Intelligent coordinated control method for distributed flywheel energy storage device

Abstract

The present invention discloses an intelligent collaborative control method for a distributed flywheel energy storage device, which relates to the field of energy storage device control, wherein the method comprises: obtaining N flywheel energy storage units of a first flywheel energy storage device; obtaining N operating sample data sets; when the first flywheel energy storage device receives a first energy adjustment instruction, the first energy adjustment instruction includes a preset energy adjustment capacity; taking the preset energy adjustment capacity as a control target, and taking the N flywheel energy storage units as adaptive variables, inputting them into a collaborative control model; outputting N control parameters; inputting the N control parameters into a frequency converter to collaboratively control the motor rotation of the N flywheel energy storage units. The method solves the technical problem in the prior art that the energy storage adjustment control accuracy of the flywheel energy storage device is low, resulting in poor energy storage adjustment control effect of the flywheel energy storage device. The technical effect of improving the energy storage adjustment control effect of the flywheel energy storage device is achieved.

Description

Technical Field

[0001] The present invention relates to the field of energy storage device control, and in particular to an intelligent collaborative control method for a distributed flywheel energy storage device. Background Art

[0002] Flywheel energy storage device is an energy storage device that uses an electric motor to drive a flywheel to rotate at high speed, converts electrical energy into kinetic energy and stores it, and then uses the high-speed rotating flywheel to drive a generator to release electrical energy when needed. Flywheel energy storage device has the advantages of high energy storage density, high energy conversion efficiency, high reliability, and easy maintenance.

[0003] In the prior art, there is a technical problem that the energy storage regulation and control accuracy of the flywheel energy storage device is low, resulting in poor energy storage regulation and control effect of the flywheel energy storage device. Summary of the invention

[0004] The present application provides an intelligent collaborative control method for a distributed flywheel energy storage device. The method solves the technical problem in the prior art that the energy storage regulation and control accuracy of the flywheel energy storage device is low, resulting in poor energy storage regulation and control effect of the flywheel energy storage device. The method achieves the technical effect of improving the energy storage regulation and control accuracy of the flywheel energy storage device, improving the energy storage regulation and control effect of the flywheel energy storage device, and improving the reliability, timeliness, and adaptability of the power supply of the flywheel energy storage device.

[0005] In view of the above problems, the present application provides an intelligent collaborative control method for a distributed flywheel energy storage device.

[0006] In a first aspect, the present application provides an intelligent collaborative control method for a distributed flywheel energy storage device, wherein the method is applied to the intelligent collaborative control system for a distributed flywheel energy storage device of the present application, and the method includes: obtaining N flywheel energy storage units of a first flywheel energy storage device; collecting operating data of the N flywheel energy storage units to obtain N operating sample data sets; when the first flywheel energy storage device receives a first energy adjustment instruction, wherein the first energy adjustment instruction includes a preset energy adjustment capacity; taking the preset energy adjustment capacity as a control target, and taking the N flywheel energy storage units as adaptation variables, inputting them into a collaborative control model, wherein the collaborative control model is trained to convergence through the N operating sample data sets to control the loss fitness of the collaborative control between each flywheel energy storage unit; according to the collaborative control model, outputting N control parameters, wherein the N control parameters include N motor speeds, and the N motor speeds correspond to the N flywheel energy storage units; inputting the N control parameters into the inverter to collaboratively control the motor rotation of the N flywheel energy storage units.

[0007] In a second aspect, the present application also provides an intelligent collaborative control system for a distributed flywheel energy storage device, wherein the system includes: an energy storage unit acquisition module, the energy storage unit acquisition module is used to acquire N flywheel energy storage units of a first flywheel energy storage device; an operation data acquisition module, the operation data acquisition module is used to collect operation data of the N flywheel energy storage units to obtain N operation sample data sets; an energy regulation instruction receiving module, the energy regulation instruction receiving module is used when the first flywheel energy storage device receives a first energy regulation instruction, wherein the first energy regulation instruction includes a preset energy regulation capacity; an input module, the input module is used to use the preset energy regulation capacity as a control The cooperative control target is set, and the N flywheel energy storage units are used as adaptation variables and input into the cooperative control model, wherein the cooperative control model is trained by the N running sample data sets until convergence to obtain, and controls the loss fitness of the cooperative control between each flywheel energy storage unit; a control parameter output module, wherein the control parameter output module is used to output N control parameters according to the cooperative control model, wherein the N control parameters include N motor speeds, and the N motor speeds correspond to the N flywheel energy storage units; a cooperative control module, wherein the cooperative control module is used to input the N control parameters into the inverter to coordinate the motor rotation of the N flywheel energy storage units.

[0008] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0009] By collecting the operation data of the N flywheel energy storage units of the first flywheel energy storage device, N operation sample data sets are obtained; when the first flywheel energy storage device receives the first energy regulation instruction, the preset energy regulation capacity is used as the control target, and the N flywheel energy storage units are used as the adaptive variables, which are input into the collaborative control model, and N control parameters are output according to the collaborative control model; the N control parameters are input into the frequency converter to perform collaborative control on the motor rotation of the N flywheel energy storage units. The technical effect of improving the energy storage regulation control accuracy of the flywheel energy storage device, improving the energy storage regulation control effect of the flywheel energy storage device, and improving the reliability, timeliness, and adaptability of the power supply of the flywheel energy storage device is achieved.

[0010] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings of the embodiments of the present invention are briefly introduced below. Obviously, the drawings in the following description only relate to some embodiments of the present invention, but are not intended to limit the present invention.

[0012] Figure 1 A schematic diagram of a flow chart of an intelligent collaborative control method for a distributed flywheel energy storage device in this application;

[0013] Figure 2 A schematic diagram of a flow chart for outputting M control parameters in the intelligent collaborative control method for a distributed flywheel energy storage device in this application;

[0014] Figure 3 This is a schematic diagram of the structure of the intelligent collaborative control system used in the distributed flywheel energy storage device of this application.

[0015] Explanation of the reference numerals: energy storage unit acquisition module 11 , operation data acquisition module 12 , energy adjustment instruction receiving module 13 , input module 14 , control parameter output module 15 , collaborative control module 16 . DETAILED DESCRIPTION

[0016] The present application provides an intelligent collaborative control method for a distributed flywheel energy storage device. It solves the technical problem in the prior art that the energy storage regulation and control accuracy of the flywheel energy storage device is low, resulting in poor energy storage regulation and control effect of the flywheel energy storage device. It achieves the technical effect of improving the energy storage regulation and control accuracy of the flywheel energy storage device, improving the energy storage regulation and control effect of the flywheel energy storage device, and improving the reliability, timeliness, and adaptability of the power supply of the flywheel energy storage device.

[0017] Embodiment 1:

[0018] Please refer to the attached Figure 1 The present application provides an intelligent collaborative control method for a distributed flywheel energy storage device, wherein the method is applied to the intelligent collaborative control system for a distributed flywheel energy storage device of the present application, and the method specifically comprises the following steps:

[0019] Step S100: Acquire N flywheel energy storage units of a first flywheel energy storage device;

[0020] Step S200: collecting operation data of the N flywheel energy storage units to obtain N operation sample data sets;

[0021] Specifically, real-time operation data of N flywheel energy storage units of the first flywheel energy storage device are collected to obtain N operation sample data sets corresponding to the N flywheel energy storage units. Among them, the first flywheel energy storage device can be any flywheel energy storage device that uses the intelligent collaborative control method for distributed flywheel energy storage devices of the present application for intelligent energy storage regulation and control. The first flywheel energy storage device includes N flywheel energy storage units. And N is a positive integer greater than 1. Each flywheel energy storage unit includes a flywheel, a flywheel rotor, a frequency converter, a motor, a generator and other equipment. Each operation sample data set includes multiple flywheel operation parameters such as real-time flywheel speed, real-time motor speed, real-time flywheel storage capacity, real-time flywheel charging rate, real-time flywheel discharge rate, real-time flywheel power, and real-time flywheel resistance corresponding to each flywheel energy storage unit. The real-time flywheel storage capacity includes the real-time energy storage capacity of the flywheel energy storage unit. And each flywheel operation parameter has a corresponding time mark. The technical effect of collecting N operating sample data sets corresponding to N flywheel energy storage units is achieved, laying the foundation for subsequent energy storage regulation and control of the first flywheel energy storage device.

[0022] Step S300: When the first flywheel energy storage device receives a first energy regulation instruction, wherein the first energy regulation instruction includes a preset energy regulation capacity;

[0023] Step S400: taking the preset energy regulation capacity as a control target and the N flywheel energy storage units as adaptation variables, and inputting them into a collaborative control model, wherein the collaborative control model is obtained by training the N running sample data sets until convergence, and controls the loss fitness of the collaborative control between the flywheel energy storage units;

[0024] Step S500: outputting N control parameters according to the collaborative control model, wherein the N control parameters include N motor speeds, and the N motor speeds correspond to the N flywheel energy storage units;

[0025] Furthermore, step S500 of the present application also includes:

[0026] Step S510: performing identification according to the N running sample data sets to obtain the control sensitivity of the flywheel rotor and the control delay degree of the resistance experienced by the flywheel;

[0027] Specifically, when the first flywheel energy storage device receives the first energy regulation instruction, data extraction is performed based on N operating sample data sets to obtain N flywheel control data sets. Among them, the first energy regulation instruction is instruction information for performing energy storage regulation and control on the first flywheel energy storage device. The first energy regulation instruction includes a preset energy regulation capacity. And the preset energy regulation capacity has a frequency modulation mode identifier. The frequency modulation mode identifier is energy storage charging mode / energy storage discharge mode. When the frequency modulation mode identifier is energy storage charging mode, the corresponding preset energy regulation capacity is the preset charging capacity. When the frequency modulation mode identifier is energy storage discharge mode, the corresponding preset energy regulation capacity is the preset discharge capacity. Each flywheel control data includes multiple time identifiers in each operating sample data set, and the real-time flywheel speed, real-time motor speed, real-time flywheel power, and real-time flywheel resistance corresponding to each time identifier.

[0028] Further, N flywheel control data sets are taken as input information and input into a flywheel control analysis model. The flywheel control analysis model is used to identify the rotor control sensitivity and analyze the resistance control delay of the N flywheel control data sets, and obtain N control sensitivities and N control delays corresponding to the N flywheel energy storage units. Among them, the control sensitivity is parameter information used to characterize the control sensitivity of the flywheel rotor of the flywheel energy storage unit. The higher the control sensitivity, the stronger the control sensitivity of the flywheel rotor of the corresponding flywheel energy storage unit. The control delay is data information used to characterize the control delay of the resistance suffered by the flywheel. The larger the control delay, the stronger the control delay of the resistance suffered by the flywheel.

[0029] Exemplarily, the construction process of the flywheel control analysis model is: historical data query is performed according to N flywheel control data sets to obtain multiple flywheel control analysis sets. Each flywheel control analysis set includes N historical flywheel control data sets, and N historical control sensitivities and N historical control delays corresponding to the N historical flywheel control data sets. Then, 80% of the random data information in the multiple flywheel control analysis sets are divided into training data sets. 20% of the random data information in the multiple flywheel control analysis sets are divided into test data sets. Based on the BP neural network, the training data set is continuously self-trained and learned to a convergence state to obtain a flywheel control analysis model. Then, the test data set is used as input information and input into the flywheel control analysis model, and the parameters of the flywheel control analysis model are updated through the test data set. Among them, the BP neural network is a multi-layer feedforward neural network trained according to the error back propagation algorithm. The BP neural network includes an input layer, a multi-layer neuron, and an output layer. The BP neural network can perform forward calculations and reverse calculations. During forward calculation, the input information is processed layer by layer through multiple layers of neurons from the input layer to the output layer. The state of each layer of neurons only affects the state of the next layer of neurons. If the expected output cannot be obtained in the output layer, reverse calculation is performed to return the error signal along the original connection path, and the error signal is minimized by modifying the weights of each neuron. The flywheel control analysis model includes an input layer, a hidden layer, and an output layer. The flywheel control analysis model has the function of identifying the rotor control sensitivity and analyzing the resistance control delay of the input N flywheel control data sets.

[0030] The technical effect of obtaining N control sensitivities and N control delays corresponding to the N flywheel energy storage units by performing rotor control sensitivity identification and resistance control delay analysis on N operating sample data sets of N flywheel energy storage units is achieved, thereby improving the accuracy of energy storage regulation and control of the flywheel energy storage device.

[0031] Step S520: using the control sensitivity and the control delay as loss variables, outputting a loss function, embedding the loss function as an incentive layer in the collaborative control model, and obtaining N control losses corresponding to the N flywheel energy storage units;

[0032] Step S530: utilizing the collaborative control model to minimize the N control losses and to adapt to the target, outputting N control parameters corresponding to the N flywheel energy storage units.

[0033] Step S600: inputting the N control parameters into the frequency converter to coordinately control the rotation of the motors of the N flywheel energy storage units.

[0034] Specifically, the preset energy regulation capacity is set as the control target, and the N flywheel energy storage units are set as the adaptation variable. The N control sensitivities and N control delays corresponding to the N flywheel energy storage units are respectively used as loss variables to obtain the loss function, and the loss function is set as the excitation layer. The excitation layer is embedded in the collaborative control model to obtain the N control losses corresponding to the N flywheel energy storage units. The collaborative control model takes minimizing the sum of N control losses as the adaptation target, outputs N control parameters corresponding to the N flywheel energy storage units, and inputs the N control parameters into the inverter, and performs collaborative control of the motor rotation of the N flywheel energy storage units according to the N control parameters.

[0035] Among them, the expression of the collaborative control model is:

[0036] ;

[0037] Among them, by minimizing Output the parameters that need to be learned, Represents the real-time speed of the i-th flywheel energy storage unit, represents the speed that needs to be learned by the i-th flywheel energy storage unit; N is the total number of flywheel energy storage units;

[0038] is the control sensitivity of the i-th flywheel energy storage unit, is the control delay of the i-th flywheel energy storage unit; is the loss function for the real-time control time of the i-th flywheel energy storage unit; It is a preset time loss function based on the synergy condition in the i-th flywheel energy storage unit.

[0039] The loss variables include N control sensitivities and N control delays corresponding to the N flywheel energy storage units. The loss function includes , The incentive layer includes the loss function. N control losses and include The N control parameters include N motor speeds corresponding to the N flywheel energy storage units.

[0040] The technical effect of improving the energy storage regulation control quality of the flywheel energy storage device by conducting collaborative control analysis on the flywheel energy storage unit under a preset energy regulation capacity through a collaborative control model is achieved.

[0041] Furthermore, step S600 of the present application also includes:

[0042] Step S610: acquiring the flywheel storage capacity, the flywheel charging rate, and the flywheel discharging rate by analyzing the N running sample data sets;

[0043] Step S620: Analyze the flywheel storage capacity, the flywheel charging rate, and the flywheel discharging rate, make a prediction based on the preset energy regulation capacity, obtain the predicted duration, and schedule the operation sequence of the collaborative control model based on the predicted duration, wherein the predicted duration is the predicted energy regulation duration of the first flywheel energy storage device based on the preset energy regulation capacity.

[0044] Furthermore, step S620 of the present application also includes:

[0045] Step S621: judging the frequency modulation mode of the preset energy modulation capacity, and when the frequency modulation mode of the preset energy modulation capacity is the energy storage charging mode, outputting a first predicted duration;

[0046] Step S622: when the frequency modulation mode of the preset energy modulation capacity is the energy storage discharge mode, outputting a second predicted duration;

[0047] Step S623: Schedule the job sequence of the collaborative control model according to the first predicted duration / the second predicted duration.

[0048] Specifically, N running sample data sets are traversed to extract data of flywheel storage capacity, flywheel charging rate, and flywheel discharging rate to obtain N flywheel energy storage data sets. Each flywheel energy storage data set includes the real-time flywheel storage capacity, real-time flywheel charging rate, and real-time flywheel discharging rate in each running sample data set.

[0049] Further, when the frequency modulation mode is identified as an energy storage charging mode, the energy storage charging duration is predicted based on N flywheel energy storage data sets, the preset charging capacity in the first energy modulation instruction, and N control parameters to obtain a first predicted duration. The motor rotation of the N flywheel energy storage units is collaboratively controlled according to the first predicted duration and the N control parameters. Similarly, when the frequency modulation mode is identified as an energy storage discharge mode, the energy storage discharge duration is predicted based on N flywheel energy storage data sets, the preset discharge capacity in the first energy modulation instruction, and the N control parameters to obtain a second predicted duration. The motor rotation of the N flywheel energy storage units is collaboratively controlled according to the second predicted duration and the N control parameters.

[0050] Wherein, the predicted duration includes a first predicted duration / a second predicted duration. The first predicted duration includes, under N flywheel energy storage data sets, when the frequency modulation mode is identified as an energy storage charging mode, each of the N control parameters satisfies the predicted energy storage charging duration required for the preset charging capacity. The second predicted duration includes, under N flywheel energy storage data sets, when the frequency modulation mode is identified as an energy storage discharging mode, each of the N control parameters satisfies the predicted energy storage discharging duration required for the preset charging capacity.

[0051] The technical effect of coordinated control of motor rotation of N flywheel energy storage units according to predicted duration and N control parameters is achieved, thereby improving the comprehensiveness of energy storage regulation and control of the flywheel energy storage device.

[0052] Exemplarily, when predicting the energy storage charging duration based on N flywheel energy storage data sets, the preset charging capacity in the first energy adjustment instruction, and N control parameters, historical data query is performed according to the N flywheel energy storage data sets, the preset charging capacity, and the N control parameters to obtain multiple sets of energy storage charging duration data. Each set of energy storage charging duration data includes N historical flywheel energy storage data sets, historical preset charging capacity, N historical control parameters, and under the N historical flywheel energy storage data sets, when the frequency modulation mode is identified as the energy storage charging mode, each of the N historical control parameters satisfies the historical energy storage charging duration corresponding to the preset charging capacity. Then, based on the fully connected neural network, multiple sets of energy storage charging duration data are cross-supervised trained to obtain an energy storage charging duration prediction model. Furthermore, the N flywheel energy storage data sets, the preset charging capacity in the first energy adjustment instruction, and the N control parameters are input into the energy storage charging duration prediction model to obtain the first predicted duration. Among them, the fully connected neural network is also called a multi-layer perceptron, and the fully connected neural network is an artificial neural network structure with a relatively simple connection method. A fully connected neural network is a feedforward neural network consisting of an input layer, a hidden layer, and an output layer. Moreover, there can be multiple neurons in the hidden layer. The energy storage charging time prediction model includes an input layer, a hidden layer, and an output layer. The energy storage charging time prediction model has the function of predicting the energy storage discharge time for the input N flywheel energy storage data sets, the preset charging capacity in the first energy adjustment instruction, and N control parameters. The energy storage discharge time prediction is the same as the energy storage charging time prediction. For the sake of brevity of the specification, it will not be repeated here.

[0053] Further, as attached Figure 2 As shown, step S600 of the present application also includes:

[0054] Step S630: collecting real-time energy storage capacity of the N flywheel energy storage units of the first flywheel energy storage device, and outputting N real-time energy storage capacities;

[0055] Step S640: Compare the N real-time energy storage capacities with the preset energy regulation capacity, and output M matching flywheel energy storage units;

[0056] Step S650: Calculate the M flywheel energy storage units based on the collaborative control model and output M control parameters, where M≤N.

[0057] Specifically, N real-time energy storage capacities are extracted from N running sample data sets. The ratios of the N real-time energy storage capacities and the preset energy regulation capacities are calculated respectively to obtain N energy storage-energy regulation capacity ratios. Then, each energy storage-energy regulation capacity ratio is judged to be greater than / equal to the preset energy storage-energy regulation capacity constraint condition. When the energy storage-energy regulation capacity ratio is greater than / equal to the preset energy storage-energy regulation capacity constraint condition, the flywheel energy storage unit corresponding to the energy storage-energy regulation capacity ratio is added to the M flywheel energy storage units, and the M flywheel energy storage units are matched with N control parameters to obtain M control parameters. The first flywheel energy storage device is collaboratively controlled to rotate the motor according to the M control parameters. Among them, the N energy storage-energy regulation capacity ratios include N ratios between the N real-time energy storage capacities and the preset energy regulation capacity. The preset energy storage-energy regulation capacity constraint condition includes an energy storage-energy regulation capacity ratio threshold value pre-set and determined by the intelligent collaborative control system for distributed flywheel energy storage devices of the present application. The M flywheel energy storage units include multiple flywheel energy storage units corresponding to multiple energy storage-energy regulation capacity ratios greater than / equal to the preset energy storage-energy regulation capacity constraint conditions among the N flywheel energy storage units. And, M≤N. The M control parameters include multiple control parameters corresponding to the M flywheel energy storage units among the N control parameters.

[0058] The technical effect of obtaining M control parameters by comparing N real-time energy storage capacities with preset energy regulation capacities is achieved, thereby improving the energy storage regulation control adaptability of the flywheel energy storage device and avoiding waste of the energy storage capacity of the flywheel energy storage device.

[0059] Furthermore, step S640 of the present application also includes:

[0060] Step S641: identifying abnormal data on the N running sample data sets, and locating the abnormal flywheel energy storage unit according to the abnormal data identification result;

[0061] Step S642: Screen out the abnormal flywheel energy storage unit from the N flywheel energy storage units, and generate first abnormal warning information for reminding the operation abnormality of the flywheel energy storage unit.

[0062] Specifically, abnormal data identification is performed on N operating sample data sets to obtain abnormal data identification results, and N flywheel energy storage units are matched according to the abnormal data identification results to obtain abnormal flywheel energy storage units. Then, the abnormal flywheel energy storage unit is screened out from the N flywheel energy storage units, and the first abnormal warning information corresponding to the abnormal flywheel energy storage unit is generated. Among them, the abnormal data identification result includes N operating sample data sets, and whether abnormal data exists in each operating sample data set. And, when abnormal data exists in the operating sample data set, the abnormal data has a corresponding abnormal impact index. The higher the abnormal impact degree of the abnormal data, the greater the corresponding abnormal impact index. The abnormal flywheel energy storage unit includes a flywheel energy storage unit with abnormal data in the abnormal data identification result. The first abnormal warning information is a warning prompt information for abnormal operation of the abnormal flywheel energy storage unit.

[0063] Exemplarily, when abnormal data is identified for N operating sample data sets, historical data query is performed based on the N operating sample data sets to obtain multiple sets of abnormal data identification sets. Each set of abnormal data identification sets includes N historical operating sample data sets and historical abnormal data identification results corresponding to the N historical operating sample data sets. Then, based on the BP neural network, the multiple sets of abnormal data identification sets are cross-supervised trained to obtain an abnormal data detection model. The N operating sample data sets are input into the abnormal data detection model, and abnormal data identification is performed on the N operating sample data sets through the abnormal data detection model to obtain abnormal data identification results. The abnormal data detection model includes an input layer, a hidden layer, and an output layer.

[0064] The technical effect of identifying abnormal data of N operating sample data sets, thereby adaptively issuing abnormal warnings to N flywheel energy storage units and improving the comprehensiveness of energy storage regulation and control of the flywheel energy storage device is achieved.

[0065] In summary, the intelligent collaborative control method for a distributed flywheel energy storage device provided in this application has the following technical effects:

[0066] 1. By collecting the operation data of the N flywheel energy storage units of the first flywheel energy storage device, N operation sample data sets are obtained; when the first flywheel energy storage device receives the first energy adjustment instruction, the preset energy adjustment capacity is used as the control target, and the N flywheel energy storage units are used as the adaptive variables, which are input into the collaborative control model, and N control parameters are output according to the collaborative control model; the N control parameters are input into the frequency converter to collaboratively control the motor rotation of the N flywheel energy storage units. The technical effect of improving the energy storage regulation control accuracy of the flywheel energy storage device, improving the energy storage regulation control effect of the flywheel energy storage device, and improving the reliability, timeliness, and adaptability of the power supply of the flywheel energy storage device is achieved.

[0067] 2. By performing rotor control sensitivity identification and resistance control delay analysis on N operating sample data sets of N flywheel energy storage units, N control sensitivities and N control delays corresponding to the N flywheel energy storage units are obtained, thereby improving the accuracy of energy storage regulation and control of the flywheel energy storage device.

[0068] Embodiment 2:

[0069] Based on the intelligent collaborative control method for a distributed flywheel energy storage device in the aforementioned embodiment, the present invention also provides an intelligent collaborative control system for a distributed flywheel energy storage device, see the attached Figure 3 , the system comprising:

[0070] An energy storage unit acquisition module 11, wherein the energy storage unit acquisition module 11 is used to acquire N flywheel energy storage units of a first flywheel energy storage device;

[0071] An operation data acquisition module 12, wherein the operation data acquisition module 12 is used to collect operation data of the N flywheel energy storage units to obtain N operation sample data sets;

[0072] An energy regulation instruction receiving module 13, wherein the energy regulation instruction receiving module 13 is used for receiving a first energy regulation instruction when the first flywheel energy storage device receives the first energy regulation instruction, wherein the first energy regulation instruction includes a preset energy regulation capacity;

[0073] An input module 14, wherein the input module 14 is used to input the preset energy regulation capacity as a control target and the N flywheel energy storage units as adaptation variables into a collaborative control model, wherein the collaborative control model is obtained by training the N running sample data sets until convergence, and controls the loss fitness of the collaborative control between the flywheel energy storage units;

[0074] A control parameter output module 15, the control parameter output module 15 is used to output N control parameters according to the collaborative control model, wherein the N control parameters include N motor speeds, and the N motor speeds correspond to the N flywheel energy storage units;

[0075] The collaborative control module 16 is used to input the N control parameters into the frequency converter to perform collaborative control on the motor rotation of the N flywheel energy storage units.

[0076] Furthermore, the system also includes:

[0077] A flywheel identification module, the flywheel identification module is used to identify the N running sample data sets to obtain the control sensitivity of the flywheel rotor and the control delay degree of the resistance experienced by the flywheel;

[0078] A control loss acquisition module, the control loss acquisition module is used to use the control sensitivity and the control delay as loss variables, output a loss function, embed the loss function as an incentive layer in the collaborative control model, and obtain N control losses corresponding to the N flywheel energy storage units;

[0079] A control parameter determination module is used to utilize the collaborative control model to minimize the N control losses and to output N control parameters corresponding to the N flywheel energy storage units to meet the target.

[0080] The expression of the collaborative control model is as follows:

[0081] ;

[0082] Among them, by minimizing Output the parameters that need to be learned, Represents the real-time speed of the i-th flywheel energy storage unit, represents the speed that needs to be learned by the i-th flywheel energy storage unit; N is the total number of flywheel energy storage units;

[0083] is the control sensitivity of the i-th flywheel energy storage unit, is the control delay of the i-th flywheel energy storage unit; is the loss function for the real-time control time of the i-th flywheel energy storage unit; It is a preset time loss function based on the synergy condition in the i-th flywheel energy storage unit.

[0084] Furthermore, the system also includes:

[0085] A first execution module, the first execution module is used to obtain the flywheel storage capacity, the flywheel charging rate and the flywheel discharging rate by analyzing the N running sample data sets;

[0086] A predicted duration acquisition module, the predicted duration acquisition module is used to analyze the flywheel storage capacity, the flywheel charging rate and the flywheel discharging rate, predict with the preset energy regulation capacity as the target, obtain the predicted duration, and schedule the job timing of the collaborative control model based on the predicted duration;

[0087] The predicted duration is a predicted energy regulation duration of the first flywheel energy storage device under the preset energy regulation capacity condition.

[0088] Furthermore, the system also includes:

[0089] A first predicted duration output module, the first predicted duration output module is used to judge the frequency modulation mode of the preset energy modulation capacity, and output a first predicted duration when the frequency modulation mode of the preset energy modulation capacity is an energy storage charging mode;

[0090] A second predicted duration output module, the second predicted duration output module is used to output a second predicted duration when the frequency modulation mode of the preset energy modulation capacity is an energy storage discharge mode;

[0091] A second execution module, the second execution module is used to schedule the job timing of the collaborative control model according to the first predicted duration / second predicted duration.

[0092] Furthermore, the system also includes:

[0093] A real-time energy storage capacity determination module, the real-time energy storage capacity determination module is used to collect real-time energy storage capacity of the N flywheel energy storage units of the first flywheel energy storage device and output N real-time energy storage capacities;

[0094] A third execution module, the third execution module is used to compare the N real-time energy storage capacities with the preset energy regulation capacity, and output M matching flywheel energy storage units;

[0095] A fourth execution module, the fourth execution module is used to calculate the M flywheel energy storage units based on the collaborative control model, and output M control parameters, wherein M≤N.

[0096] Furthermore, the system also includes:

[0097] An abnormality locating module, the abnormality locating module is used to identify abnormal data on the N running sample data sets, and locate the abnormal flywheel energy storage unit according to the abnormal data identification result;

[0098] A warning information generation module is used to screen out the abnormal flywheel energy storage unit from the N flywheel energy storage units and generate first abnormal warning information for reminding the operation abnormality of the flywheel energy storage unit.

[0099] The intelligent collaborative control system for a distributed flywheel energy storage device provided in an embodiment of the present invention can execute the intelligent collaborative control method for a distributed flywheel energy storage device provided in any embodiment of the present invention, and has functional modules and beneficial effects corresponding to the execution method.

[0100] The modules included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be realized; in addition, the specific names of the functional modules are only for the convenience of distinguishing each other and are not used to limit the scope of protection of the present invention.

[0101] The present application provides an intelligent collaborative control method for a distributed flywheel energy storage device, wherein the method is applied to an intelligent collaborative control system of a distributed flywheel energy storage device, and the method includes: acquiring N operational sample data sets by collecting operational data of N flywheel energy storage units of a first flywheel energy storage device; when the first flywheel energy storage device receives a first energy regulation instruction, taking a preset energy regulation capacity as a control target, taking N flywheel energy storage units as adaptive variables, inputting them into a collaborative control model, and outputting N control parameters according to the collaborative control model; inputting the N control parameters into a frequency converter, and collaboratively controlling the motor rotation of the N flywheel energy storage units. The technical problem that the energy storage regulation control accuracy of a flywheel energy storage device is low, resulting in poor energy storage regulation control effect of the flywheel energy storage device, is solved in the prior art. The technical effect of improving the energy storage regulation control accuracy of a flywheel energy storage device, improving the energy storage regulation control effect of a flywheel energy storage device, and improving the reliability, timeliness, and adaptability of power supply of a flywheel energy storage device is achieved.

[0102] Note that the above are only preferred embodiments of the present invention and the technical principles used. Those skilled in the art will understand that the present invention is not limited to the specific embodiments described herein, and that various obvious changes, readjustments and substitutions can be made by those skilled in the art without departing from the scope of protection of the present invention. Therefore, although the present invention has been described in more detail through the above embodiments, the present invention is not limited to the above embodiments, and may include more other equivalent embodiments without departing from the concept of the present invention, and the scope of the present invention is determined by the scope of the appended claims.

Claims

1. An intelligent collaborative control method for a distributed flywheel energy storage device, characterized in that: The method comprises: Obtaining N flywheel energy storage units of a first flywheel energy storage device; Collecting operation data of the N flywheel energy storage units to obtain N operation sample data sets; When the first flywheel energy storage device receives a first energy regulation instruction, wherein the first energy regulation instruction includes a preset energy regulation capacity; The preset energy regulation capacity is used as a control target, and the N flywheel energy storage units are used as adaptation variables, which are input into a collaborative control model, wherein the collaborative control model is obtained by training the N running sample data sets until convergence, and controls the loss fitness of the collaborative control between the flywheel energy storage units; Outputting N control parameters according to the collaborative control model, wherein the N control parameters include N motor speeds, and the N motor speeds correspond to the N flywheel energy storage units; Inputting the N control parameters into a frequency converter to coordinately control the rotation of the motors of the N flywheel energy storage units; The method further comprises: Identify according to the N running sample data sets to obtain the control sensitivity of the flywheel rotor and the control delay degree of the resistance experienced by the flywheel; The control sensitivity and the control delay are respectively used as loss variables, a loss function is output, and the loss function is embedded in the collaborative control model as an incentive layer to obtain N control losses corresponding to the N flywheel energy storage units respectively; Using the collaborative control model, to minimize the N control losses and to adapt to the target, output N control parameters corresponding to the N flywheel energy storage units; The expression of the collaborative control model is as follows: ， Among them, by minimizing Output the parameters that need to be learned, Represents the real-time speed of the i-th flywheel energy storage unit, represents the speed that needs to be learned by the i-th flywheel energy storage unit; N is the total number of flywheel energy storage units; is the control sensitivity of the i-th flywheel energy storage unit, is the control delay of the i-th flywheel energy storage unit; is the loss function for the real-time control time of the i-th flywheel energy storage unit; It is a preset time loss function based on the synergy condition in the i-th flywheel energy storage unit.

2. The method according to claim 1, characterized in that The method further comprises: By analyzing the N running sample data sets, a flywheel storage capacity, a flywheel charging rate, and a flywheel discharging rate are obtained; Analyze the flywheel storage capacity, the flywheel charging rate, and the flywheel discharging rate, predict with the preset energy regulation capacity as the target, obtain the predicted duration, and schedule the operation sequence of the collaborative control model based on the predicted duration; The predicted duration is a predicted energy regulation duration of the first flywheel energy storage device under the preset energy regulation capacity condition.

3. The method according to claim 2, characterized in that The method further comprises: Determining the frequency modulation mode of the preset energy modulation capacity, and when the frequency modulation mode of the preset energy modulation capacity is an energy storage charging mode, outputting a first predicted duration; When the frequency modulation mode of the preset energy modulation capacity is the energy storage discharge mode, outputting a second predicted duration; The job timing of the collaborative control model is scheduled according to the first predicted duration / the second predicted duration.

4. The method according to claim 1, characterized in that The method further comprises: Performing real-time energy storage collection on the N flywheel energy storage units of the first flywheel energy storage device, and outputting N real-time energy storage capacities; Compare the N real-time energy storage capacities with the preset energy regulation capacity, and output M matching flywheel energy storage units; The M flywheel energy storage units are calculated based on the collaborative control model, and M control parameters are output, wherein M≤N.

5. The method according to claim 4, characterized in that The method further comprises: By performing abnormal data identification on the N running sample data sets, the abnormal flywheel energy storage unit is located according to the abnormal data identification result; The abnormal flywheel energy storage unit is screened out from the N flywheel energy storage units, and first abnormal warning information is generated to remind the operator of abnormal operation of the flywheel energy storage unit.

6. An intelligent collaborative control system for a distributed flywheel energy storage device, characterized in that: The system is used to perform the method according to any one of claims 1 to 5, and the system comprises: An energy storage unit acquisition module, the energy storage unit acquisition module is used to acquire N flywheel energy storage units of the first flywheel energy storage device; An operation data acquisition module, wherein the operation data acquisition module is used to collect operation data of the N flywheel energy storage units to obtain N operation sample data sets; An energy regulation instruction receiving module, the energy regulation instruction receiving module is used for receiving a first energy regulation instruction when the first flywheel energy storage device receives a first energy regulation instruction, wherein the first energy regulation instruction includes a preset energy regulation capacity; An input module, the input module is used to input the preset energy regulation capacity as a control target and the N flywheel energy storage units as adaptation variables into a collaborative control model, wherein the collaborative control model is obtained by training the N running sample data sets until convergence, and controls the loss fitness of the collaborative control between the flywheel energy storage units; A control parameter output module, the control parameter output module is used to output N control parameters according to the collaborative control model, wherein the N control parameters include N motor speeds, and the N motor speeds correspond to the N flywheel energy storage units; A collaborative control module is used to input the N control parameters into a frequency converter to collaboratively control the motor rotation of the N flywheel energy storage units.

Images