PID, and outputs the current gate opening value of the system;
the multi-input multi-output Fourier series neural network calculates an approximate value of the jacobian matrix system according to the neural network connection weight, then calculates the self-adaptive neural network bias and the neural network connection weight according to the neural network self-adaptive equation, and finally calculates gain parameters of the PID controller according to the self-adaptive neural network bias and the neural network connection weight, and outputs the gain parameters to the PID controller.
The optimal objective function established by the PID controller is specifically:
e(k)＝R(k)-y(k)
wherein E (k) is an optimal objective function, E (k) is a difference between a current gate opening value and a target gate opening value, R (k) is the current gate opening value, and y (k) is the target gate opening value.
The neural network adaptive equation is specifically:
wherein,bias for adaptive tuned neural network, +.>For the original neural network bias, eta is the learning rate, e (k) is the difference between the current gate opening value and the target gate opening value, y (k) is the target gate opening value, u (k) is the gate opening value after PID adjustment, O h (k) For the gain parameter of the PID controller, +. >Connecting weights for the self-adaptive neural network, < ->Connecting weights for the original neural network, H j Is an intermediate variable.
The calculation mode for calculating the gain parameter of the PID controller according to the self-adaptive regulated network bias and the network connection weight is as follows:
wherein O is h Is a gain parameter of the PID controller,is self-adaptive after adjustmentN, N 1 For the length of the sequence, N mc For the sequence length, l is the number of product nodes, < ->Connecting weights for the self-adaptive adjusted neural network, H j Is an intermediate variable.
Specifically, the control subsystem receives the gate opening uploaded by the simulation subsystem, a preset value of the gate opening is obtained, the controller carries out linkage control on the related gate, and the monitoring equipment uploads the gate opening to the control subsystem in real time to realize real-time feedback.
The feedback system is to arrange monitoring equipment at each position of the channel to acquire data such as gate opening, channel water depth, gate passing flow and the like. The feedback system mainly has two data feedback directions, one is that when the control subsystem controls the gate, the feedback system monitors the change condition of the opening of the gate in real time, the data is uploaded to the control subsystem, and the controller timely adjusts the control strategy according to the data. On the other hand, after the control subsystem finishes the control process, the actual gate opening and water level condition are uploaded to the simulation subsystem, and the hydrodynamic model is utilized for simulation, so that the control result is intuitively displayed.
If the control result reaches the expected requirement, the control is ended; if the control result shows that the channel water level does not meet the expected requirement, the DE-MPC method is continuously utilized to obtain the opening of the gate, the subsequent steps are repeated, and the channel gate is continuously controlled until the expected requirement is met.
In order to realize the linkage control of the gate, the embodiment uses an Adaptive Fourier Series Neural Network PID (AFSNNPID) control method by combining the characteristics of multiple targets, nonlinearity and time variability of channel control, and the method can realize the functions of parameter adjustment and control.
Discrete form of PID controller:
where u (K) is the gate opening value, K is the time step, e (K) is the gate opening error, K p ，K i ，K d Is the gain parameter of PID, affects the control efficiency of the gate, T s Is the sampling period.
The present embodiment uses two Fourier Series Neural Networks (FSNNs) to implement a gate linkage controller, the FSNN on the right being the simulator FSNN, which is a Multiple Input Single Output (MISO) FSNN, allowing the dynamic behavior of the system to be simulated.
Input vector X of simulator FSNN e ＝[x 1 ，x 2 ，x 3 ，...，x m ]The definition is as follows:
X e ＝[u(k)，u(k-1)，...，u(k-b e )，y(k-1)，y(k-2)，...，y(k-a e )]
wherein m is e ＝1+b e +a e Is the input number, b, of the simulator FSNN e A is the input quantity of the gate opening value e Is the input number of the target gate opening value.
EmulatorThe output of (2) is given by:
the connection weight is adjusted by the following formula:
W o (k)＝W 0 (k-1)+ηe h (k)
wherein n is 1 For the length of the sequence, n m For the length of the sequence, e h (k) The error at time step k is output for h.
The FSNN on the left is a Multiple Input Multiple Output (MIMO) FSNN with three outputs (o 1 ，o 2 And o 3 ). It gives the controller gain, K p 、K i 、K d For three parameters of PID controller, let o 1 ＝K p ，o 2 ＝K i ，o 3 ＝K d The input vector of the network is:
X c ＝[e(k)，e(k-1)，...，e(k-b c )，u(k-1)，u(k-2)，...，u(k-a c )]
wherein m is c ＝1+b c +a c Is the number of input vectors, b c To input the number of errors, a c Is the number of input openings.
The output of FSNN is:
H 1 ＝cos(n 1 ω 1 x 1 )cos(n 2 ω 2 x 2 )...cos(n mc-1 ω mc-1 x mc-1 )cos(n mc ω mc x mc )
H 2 ＝cos(n 1 ω 1 x 1 )cos(n 2 ω 2 x 2 )...cos(n mc-1 ω mc - 1 x mc-1 )sin(n mc ω mc x mc )
…
H l-1 ＝sin(n 1 ω 1 x 1 )sin(n 2 ω 2 x 2 )...sin(n mc-1 ω mc-1 x mc-1 )cos(n mc ω mc x mc )
H l ＝sin(n 1 ω 1 x 1 )sin(n 2 ω 2 x 2 )...sin(n mc-1 ω mc-1 x mc-1 )sin(n mc ω mc x mc )
wherein,and->The bias of the connection weight and the h MISO FSNN, respectively.Is the frequency weight, T i Is input x i Range (x) i ∈[0T i ])，l＝2 m Is the number of product nodes, < > and->Is the connection weight (state weight) between the hidden layer and the output layer, W 0 Is the network bias, N i Is the sequence length. Weight number->By->Given.
The FSNN connection weights, given the PID controller gains, are adjusted to minimize the following objective function:
e(k)＝R(k)-y(k)
where y (k) is a gate opening degree which is a system output, and R (k) is a target gate opening degree which is a reference value.
The adaptation rules are derived using the delta rules as follows:
the calculation method comprises the following steps:
wherein,a jacobian matrix representing time k is estimated using the FSNN model.
To obtain fast convergence and good control performance of the control algorithm, the FSNN model must have sufficient accuracy, and a large estimation error may cause convergence or divergence of the control algorithm. The resulting jacobian matrix system is as follows:
finally, the adaptive equation is as follows:
when the control subsystem operates, initial gate opening and control parameters are firstly obtained, then the system outputs a difference value between the gate opening and the expected opening, firstly an approximation value of the jacobian matrix system is calculated, and then a new value is calculatedValue, recalculate new ++>Finally calculating the control rule value to enter the channel gateAnd (5) row control.
The AFSNNPID controller is utilized to carry out linkage control on the gate in the channel, so that the problem that the traditional PID controller cannot process the multi-target nonlinearity of the channel can be solved, the automatic adjustment of parameters is realized, the adjustment speed of the gate is accelerated, and the channel and the gate are regarded as an organic whole. On the other hand, the intelligent control of the gate can be realized by the control system, so that the labor cost is reduced, the safety coefficient of the gate adjusting work is improved, more choices are provided for the adjusting mode of the channel gate, and the risk coping capacity of the channel is improved.
The simulation module and the control module in the water supply-power generation parallel linkage control method system and the device are matched with each other, and the water level of the channel is efficiently and automatically controlled through the steps of simulation, control, feedback and adjustment, so that the risk coping capacity of the channel and the flexibility degree of regulation and control are improved.
The water supply-power generation parallel linkage control system provided by the invention can comprehensively sense and efficiently regulate the water quantity states of various channels such as a single channel, multiple channels, a water supply-power generation channel and the like, improves the risk coping capacity of the channels and the stability of water supply, reduces the labor cost and improves the economic benefit.
The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
The principles and embodiments of the present invention have been described in detail with reference to specific examples, which are provided to facilitate understanding of the method and core ideas of the present invention; meanwhile, as those skilled in the art will have variations in the specific embodiments and application scope in accordance with the ideas of the present invention, the present description should not be construed as limiting the present invention in view of the above.
Those of ordinary skill in the art will recognize that the embodiments described herein are for the purpose of aiding the reader in understanding the principles of the present invention and should be understood that the scope of the invention is not limited to such specific statements and embodiments. Those of ordinary skill in the art can make various other specific modifications and combinations from the teachings of the present disclosure without departing from the spirit thereof, and such modifications and combinations remain within the scope of the present disclosure.
Claims (10)
1. The utility model provides a multichannel parallel linkage control system of water network which characterized in that includes:
the multichannel parallel water delivery subsystem is used for delivering water in a multichannel parallel water delivery mode;
the parallel linkage simulation subsystem for water supply and power generation is used for acquiring water delivery state information of the multi-channel parallel water delivery subsystem, simulating channel water level by utilizing a hydrodynamic model and combining a water level error model prediction control method, and acquiring gate opening information required by controlling the multi-channel parallel water delivery subsystem to reach a preset water level;
the water supply and power generation parallel linkage control subsystem is used for acquiring gate opening information output by the water supply and power generation parallel linkage simulation subsystem, calculating a control rule value by using a PID control method of the self-adaptive Fourier series neural network, controlling the channel gate according to the control rule value, and feeding back water delivery state information of the multi-channel parallel water delivery subsystem to the water supply and power generation parallel linkage simulation subsystem.
2. The water network multichannel parallel linkage control system according to claim 1, wherein the multichannel parallel water delivery subsystem specifically comprises:
the water inlet of the first main channel and the water inlet of the second main channel are respectively provided with a water diversion gate, and a generator set is arranged in the first main channel; the water outlets of the first main channel and the second main channel are communicated with the stilling pool, and the water outlet of the stilling pool is communicated with the third main channel.
3. The water network multichannel parallel linkage control system according to claim 2, wherein the water delivery state information of the multichannel parallel water delivery subsystem obtained by the water supply power generation parallel linkage simulation subsystem specifically comprises:
gate opening information of the water diversion gate, water level information of the first main channel, the second main channel and the third main channel, and flow information of the generator set in the first main channel.
4. The water network multichannel parallel linkage control system according to claim 1 or 3, wherein the water supply power generation parallel linkage simulation subsystem simulates channel water level by using a hydrodynamic model in combination with a water level error model prediction control method, and the gate opening information required for controlling the multichannel parallel water delivery subsystem to reach a preset water level is specifically obtained by:
The water level information of each main channel and the gate opening information of each water diversion gate are read by utilizing a hydrodynamic model, the gate opening information of each water diversion gate is used as an input variable, the water level information of each main channel is used as an output variable, and the change of the channel water level is regulated by controlling the change of the gate opening;
meanwhile, the water level information of each main channel at the current moment is obtained by utilizing a water level error model prediction control method, Q, R parameter values are determined according to the obtained water level information, a time step to be predicted, a prediction interval and an expected value of an error are set, a gate opening degree information is used as a control variable to construct an optimization objective function of the water level error, an integrator delay model is used for establishing a differential error equation of the stilling pool, a corresponding control strategy is generated according to the range of the prediction interval, the water level error value of the prediction interval is obtained, and iterative optimization is carried out according to the comparison result of the water level error value of the prediction interval and the actual water level error value.
5. The water network multichannel parallel linkage control system according to claim 4, wherein the optimization objective function for constructing the water level error by using the gate opening information as the control variable is specifically:
Wherein U is * For the control sequence of the gate opening actual application, U is the control sequence of the predicted gate opening, x 0 For initial state of system evolution according to applied action sequence, J (U, x 0 ) As a cost function, k is a time step, N h For the predicted water level value, x (k) is the error value at time step k, Q, R is a constant weight matrix of the quadratic deviation penalty, Q l Constant weighting matrix for linear penalty, T is transposed sign, u (k) is gate opening, x (N) h ) For prediction intervalAnd (3) final error between the internal control result and the target.
6. The water network multichannel parallel linkage control system of claim 4, wherein the differential error equation for establishing the stilling pool by adopting the integrator delay model is specifically as follows:
wherein D is j For differential error at the absorption cell j, e j For the water level error at the j position of the stilling pool, n is the total number of stilling pools, e i Is the water level error at the other stilling pool except the stilling pool j.
7. The water network multichannel parallel linkage control system according to claim 1 or 3, wherein the water supply power generation parallel linkage control subsystem calculates a control law value by using a self-adaptive fourier series neural network PID control method specifically comprises:
Constructing a gate linkage controller by using a PID controller, a multi-input multi-output Fourier series neural network, a multi-input single-output Fourier series neural network and a system controller;
the PID controller establishes an optimal objective function according to the difference value between the input current gate opening value and the target gate opening value, and combines the gain parameters output by the multi-input multi-output Fourier series neural network to output the gate opening value subjected to PID regulation to the system controller;
the system controller controls the water diversion gate according to the gate opening value regulated by the PID, and outputs the current gate opening value of the system;
the multi-input multi-output Fourier series neural network calculates an approximate value of the jacobian matrix system according to the neural network connection weight, then calculates the self-adaptive neural network bias and the neural network connection weight according to the neural network self-adaptive equation, and finally calculates gain parameters of the PID controller according to the self-adaptive neural network bias and the neural network connection weight, and outputs the gain parameters to the PID controller.
8. The water network multichannel parallel linkage control system of claim 7, wherein the optimal objective function established by the PID controller is specifically:
e(k)＝R(k)-y(k)
Wherein E (k) is an optimal objective function, E (k) is a difference between a current gate opening value and a target gate opening value, R (k) is the current gate opening value, and y (k) is the target gate opening value.
9. The water network multichannel parallel linkage control system of claim 7, wherein the neural network adaptive equation is specifically:
wherein,bias for adaptive tuned neural network, +.>For the original neural network bias, eta is the learning rate, e (k) is the difference between the current gate opening value and the target gate opening value, y (k) is the target gate opening value, u (k) is the gate opening value after PID adjustment, O h (k) For the gain parameter of the PID controller, +.>Connecting weights for the self-adaptive neural network, < -> Connecting weights for the original neural network, H j Is an intermediate variable.
10. The water network multichannel parallel linkage control system according to claim 7, wherein the calculation mode for calculating the gain parameter of the PID controller according to the adaptively adjusted network bias and the network connection weight is as follows:
wherein O is h Is a gain parameter of the PID controller,for self-adaptive regulated neural network bias, N 1 N is the number of channels of the Lay-1 layer of the neural network mc The number of channels of the layer of the neural network Lay-mc is the number of nodes of the product, and l is the number of nodes of the product>Connecting weights for the self-adaptive adjusted neural network, H j Is an intermediate variable.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311393936.0A CN117348400A (en) | 2023-10-25 | 2023-10-25 | Water network multichannel parallel linkage control system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311393936.0A CN117348400A (en) | 2023-10-25 | 2023-10-25 | Water network multichannel parallel linkage control system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117348400A true CN117348400A (en) | 2024-01-05 |
Family
ID=89359247
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311393936.0A Pending CN117348400A (en) | 2023-10-25 | 2023-10-25 | Water network multichannel parallel linkage control system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117348400A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117824788A (en) * | 2024-03-05 | 2024-04-05 | 河海大学 | Water level monitoring and analyzing system |
| CN118131825A (en) * | 2024-05-06 | 2024-06-04 | 河北工程大学 | Water level control method and device for throttle valve |
-
2023
- 2023-10-25 CN CN202311393936.0A patent/CN117348400A/en active Pending
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117824788A (en) * | 2024-03-05 | 2024-04-05 | 河海大学 | Water level monitoring and analyzing system |
| CN117824788B (en) * | 2024-03-05 | 2024-05-28 | 河海大学 | Water level monitoring and analyzing system |
| CN118131825A (en) * | 2024-05-06 | 2024-06-04 | 河北工程大学 | Water level control method and device for throttle valve |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP7346705B2 (en) | Water level prediction and control method for series water channel based on fuzzy neural network | |
| CN117348400A (en) | Water network multichannel parallel linkage control system | |
| CN107506909B (en) | Cascade reservoir hydropower station group scheduling control system and method for fish habitat protection | |
| WO2024109103A1 (en) | Joint control method for gates, apparatus, and parallel water supply and power generation system | |
| CN103591637B (en) | A kind of central heating secondary network runing adjustment method | |
| CN111967666B (en) | Comprehensive cooperative scheduling system and scheduling method for river and lake water system | |
| CN109272245A (en) | The real-time auto-allocation method of step power station floodgate aperture and system | |
| CN109447405A (en) | A kind of library multi-stag step library group's short-term plan formulating method undertaking peak regulation task | |
| CN104268653B (en) | Cascade reservoir optimal scheduling method based on ESP | |
| CN106505635A (en) | Abandon the minimum active power dispatch model of wind and scheduling system | |
| CN103886388A (en) | Multi-cycle generation scheduling coordinated optimization and closed-loop control method | |
| CN106026084B (en) | A kind of AGC power dynamic allocation methods based on virtual power generation clan | |
| Liao et al. | Daily peak shaving operation of cascade hydropower stations with sensitive hydraulic connections considering water delay time | |
| Puig et al. | Model predictive control of combined irrigation and water supply systems: Application to the Guadiana river | |
| CN111612292A (en) | Cascade hydropower station dispatching control system and method based on key water level control | |
| Hoshimov et al. | Forecasting the electric consumption of objects using artificial neural networks | |
| CN115313403A (en) | Real-time voltage regulation and control method based on deep reinforcement learning algorithm | |
| CN116256974A (en) | Intelligent control method, system and device for complex water supply network diversion water outlet joint automation | |
| CN109325610B (en) | Piecewise linear scheduling method for multi-water-source cascade pump station water transfer system | |
| Ma et al. | Multi-objective solution and decision-making framework for coordinating the short-term hydropeaking-navigation-production conflict of cascade hydropower reservoirs | |
| CN104750127A (en) | Control method for four-holding water tank system | |
| Yuan et al. | Self-optimization system dynamics simulation of real-time short term cascade hydropower system considering uncertainties | |
| CN115986850B (en) | Transmission and distribution collaborative optimization scheduling method considering multisource balance dynamic full response | |
| Zhang et al. | Self-optimization simulation model of short-term cascaded hydroelectric system dispatching based on the daily load curve | |
| Moosavian et al. | Sequential quadratic programming and analytic hierarchy process for nonlinear multiobjective optimization of a hydropower network |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |