Hybrid energy storage system power distribution method and system, terminal and storage medium

Abstract

The invention relates to the technical field of hybrid energy storage systems, and particularly provides a hybrid energy storage system power distribution method and system, a terminal and a storage medium, and the method comprises the steps: calculating hybrid energy storage power based on wind power plant generation power and grid-connected power; the historical hybrid energy storage power is decomposed based on an EMD algorithm, and historical optimal EMD parameters are determined according to historical hybrid energy storage system features and decomposition results; a BiLSTM model is trained; obtaining a current optimal EMD parameter based on the current hybrid energy storage power and a BiLSTM model, and decomposing the hybrid energy storage power into a plurality of MIF components based on an EMD algorithm; and dividing the multiple MIF components into a high-frequency part and a low-frequency part based on a preset frequency threshold, reconstructing the high-frequency part to obtain flywheel energy storage system power, and reconstructing the low-frequency part to obtain battery energy storage system power. Cooperative operation of the wind power plant and the hybrid energy storage system is optimized, and the performance and efficiency of the energy storage system are improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of hybrid energy storage systems, and particularly relates to a power distribution method, system, terminal and storage medium for a hybrid energy storage system. Background Art

[0002] The existing requirements for the active power of a wind farm mainly include: the wind farm should have the ability to control active power, the maximum power change rate of the wind farm, and the ability to control the active power of the wind farm under the emergency state of the power grid.

[0003] Adding an energy storage system to a wind farm can effectively smooth and suppress the fluctuation of wind power, and improve the power consumption capacity of the power system for wind power. According to the characteristics of the energy storage, it is divided into energy-type energy storage and power-type energy storage.

[0004] The energy-type energy storage system is represented by the battery energy storage system, which has the advantage of large energy density, but the disadvantages of small power density and long response time, and is usually used to handle power fluctuations with high energy and low frequency; the power-type energy storage system is represented by supercapacitors, superconducting magnets and flywheel energy storage, which has the advantages of large power density and frequent charge and discharge, but the disadvantage of small energy density, and is usually used to handle power fluctuations with low energy and high frequency. Summary of the Invention

[0005] In view of the above deficiencies of the prior art, the present invention provides a power distribution method, system, terminal and storage medium for a hybrid energy storage system to solve the above technical problems.

[0006] In a first aspect, the present invention provides a power distribution method for a hybrid energy storage system, including: S1, collecting the wind farm power generation and the characteristics of the hybrid energy storage system, extracting the grid-connected power from the wind farm power generation based on the adaptive sliding average method, and calculating the hybrid energy storage power based on the wind farm power generation and the grid-connected power; S2, obtaining the historical hybrid energy storage power and the historical hybrid energy storage system characteristics based on S1, decomposing the historical hybrid energy storage power based on the EMD algorithm, and determining the historical optimal EMD parameters according to the historical hybrid energy storage system characteristics and the decomposition results; S3, training a BiLSTM model based on the historical hybrid energy storage power, the historical optimal EMD parameters and the historical hybrid energy storage system characteristics; S4, obtaining the current optimal EMD parameters based on the current hybrid energy storage power and the BiLSTM model, and decomposing the hybrid energy storage power into multiple MIF components based on the EMD algorithm of the current optimal EMD parameters; S5, dividing the multiple MIF components into a high-frequency part and a low-frequency part based on a preset frequency threshold, reconstructing the high-frequency part to obtain the flywheel energy storage system power, and reconstructing the low-frequency part to obtain the battery energy storage system power.

[0007] In an optional embodiment, in step S1, extracting the grid-connected power from the wind farm power generation based on the adaptive moving average method specifically includes: Obtain the time series of the wind farm power generation, and set the initial moving window size based on the frequency range of the wind farm power generation fluctuation and the desired smoothing degree; Initialize the weights of each time point data when calculating the average value, and each time point data has the same weight within the window; Calculate the moving average value of each time point data according to the set moving window size and weights. The moving average values of all time point data form the grid-connected power sequence, and the moving average value of each time point data is calculated as follows:

[0008] where i is the summation index, j is the time point, is the moving average value of the jth time point, is the data of the (j + i)th time point, m is the summation boundary, N is the window size, N = 2m + 1, is the weight of the data of the ith time point.

[0009] In an optional embodiment, the specific steps of step S3 include: Construct a data set based on the historical hybrid energy storage power, the historical optimal EMD parameters, and the historical hybrid energy storage system characteristics, and divide the data set into a training set, a test set, and a validation set; The input feature vector of the BiLSTM model is the historical hybrid energy storage power and the historical hybrid energy storage system characteristics. Determine the dimension of the input feature vector and set the number of input layer nodes. The output feature vector of the BiLSTM model is the historical optimal EMD parameters. Determine the dimension of the output feature vector and set the number of output layer nodes; Initialize the weights and biases of the BiLSTM model; For each sample in the training set, input the input feature vector into the BiLSTM model, and calculate the predicted output vector of the model through forward propagation; Calculate the loss value between the predicted output vector and the actual output feature vector according to the loss function, and use the backpropagation algorithm to calculate the gradients of the loss function with respect to the model weights and biases, and update the model weights and biases using the gradient information; Calculate the evaluation index based on the validation set combined with the root mean square error function every preset number of training times, and adjust the model structure based on the change trend of the evaluation index; Finally evaluate the trained BiLSTM model based on the test set.

[0010] In an alternative embodiment, in the forward propagation calculation, it specifically includes: Performing a forward LSTM calculation on the input feature vector to obtain the output result of the forward layer, specifically:

[0011] Wherein, is the output result of the forward layer at time t, is the output result of the forward layer at time t-1, is the weight coefficient between the input layer and the forward layer, is the weight coefficient between the forward propagation unit at time t-1 and the forward propagation unit at time t, is the calculation function inside the neuron; Performing a backward LSTM calculation on the input feature vector to obtain the output result of the backward layer, specifically:

[0012] Wherein, is the output result of the backward layer at time t, is the output result of the backward layer at time t+1, is the weight coefficient between the input layer and the backward layer, is the weight coefficient between the forward propagation unit at time t+1 and the forward propagation unit at time t; Weightedly fusing the output results of the forward layer and the backward layer to obtain the output feature vector of the BiLSTM model;

[0013] Wherein, is the weight coefficient between the forward layer and the output layer, is the weight coefficient between the backward layer and the output layer, is the function for calculating the total result of the forward layer and the backward layer, is the output feature vector of the BiLSTM model.

[0014] In an alternative embodiment, the EMD algorithm based on the current optimal EMD parameters decomposes the hybrid energy storage power into multiple MIF components, which specifically includes: S4-1, obtaining the hybrid energy storage power sequence, finding all the maximum points and minimum points of the hybrid energy storage power sequence, and respectively fitting the maximum envelope line and the minimum envelope line of the signal through a cubic spline function; S4-2, calculating the average value based on the maximum envelope line and the minimum envelope line; S4-3, subtracting the average value from the hybrid energy storage power sequence to calculate the low-frequency signal sequence; S4-4. Determine whether the number of zeros and the number of extreme points in the low-frequency-removed signal sequence are equal or differ by at most one, and determine whether the means of the maximum envelope and the minimum envelope in the low-frequency-removed signal sequence are equal and are 0; S4-5. If the above judgments on the low-frequency-removed signal sequence are all negative, then replace the hybrid energy storage power sequence with the low-frequency-removed signal sequence, and repeat steps S4-1, S4-2, S4-3, and S4-4 until the above judgments on the low-frequency-removed signal sequence are all positive, and obtain the first IMF component; S4-6. Subtract the IMF component from the hybrid energy storage power sequence to obtain a residual signal sequence, replace the hybrid energy storage power sequence with the residual signal sequence, and repeat the execution of S4-1, S4-2, S4-3, and S4-4 to obtain the second IMF component; S4-7. Based on the optimal decomposition level and the optimal iteration termination parameter, repeat S4-6 to obtain multiple IMF components and an inseparable sequence.

[0015] In an alternative embodiment, in step S5, The power of the flywheel energy storage system obtained by reconstructing the high-frequency part is calculated as:

[0016] The power of the battery energy storage system obtained by reconstructing the low-frequency part is calculated as:

[0017] Wherein, is the power of the flywheel energy storage system, is the high-frequency reconstruction power, is the power of the battery energy storage system, is the low-frequency reconstruction power, is the high-low frequency demarcation point corresponding to the frequency threshold, is the kth IMF component.

[0018] In an alternative embodiment, in step S1, obtaining the characteristics of the hybrid energy storage system specifically includes: In the battery energy storage system, calculate the charge and discharge efficiency of the battery, calculate the first-order difference of the battery SOC to obtain the change characteristics of the battery charge and discharge state, and calculate the mean and variance of the battery temperature to obtain the change characteristics of the TV temperature; In the flywheel energy storage system, calculate the change rate of the flywheel speed, and calculate the mean and variance of the change of the flywheel charge and discharge power to obtain the change characteristics of the flywheel charge and discharge power.

[0019] In a second aspect, the present invention provides a power distribution system for a hybrid energy storage system. When the system is implemented, it executes the above-mentioned power distribution method for the hybrid energy storage system. The hybrid energy storage system includes a battery energy storage system and a flywheel energy storage system. The system includes: The hybrid energy storage power acquisition module collects the power generation power of the wind farm and the characteristics of the hybrid energy storage system, extracts the grid-connected power from the wind farm power generation power based on the adaptive sliding average method, and calculates the hybrid energy storage power based on the wind farm power generation power and the grid-connected power; The historical data acquisition module obtains the historical hybrid energy storage power and the historical hybrid energy storage system characteristics based on S1, decomposes the historical hybrid energy storage power based on the EMD algorithm, and determines the historical optimal EMD parameters according to the historical hybrid energy storage system characteristics and the decomposition results; The BiLSTM model training module trains the BiLSTM model based on the historical hybrid energy storage power, the historical optimal EMD parameters, and the historical hybrid energy storage system characteristics; The hybrid energy storage power decomposition module obtains the current optimal EMD parameters based on the current hybrid energy storage power and the BiLSTM model, and decomposes the hybrid energy storage power into multiple MIF components based on the EMD algorithm of the current optimal EMD parameters; The high and low frequency reconstruction module divides multiple MIF components into a high frequency part and a low frequency part based on a preset frequency threshold, reconstructs the high frequency part to obtain the flywheel energy storage system power, and reconstructs the low frequency part to obtain the battery energy storage system power.

[0020] In a third aspect, a terminal is provided, including: A processor and a memory, wherein, The memory is used to store a computer program, The processor is used to call and run the computer program from the memory, so that the terminal executes the method of the above terminal.

[0021] In a fourth aspect, a computer-readable storage medium is provided, and instructions are stored in the computer-readable storage medium, and when it runs on a computer, it causes the computer to execute the methods described in the above aspects.

[0022] The beneficial effects of the present invention are that the hybrid energy storage system power distribution method, system, terminal and storage medium provided by the present invention extract the grid-connected power to calculate the hybrid energy storage power, use historical data to determine the optimal parameters and train the model, then combine the current data to obtain the current optimal parameters to decompose the power, and finally divide and reconstruct according to the frequency threshold to obtain the flywheel energy storage system power and the battery energy storage system power. This process comprehensively considers various factors and data, improves the accuracy and rationality of the hybrid energy storage system power distribution, optimizes the coordinated operation of the wind farm and the hybrid energy storage system, and improves the performance and efficiency of the energy storage system.

[0023] In addition, the design principle of the present invention is reliable, the structure is simple, and it has a very wide application prospect. Description of the Drawings

[0024] To more clearly illustrate the technical solution of the present invention, the accompanying drawings required in the description will be briefly introduced below. Obviously, the accompanying drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other accompanying drawings can be obtained based on these drawings.

[0025] Figure 1 It is a schematic flowchart of the power distribution method for the hybrid energy storage system according to an embodiment of the present invention.

[0026] Figure 2 It is a schematic structural diagram of the hybrid energy storage system according to an embodiment of the present invention.

[0027] Figure 3 It is a schematic diagram of power distribution according to an embodiment of the present invention.

[0028] Figure 4 It is a structural diagram of the storage unit of BiLSTM according to an embodiment of the present invention.

[0029] Figure 5 It is a flowchart of the EMD algorithm according to an embodiment of the present invention.

[0030] Figure 6 It is a schematic block diagram of the power distribution system for the hybrid energy storage system according to an embodiment of the present invention.

[0031] Figure 7 It is a schematic structural diagram of a terminal provided by an embodiment of the present invention. Detailed implementation manners

[0032] In order to enable those skilled in the art of the present technology to better understand the technical solutions in the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, rather than all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0033] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the technical field of the present invention. The terms used in the description of the present invention herein are only for the purpose of describing specific embodiments, and are not intended to limit the present invention.

[0034] The power distribution method for the hybrid energy storage system provided by the embodiments of the present invention is executed by a computer device. Correspondingly, the power distribution system for the hybrid energy storage system runs in the computer device.

[0035] Figure 1It is a schematic flowchart of the power distribution method of the hybrid energy storage system according to an embodiment of the present invention. Among them, Figure 1 The execution subject can be a power distribution system of a hybrid energy storage system. According to different requirements, the order of the steps in this flowchart can be changed, and some can be omitted.

[0036] As Figure 1 shown, the method includes: S1, collect the wind farm power generation power and the characteristics of the hybrid energy storage system, extract the grid-connected power from the wind farm power generation power based on the adaptive sliding average method, and calculate the hybrid energy storage power based on the wind farm power generation power and the grid-connected power; The adaptive sliding average method can effectively smooth the fluctuations in the wind farm power generation power, more accurately extract the power suitable for grid connection, and provide a reasonable basis for the subsequent calculation of the hybrid energy storage system power. At the same time, considering the characteristics of the hybrid energy storage system, the calculated hybrid energy storage power is more in line with the actual situation of the system, which helps to improve the accuracy and adaptability of the overall system power management.

[0037] S2, based on the historical hybrid energy storage power and the historical hybrid energy storage system characteristics obtained in S1, decompose the historical hybrid energy storage power based on the EMD algorithm, and determine the historical optimal EMD parameters according to the historical hybrid energy storage system characteristics and the decomposition results; By decomposing historical data through the EMD algorithm, the internal composition and change law of the hybrid energy storage power can be deeply understood. Determining the optimal parameters in combination with the system characteristics can make the decomposition process more accurately reflect the actual situation, improve the accuracy of subsequent model training and power analysis, and provide strong support for the refined management of the hybrid energy storage system. S3, train the BiLSTM model based on the historical hybrid energy storage power, the historical optimal EMD parameters and the historical hybrid energy storage system characteristics; The BiLSTM model can effectively learn the time series information and complex patterns in the historical data. Combining the optimal parameters and system characteristics determined above, the model can better capture the change trend and law of the hybrid energy storage power. This helps to improve the accuracy and reliability of the model's prediction of the hybrid energy storage power, and provides an effective decision-making basis for the real-time control and optimization of the system.

[0038] S4, obtain the current optimal EMD parameters based on the current hybrid energy storage power and the BiLSTM model, and decompose the hybrid energy storage power into multiple MIF components based on the EMD algorithm of the current optimal EMD parameters; Adjust the decomposition parameters in real time according to the current system state to make the power decomposition more in line with the current actual situation. The obtained multiple MIF components can more finely reflect the composition structure of the hybrid energy storage power, provide an accurate basis for subsequent division and reconstruction according to the frequency threshold, and help to improve the accuracy and rationality of the energy storage system power distribution.

[0039] S5, divide multiple MIF components into a high-frequency part and a low-frequency part based on a preset frequency threshold, reconstruct the high-frequency part to obtain the power of the flywheel energy storage system, and reconstruct the low-frequency part to obtain the power of the battery energy storage system.

[0040] The high-frequency part is suitable for the flywheel energy storage system, and the low-frequency part is suitable for the battery energy storage system. This can give full play to the respective advantages of the flywheel energy storage system and the battery energy storage system, improve the performance and efficiency of the entire hybrid energy storage system, and achieve the coordinated and optimized operation of the wind farm and the hybrid energy storage system.

[0041] Optionally, as an embodiment of the present invention, refer to Figure 2 and Figure 3 , a wind power, flywheel-battery hybrid energy storage system, and a grid-connected structure. This structure mainly consists of three parts: wind power generation, a hybrid energy storage system composed of an energy type (battery) and a power type (flywheel), a control system, and an intelligent sensor system. The intelligent sensor system consists of: voltage sensor acquisition, power sensor acquisition, a wireless transmission module, etc. The collected signals are sent to the big data cloud through the 4G / 5G wireless transmission module, and after being processed by the data in the cloud, they are fed back to the controller to execute corresponding commands. The role of the control system is to control the charge and discharge state of the hybrid energy storage system in real time under the constraints of the wind power grid connection standard, so as to effectively smooth and suppress the wind power output power.

[0042] Power relationship of the hybrid energy storage system:

[0043] In the formula, is the active power generated by the wind farm / MW; is the grid-connected power; is the charge and discharge power of the lithium battery system in the hybrid energy storage system; is the charge and discharge power of the flywheel system in the hybrid energy storage system; Charge and discharge power of the hybrid energy storage system; when is positive, it means that the hybrid energy storage system is charging, is negative, indicating that the hybrid energy storage system is discharging.

[0044] Optionally, as an embodiment of the present invention, extracting the grid-connected power from the wind farm power generation based on the adaptive sliding average method specifically includes: Obtain the time series of the wind farm power generation, and set the initial sliding window size based on the frequency range of the wind farm power generation fluctuation and the desired smoothing degree; Initialize the weights of each time point data when calculating the average value, and each time point data has the same weight within the window; Calculate the moving average of the data at each time point according to the set sliding window size and weight. The moving averages of the data at all time points form the grid-connected power sequence. The moving average of the data at each time point is calculated as follows:

[0045] where i is the summation index, j is the time point, is the moving average at the j-th time point, is the data at the (j + i)-th time point, m is the summation boundary, N is the window size, and N = 2m + 1, is the weight of the data at the i-th time point.

[0046] Optionally, as an embodiment of the present invention, in step S1, obtaining the characteristics of the hybrid energy storage system specifically includes: In the battery energy storage system, calculate the charge and discharge efficiency of the battery, calculate the first-order difference of the battery SOC to obtain the change characteristics of the battery charge and discharge state, and calculate the mean and variance of the battery temperature to obtain the change characteristics of the battery temperature; In the flywheel energy storage system, calculate the change rate of the flywheel speed, and calculate the mean and variance of the change of the flywheel charge and discharge power to obtain the change characteristics of the flywheel charge and discharge power.

[0047] Optionally, as an embodiment of the present invention, the specific steps of step S3 include: Construct a data set based on the historical hybrid energy storage power, historical optimal EMD parameters, and historical hybrid energy storage system characteristics, and divide the data set into a training set, a test set, and a validation set; The input feature vector of the BiLSTM model is the historical hybrid energy storage power and historical hybrid energy storage system characteristics. Determine the dimension of the input feature vector and set the number of nodes in the input layer. The output feature vector of the BiLSTM model is the historical optimal EMD parameters. Determine the dimension of the output feature vector and set the number of nodes in the output layer; Initialize the weights and biases of the BiLSTM model; For each sample in the training set, input the input feature vector into the BiLSTM model, and calculate the predicted output vector of the model through forward propagation; Calculate the loss value between the predicted output vector and the actual output feature vector according to the loss function, and use the backpropagation algorithm to calculate the gradients of the loss function with respect to the model weights and biases, and update the model weights and biases using the gradient information; Calculate the evaluation index based on the validation set in combination with the root mean square error function every preset number of training times, and adjust the model structure based on the change trend of the evaluation index; Finally evaluate the trained BiLSTM model based on the test set.

[0048] Optionally, as an embodiment of the present invention, referring to Figure 4 , in the forward propagation calculation, specifically including: Performing a forward LSTM calculation on the input feature vector to obtain the output result of the forward layer, specifically:

[0049] Wherein, is the output result of the forward layer at time t, is the output result of the forward layer at time t-1, is the weight coefficient between the input layer and the forward layer, is the weight coefficient between the forward propagation unit at time t-1 and the forward propagation unit at time t, is the calculation function inside the neuron; Performing a backward LSTM calculation on the input feature vector to obtain the output result of the backward layer, specifically:

[0050] Wherein, is the output result of the backward layer at time t, is the output result of the backward layer at time t+1, is the weight coefficient between the input layer and the backward layer, is the weight coefficient between the forward propagation unit at time t+1 and the forward propagation unit at time t; Weightedly fusing the output results of the forward layer and the backward layer to obtain the output feature vector of the BiLSTM model;

[0051] Wherein, is the weight coefficient between the forward layer and the output layer, is the weight coefficient between the backward layer and the output layer, is the function for calculating the total result of the forward layer and the backward layer, is the output feature vector of the BiLSTM model.

[0052] Optionally, as an embodiment of the present invention, referring to Figure 5 , the EMD algorithm based on the current optimal EMD parameters decomposes the hybrid energy storage power into multiple MIF components, specifically including: S4-1, obtaining the hybrid energy storage power sequence , finding all the maximum points and minimum points of the hybrid energy storage power sequence, and respectively fitting the maximum envelope and the minimum envelope of the signal through a cubic spline function; S4-2, calculate the average value based on the maximum envelope and the minimum envelope ;

[0053] S4-3, subtract the average value from the hybrid energy storage power sequence to calculate the low-frequency signal sequence ;

[0054] S4-4, judge whether the number of zeros and the number of extreme points in the low-frequency signal sequence are equal or at most differ by one, judge whether the means of the maximum envelope and the minimum envelope in the low-frequency signal sequence are equal and are 0, and calculate as;

[0055] S4-5, if the above judgments on the low-frequency signal sequence are all negative, then replace the low-frequency signal sequence with the hybrid energy storage power sequence, and repeat steps S4-1, S4-2, S4-3, S4-4 until the above judgments on the low-frequency signal sequence are all positive to obtain the first IMF component;

[0056] S4-6, subtract the IMF component from the hybrid energy storage power sequence to obtain the residual signal sequence, replace the hybrid energy storage power sequence with the residual signal sequence, and repeat S4-1, S4-2, S4-3, S4-4 to obtain the second IMF component;

[0057] S4-7, repeat S4-6 based on the optimal decomposition level and the optimal iteration termination parameter to obtain multiple IMF components and an inseparable sequence.

[0058]

[0059] The expression of the EMD algorithm is: 。

[0060] Optionally, as an embodiment of the present invention, in step S5, The power of the flywheel energy storage system obtained by reconstructing the high-frequency part is calculated as:

[0061] The power of the battery energy storage system obtained by reconstructing the low-frequency part is calculated as:

[0062] Among them, is the power of the flywheel energy storage system, is the high-frequency reconstruction power, is the power of the battery energy storage system, is the low-frequency reconstruction power, is the high-low frequency demarcation point corresponding to the frequency threshold, is the k-th IMF component.

[0063] In some embodiments, the hybrid energy storage system power distribution system may include multiple functional modules composed of computer program segments. The computer programs of each program segment in the hybrid energy storage system power distribution system can be stored in the memory of the computer device and executed by at least one processor to execute (see details in Figure 1 description) the functions of hybrid energy storage system power distribution.

[0064] In this embodiment, according to the functions it performs, the hybrid energy storage system power distribution system can be divided into multiple functional modules, as Figure 6 shown. The functional modules of the system may include: a hybrid energy storage power acquisition module, a historical data acquisition module, a BiLSTM model training module, a hybrid energy storage power decomposition module, and a high-low frequency reconstruction module. The module referred to in the present invention means a series of computer program segments that can be executed by at least one processor and can complete fixed functions, and are stored in the memory. In this embodiment, the functions of each module will be described in detail in subsequent embodiments. The system includes: The hybrid energy storage power acquisition module collects the wind farm power generation and the characteristics of the hybrid energy storage system, extracts the grid-connected power from the wind farm power generation based on the adaptive sliding average method, and calculates the hybrid energy storage power based on the wind farm power generation and the grid-connected power; The historical data acquisition module obtains the historical hybrid energy storage power and the historical hybrid energy storage system characteristics based on S1, decomposes the historical hybrid energy storage power based on the EMD algorithm, and determines the historical optimal EMD parameters according to the historical hybrid energy storage system characteristics and the decomposition results; The BiLSTM model training module trains the BiLSTM model based on the historical hybrid energy storage power, the historical optimal EMD parameters, and the historical hybrid energy storage system characteristics; The hybrid energy storage power decomposition module obtains the current optimal EMD parameters based on the current hybrid energy storage power and the BiLSTM model, and decomposes the hybrid energy storage power into multiple MIF components based on the EMD algorithm of the current optimal EMD parameters; The high-low frequency reconstruction module divides multiple MIF components into a high-frequency part and a low-frequency part based on a preset frequency threshold, reconstructs the high-frequency part to obtain the flywheel energy storage system power, and reconstructs the low-frequency part to obtain the battery energy storage system power.

[0065] By collecting relevant data and performing a series of processes, including extracting the grid-connected power to calculate the hybrid energy storage power, using historical data to determine optimal parameters and train a model, then decomposing the power in combination with current data, and finally reconstructing by frequency division to obtain the power of different energy storage systems. This architecture comprehensively considers various factors and data, improves the accuracy and rationality of the power distribution of the hybrid energy storage system, optimizes the coordinated operation of the wind farm and the hybrid energy storage system, enhances the performance and efficiency of the energy storage system, and provides strong support for the effective management and control of the energy storage system.

[0066] Figure 7 FIG. 4 is a schematic structural diagram of a terminal provided in an embodiment of the present invention, and the terminal can be used to execute the method for power distribution of a hybrid energy storage system provided in the embodiment of the present invention.

[0067] Among them, the terminal may include: a processor, a memory, and a communication unit. These components communicate through one or more buses. Those skilled in the art can understand that the structure of the server shown in the figure does not constitute a limitation to the present invention. It can be a bus structure, a star structure, and may also include more or fewer components than shown in the figure, or combine some components, or have different component arrangements.

[0068] Among them, the memory can be used to store the execution instructions of the processor. The memory can be implemented by any type of volatile or non-volatile storage terminal or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk. When the execution instructions in the memory are executed by the processor, the terminal can execute some or all of the steps in the above method embodiments.

[0069] The processor is the control center of the storage terminal, connects various parts of the entire electronic terminal through various interfaces and lines, and executes various functions and / or processes data of the electronic terminal by running or executing software programs and / or modules stored in the memory, and calling data stored in the memory. The processor can be composed of an integrated circuit (IC). For example, it can be composed of a single packaged IC, or composed of multiple packaged ICs with the same or different functions connected together. For example, the processor may only include a central processing unit (CPU). In the embodiment of the present invention, the CPU can be a single operation core or include multiple operation cores.

[0070] A communication unit for establishing a communication channel so that the storage terminal can communicate with other terminals, receiving user data sent by other terminals or sending user data to other terminals.

[0071] The present invention also provides a computer-readable storage medium. The computer storage medium can store a program, and when the program is executed, it can include some or all of the steps in the embodiments provided by the present invention. The storage medium can be a magnetic disk, an optical disk, a read-only memory (ROM), a random access memory (RAM), etc.

[0072] Therefore, the technical effects achievable by this embodiment can be referred to the descriptions above and will not be elaborated here.

[0073] Those skilled in the art can clearly understand that the technologies in the embodiments of the present invention can be implemented by means of software plus a necessary general hardware platform. Based on such an understanding, the technical solutions in the embodiments of the present invention, in essence, or the part that contributes to the prior art can be embodied in the form of a software product. The computer software product is stored in a storage medium such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk, etc., which can store program codes, and includes several instructions to enable a computer terminal (which can be a personal computer, a server, or a second terminal, a network terminal, etc.) to execute all or part of the steps of the methods described in the embodiments of the present invention.

[0074] For the same or similar parts among the various embodiments in this specification, reference can be made to each other. In particular, for the terminal embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and the relevant parts can refer to the descriptions in the method embodiments.

[0075] In several embodiments provided by the present invention, it should be understood that the disclosed systems and methods can be implemented in other ways. For example, the system embodiments described above are only illustrative. For example, the division of the modules is only a logical function division. In actual implementation, there can be other division methods. For example, multiple modules or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed couplings or direct couplings or communication connections to each other can be through some interfaces. The indirect couplings or communication connections of the systems or modules can be in electrical, mechanical or other forms.

[0076] The module described as a separation component may or may not be physically separated. The component shown as a module may or may not be a physical module, that is, it may be located in one place or may be distributed across multiple network modules. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0077] In addition, in each embodiment of the present invention, each functional module may be integrated into one processing module, may exist separately as individual physical modules, or two or more modules may be integrated into one module.

[0078] Although the present invention has been described in detail by referring to the accompanying drawings and in combination with the preferred embodiments, the present invention is not limited thereto. Without departing from the spirit and essence of the present invention, those of ordinary skill in the art can make various equivalent modifications or substitutions to the embodiments of the present invention, and these modifications or substitutions should all be within the scope of the present invention. / Any person skilled in the art within the technical scope disclosed by the present invention can easily think of changes or substitutions, and all should be covered within the protection scope of the present invention.

Claims

1. A power distribution method for a hybrid energy storage system, characterized in that, The hybrid energy storage system includes a battery energy storage system and a flywheel energy storage system, and comprises the following steps: S1. Collect the wind farm power generation power and the characteristics of the hybrid energy storage system, extract the grid-connected power from the wind farm power generation power based on the adaptive sliding average method, and calculate the hybrid energy storage power based on the wind farm power generation power and the grid-connected power; S2. Based on the historical hybrid energy storage power and the historical hybrid energy storage system characteristics obtained in S1, decompose the historical hybrid energy storage power based on the EMD algorithm, and determine the historical optimal EMD parameters according to the historical hybrid energy storage system characteristics and the decomposition result; S3. Train the BiLSTM model based on the historical hybrid energy storage power, the historical optimal EMD parameters and the historical hybrid energy storage system characteristics; S4. Obtain the current optimal EMD parameters based on the current hybrid energy storage power and the BiLSTM model, and decompose the hybrid energy storage power into multiple MIF components based on the EMD algorithm with the current optimal EMD parameters; S5. Divide the multiple MIF components into a high-frequency part and a low-frequency part based on a preset frequency threshold, reconstruct the high-frequency part to obtain the flywheel energy storage system power, and reconstruct the low-frequency part to obtain the battery energy storage system power.

2. The power distribution method of the hybrid energy storage system according to claim 1, wherein In step S1, extracting the grid-connected power from the wind farm power generation power based on the adaptive sliding average method specifically includes: Obtain the time series of the wind farm power generation power, and set the initial sliding window size based on the frequency range of the wind farm power generation power fluctuation and the desired smoothing degree; Initialize the weights of each time point data when calculating the average value, and each time point data has the same weight within the window; Calculate the sliding average value of each time point data according to the set sliding window size and weight, and the sliding average values of all time point data form the grid-connected power sequence. The sliding average value of each time point data is calculated as follows: where i is the summation index, j is the time point, is the moving average at the j-th time point, is the data at the (j + i)-th time point, m is the summation boundary, N is the window size, N = 2m + 1, is the weight of the data at the i-th time point.

3. The power distribution method of the hybrid energy storage system according to claim 1, characterized in that The specific steps of step S3 include: Construct a data set based on the historical hybrid energy storage power, the historical optimal EMD parameters and the historical hybrid energy storage system characteristics, and divide the data set into a training set, a test set and a validation set; The input feature vector of the BiLSTM model is the historical hybrid energy storage power and the historical hybrid energy storage system characteristics. Determine the dimension of the input feature vector and set the number of input layer nodes. The output feature vector of the BiLSTM model is the historical optimal EMD parameters. Determine the dimension of the output feature vector and set the number of output layer nodes; Initialize the weights and biases of the BiLSTM model; For each sample in the training set, input the input feature vector into the BiLSTM model, and calculate the predicted output vector of the model through forward propagation; Calculate the loss value between the predicted output vector and the actual output feature vector according to the loss function, and use the backpropagation algorithm to calculate the gradients of the loss function with respect to the model weights and biases, and update the model weights and biases using the gradient information; Calculate the evaluation index based on the validation set in combination with the root mean square error function every preset number of training times, and adjust the model structure based on the change trend of the evaluation index; Perform a final evaluation on the trained BiLSTM model based on the test set.

4. The power distribution method of the hybrid energy storage system according to claim 3, characterized in that In the forward propagation calculation, it specifically includes: Perform forward LSTM calculation on the input feature vector to obtain the output result of the forward layer, specifically: Among them, is the output result of the forward layer at time t, is the output result of the forward layer at time t-1, is the weight coefficient between the input layer and the forward layer, is the weight coefficient between the forward propagation unit at time t-1 and the forward propagation unit at time t, is the calculation function inside the neuron; Perform backward LSTM calculation on the input feature vector to obtain the output result of the backward layer, specifically: Among them, is the output result of the backward layer at time t, is the output result of the backward layer at time t+1, is the weight coefficient between the input layer and the backward layer, is the weight coefficient between the forward propagation unit at time t+1 and the forward propagation unit at time t; Weightedly fuse the output results of the forward layer and the backward layer to obtain the output feature vector of the BiLSTM model; Among them, is the weight coefficient between the forward layer and the output layer, is the weight coefficient between the backward layer and the output layer, is a function for calculating the total result of the forward layer and the backward layer, is the output feature vector of the BiLSTM model.

5. The power distribution method of the hybrid energy storage system according to claim 3, wherein The EMD algorithm based on the current optimal EMD parameters decomposes the hybrid energy storage power into multiple MIF components, specifically including: S4-1: Obtain the hybrid energy storage power sequence, find all the maximum and minimum points of the hybrid energy storage power sequence, and respectively fit the maximum envelope line and the minimum envelope line of the signal through a cubic spline function; S4-2: Calculate the average value based on the maximum envelope line and the minimum envelope line; S4-3: Subtract the average value from the hybrid energy storage power sequence to calculate the low-frequency removed signal sequence; S4-4: Judge whether the number of zero points and the number of extreme points in the low-frequency removed signal sequence are equal or at most differ by one, and judge whether the means of the maximum envelope line and the minimum envelope line in the low-frequency removed signal sequence are equal and are 0; S4-5: If the above judgments on the low-frequency removed signal sequence are all negative, then replace the hybrid energy storage power sequence with the low-frequency removed signal sequence, and repeat steps S4-1, S4-2, S4-3, S4-4 until the above judgments on the low-frequency removed signal sequence are all positive to obtain the first IMF component; S4-6: Subtract the IMF component from the hybrid energy storage power sequence to obtain the remaining signal sequence, replace the hybrid energy storage power sequence with the remaining signal sequence, and repeat the execution of S4-1, S4-2, S4-3, S4-4 to obtain the second IMF component; S4-7: Repeat S4-6 based on the optimal decomposition layer number and the optimal iteration termination parameter to obtain multiple IMF components and an inseparable sequence.

6. The power distribution method of the hybrid energy storage system according to claim 1, wherein In step S5, The power of the flywheel energy storage system obtained by reconstructing the high-frequency part is calculated as: The power of the battery energy storage system obtained by reconstructing the low-frequency part is calculated as: Among them, is the power of the flywheel energy storage system, is the high-frequency reconstruction power, is the power of the battery energy storage system, is the low-frequency reconstruction power, is the high-low frequency demarcation point corresponding to the frequency threshold, is the k-th IMF component.

7. The power distribution method of the hybrid energy storage system according to claim 1, wherein In step S1, obtaining the characteristics of the hybrid energy storage system specifically includes: In the battery energy storage system, calculate the charge and discharge efficiency of the battery, calculate the first-order difference of the battery SOC to obtain the change characteristics of the battery charge and discharge state, and calculate the mean value and variance of the battery temperature to obtain the change characteristics of the TV temperature; In the flywheel energy storage system, calculate the flywheel speed change rate, and calculate the mean value and variance of the flywheel charge and discharge power change to obtain the flywheel charge and discharge power change characteristics.

8. A power distribution method for a hybrid energy storage system, characterized in that When the system is implemented, execute the hybrid energy storage system power distribution method described in any one of claims 1-7. The hybrid energy storage system includes a battery energy storage system and a flywheel energy storage system. The system includes: A hybrid energy storage power acquisition module, which collects the wind farm power generation power and the characteristics of the hybrid energy storage system, extracts the grid-connected power from the wind farm power generation power based on the adaptive moving average method, and calculates the hybrid energy storage power based on the wind farm power generation power and the grid-connected power; A historical data acquisition module, which obtains the historical hybrid energy storage power and the historical hybrid energy storage system characteristics based on S1, decomposes the historical hybrid energy storage power based on the EMD algorithm, and determines the historical optimal EMD parameters according to the historical hybrid energy storage system characteristics and the decomposition results; The BiLSTM model training module trains a BiLSTM model based on historical hybrid energy storage power, historical optimal EMD parameters, and historical hybrid energy storage system characteristics; The hybrid energy storage power decomposition module obtains current optimal EMD parameters based on the current hybrid energy storage power and the BiLSTM model, and decomposes the hybrid energy storage power into multiple MIF components by the EMD algorithm based on the current optimal EMD parameters; The high- and low-frequency reconstruction module divides multiple MIF components into a high-frequency part and a low-frequency part based on a preset frequency threshold, reconstructs the high-frequency part to obtain the flywheel energy storage system power, and reconstructs the low-frequency part to obtain the battery energy storage system power.

9. A terminal, characterized in that, Comprising: A memory for storing a hybrid energy storage system power distribution program; A processor for implementing the steps of the hybrid energy storage system power distribution method according to any one of claims 1-7 when executing the hybrid energy storage system power distribution program.

10. A computer-readable storage medium, characterized in that, The hybrid energy storage system power distribution program is stored on the readable storage medium, and when the hybrid energy storage system power distribution program is executed by the processor, the steps of the hybrid energy storage system power distribution method according to any one of claims 1-7 are implemented.

Images