Method and device for determining capacity distribution accuracy of super-capacity energy storage system

Abstract

The embodiment of the invention provides a capacity distribution accuracy determination method and device for a super-capacity energy storage system, and relates to the technical field of power grid frequency modulation, and the method comprises the steps: obtaining a frequency modulation instruction and a decomposition layer number adopted by capacity distribution; repeating for multiple times, correcting the penalty factor, and ending the circulation until a second prediction error corresponding to the second prediction decomposition purity is smaller than a second prediction error obtained in the previous circulation; and determining that the second prediction decomposition purity obtained in the last cycle represents the capacity distribution accuracy of the super-capacity energy storage system, wherein the smaller the second prediction decomposition purity is, the higher the capacity distribution accuracy is. And an efficient and accurate evaluation and determination method is provided for the capacity distribution accuracy of the energy storage equipment in a super-capacity energy storage system.

Description

Technical Field

[0001] The present disclosure relates to the technical field of power grid frequency regulation, and in particular to a method and device for determining the capacity allocation accuracy of an ultra-capacity energy storage system. Background Art

[0002] The existing frequency regulation of thermal power and hydropower units has many limitations, and the installation of energy storage solutions in thermal power plants is highly favored. At present, in major regional power grids in my country, large hydropower and thermal power units are mainly used as power grid frequency regulation power sources, and the frequency regulation power output is adjusted to respond to system frequency changes, but the frequency regulation of hydropower and thermal power units has certain limitations. Energy storage frequency regulation is becoming a new auxiliary means of frequency regulation with its fast and accurate power response capability, but limited by the nominal cycle life of lithium batteries of about 5,000 times, most of the existing energy storage frequency regulation projects only respond to AGC (Automatic Generation Control) secondary frequency regulation, but not primary frequency regulation. Supercapacitors rely on their high power and long life characteristics to adapt to the short-term high frequency and high power frequency regulation needs of the power grid, and can make up for the shortcomings of ultra-short-term and short-term energy storage. However, there is currently no efficient and accurate solution to determine the accuracy of capacity allocation involved in supercapacitor energy storage systems. Summary of the invention

[0003] The embodiments of the present disclosure provide a method and device for determining the accuracy of capacity allocation in an overcapacity energy storage system, so as to solve the problem that there is no solution for evaluating and determining the accuracy of capacity allocation in an overcapacity energy storage system.

[0004] Based on the above problems, in a first aspect, a method for determining the capacity allocation accuracy of an ultra-capacity energy storage system is provided, comprising: Obtain frequency modulation instructions and decomposition layers used for capacity allocation; The method repeats the loop multiple times, and in each loop, the frequency modulation instruction and the number of decomposition layers are input into a pre-trained prediction model to obtain a first prediction decomposition purity and a corresponding first prediction error and store them in historical data; the frequency modulation instruction and the number of decomposition layers are input into a pre-trained prediction model and a penalty factor of the model is set to obtain a second prediction decomposition purity and a corresponding second prediction error and store them in historical data; based on the historical data, a first average error coefficient corresponding to the first prediction error and a second average error coefficient corresponding to the second prediction error are determined; based on the first average error coefficient and the second average error coefficient, the penalty factor is corrected; and the loop ends when the second prediction error corresponding to the second prediction decomposition purity is less than the second prediction error obtained in the previous loop; Determine that the second prediction decomposition purity obtained in the last cycle represents the capacity allocation accuracy of the super-capacity energy storage system, and the smaller the second prediction decomposition purity is, the higher the capacity allocation accuracy is.

[0005] In combination with the first aspect, in a possible implementation, the pre-trained prediction model takes the corresponding frequency modulation instructions and decomposition layer number in the historical data as input, predicts the decomposition purity as output, and trains the model based on the true value of the decomposition purity.

[0006] In combination with the first aspect, in a possible implementation manner, inputting the frequency modulation instruction and the number of decomposition layers into a pre-trained prediction model to obtain a first prediction decomposition purity and a corresponding first prediction error includes: Inputting the frequency modulation instruction and the number of decomposition layers into a pre-trained prediction model to obtain a first prediction decomposition purity; The actual decomposition purity is calculated based on the frequency modulation instruction and the number of decomposition layers; The first prediction error is obtained by subtracting the first predicted decomposition purity from the actual decomposition purity.

[0007] In combination with the first aspect, in a possible implementation manner, the actual decomposition purity is calculated by the following formula: , in, Characterizes the actual decomposition purity, Characterizes the number of decomposition levels, Characterize the decomposition The subsequences are mapped to the peak frequencies in the frequency domain, Characterize the linear independence analysis function in the time domain, Characterize the decomposition subsequences.

[0008] In combination with the first aspect, in a possible implementation manner, determining, based on historical data, a first average error coefficient corresponding to the first prediction error and a second average error coefficient corresponding to the second prediction error includes: Averaging the first prediction errors stored in the historical data for multiple cycles to obtain the first average error coefficient; The second prediction error stored in the historical data in multiple cycles is averaged to obtain the second average error coefficient.

[0009] In combination with the first aspect, in a possible implementation manner, the correcting the penalty factor based on the first average error coefficient and the second average error coefficient includes: If the first average error coefficient is greater than the second average error coefficient, the penalty factor α is modified; otherwise, the penalty factor is directly used as the modified penalty factor; The revised formula is as follows: , in, Characterizes the punishment factor, Characterizes the first average error coefficient, Characterizes the second mean error coefficient, Characterize the number of corrections, Characterizes the natural constants, Characterizing the PADE approximation.

[0010] In a second aspect, a device for determining the capacity allocation accuracy of an ultra-capacity energy storage system is provided, comprising: A data acquisition module, used to obtain frequency modulation instructions and decomposition layers used for capacity allocation; A loop correction module, for looping multiple times, each loop inputting the frequency modulation instruction and the number of decomposition layers into a pre-trained prediction model, obtaining a first prediction decomposition purity and a corresponding first prediction error and storing them in historical data; inputting the frequency modulation instruction and the number of decomposition layers into a pre-trained prediction model and setting a penalty factor of the model, obtaining a second prediction decomposition purity and a corresponding second prediction error and storing them in historical data; determining a first average error coefficient corresponding to the first prediction error and a second average error coefficient corresponding to the second prediction error based on historical data; correcting the penalty factor based on the first average error coefficient and the second average error coefficient; until the second prediction error corresponding to the second prediction decomposition purity is less than the second prediction error obtained in the previous loop, the loop is terminated; The accuracy determination module is used to determine that the second predicted decomposition purity obtained in the last cycle represents the accuracy of capacity allocation of the super-capacity energy storage system, and the smaller the second predicted decomposition purity is, the higher the capacity allocation accuracy is.

[0011] In combination with the second aspect, in a possible implementation, in the loop correction module, the pre-trained prediction model takes the corresponding frequency modulation instructions and decomposition layer number in the historical data as input, predicts the decomposition purity as output, and trains the model based on the true value of the decomposition purity.

[0012] In combination with the second aspect, in a possible implementation manner, in the cyclic correction module, inputting the frequency modulation instruction and the number of decomposition layers into a pre-trained prediction model to obtain a first prediction decomposition purity and a corresponding first prediction error includes: Inputting the frequency modulation instruction and the number of decomposition layers into a pre-trained prediction model to obtain a first prediction decomposition purity; The actual decomposition purity is calculated based on the frequency modulation instruction and the number of decomposition layers; The first prediction error is obtained by subtracting the first predicted decomposition purity from the actual decomposition purity.

[0013] In conjunction with the second aspect, in a possible implementation manner, in the circulation correction module, the actual decomposition purity is calculated by the following formula: , in, Characterizes the actual decomposition purity, Characterizes the number of decomposition levels, Characterize the decomposition The subsequences are mapped to the peak frequencies in the frequency domain, Characterize the linear independence analysis function in the time domain, Characterize the decomposition subsequences.

[0014] In combination with the second aspect, in a possible implementation manner, in the cyclic correction module, determining a first average error coefficient corresponding to the first prediction error and a second average error coefficient corresponding to the second prediction error based on historical data includes: Averaging the first prediction errors stored in the historical data for multiple cycles to obtain the first average error coefficient; The second prediction error stored in the historical data in multiple cycles is averaged to obtain the second average error coefficient.

[0015] In combination with the second aspect, in a possible implementation manner, in the cyclic correction module, the correction of the penalty factor based on the first average error coefficient and the second average error coefficient includes: If the first average error coefficient is greater than the second average error coefficient, the penalty factor α is modified; otherwise, the penalty factor is directly used as the modified penalty factor; The revised formula is as follows: , in, Characterizes the punishment factor, Characterizes the first average error coefficient, Characterizes the second mean error coefficient, Characterize the number of corrections, Characterizes the natural constants, Characterizing the PADE approximation.

[0016] In a third aspect, a computer device is provided, comprising: a processor, a memory and a bus, the memory storing machine-readable instructions executable by the processor, and when the computer device is running, the processor communicates with the memory via the bus, and when the machine-readable instructions are executed by the processor, the steps of a method for determining the capacity allocation accuracy of an ultra-capacity energy storage system as described in the first aspect or in combination with any possible implementation manner of the first aspect are performed.

[0017] In a fourth aspect, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps of a method for determining the capacity allocation accuracy of an ultra-capacity energy storage system as described in the first aspect or in combination with any possible implementation manner of the first aspect are executed.

[0018] The beneficial effects of the embodiments of the present disclosure include: The disclosed embodiment provides a method and device for determining the accuracy of capacity allocation of an overcapacity energy storage system, which is applied to the overcapacity energy storage system to evaluate the allocation result of the energy storage capacity and determine the accuracy of the capacity allocation. In the disclosed embodiment, a prediction model is used to predict the decomposition purity of the frequency modulation instruction before allocation, and the punishment factor is introduced and the first average error coefficient and the second average error coefficient determined by the first prediction error and the second prediction error in the historical data are continuously corrected to correct the prediction result, so that the prediction of the decomposition purity is more accurate, and the determination of the accuracy of the capacity allocation is also more accurate. An efficient and accurate evaluation and determination method is provided for the accuracy of the allocation of the capacity of the energy storage device in the overcapacity energy storage system. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 A flow chart of a method for determining the capacity allocation accuracy of a super-capacity energy storage system provided in an embodiment of the present disclosure; Figure 2 A structural diagram of a device for determining the accuracy of capacity allocation of a super-capacity energy storage system provided in an embodiment of the present disclosure; Figure 3 A flow chart of a method for determining the capacity allocation accuracy of a super-capacity energy storage system provided in an embodiment of the present disclosure. DETAILED DESCRIPTION

[0020] The embodiments of the present disclosure provide a method and device for determining the accuracy of capacity allocation of an ultra-capacity energy storage system. The preferred embodiments of the present disclosure are described below in conjunction with the drawings of the specification. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present disclosure, and are not used to limit the present disclosure. In addition, the embodiments and features in the embodiments of the present application can be combined with each other if there is no conflict.

[0021] The present disclosure provides a method for determining the capacity allocation accuracy of an ultra-capacity energy storage system. Figure 1 As shown, including: S101, obtaining frequency modulation instructions and decomposition layer number used for capacity allocation; S102, looping for multiple times, in each loop, inputting the frequency modulation instruction and the number of decomposition layers into the pre-trained prediction model, obtaining the first prediction decomposition purity and the corresponding first prediction error and storing them in the historical data; inputting the frequency modulation instruction and the number of decomposition layers into the pre-trained prediction model and setting the penalty factor of the model, obtaining the second prediction decomposition purity and the corresponding second prediction error and storing them in the historical data; determining the first average error coefficient corresponding to the first prediction error and the second average error coefficient corresponding to the second prediction error based on the historical data; correcting the penalty factor based on the first average error coefficient and the second average error coefficient; until the second prediction error corresponding to the second prediction decomposition purity is less than the second prediction error obtained in the previous loop, then the loop ends; S103, determining that the second predicted decomposition purity obtained in the last cycle represents the capacity allocation accuracy of the super-capacity energy storage system, and the smaller the second predicted decomposition purity is, the higher the capacity allocation accuracy is.

[0022] In the embodiment of the present disclosure, when allocating the energy storage capacity of the super-capacity energy storage system, the frequency modulation instruction is first decomposed by the VMD algorithm to obtain ,in, is the total power that needs to be compensated for the frequency modulation instruction, is the i-th modal component (IMF) after decomposition, and the number of modal quantities is the decomposition level K1. is the residual component after decomposition; the high-frequency power component is obtained and low frequency power components After that, the high-frequency power component and the low-frequency power component are sent to the corresponding energy storage device, and the modal components generated by the decomposition will be aliased, affecting the accuracy of capacity allocation. The decomposition purity is the standard for defining whether each modal component is "independent" after VMD decomposition. The smaller the decomposition purity, the more independent each modal component, and the less aliasing between the modal components, which means that the capacity allocation accuracy is higher. Therefore, by continuously correcting the penalty factor of the model, the accuracy of the decomposition purity prediction is continuously improved, so that the accuracy of the energy storage system capacity allocation can be determined more accurately.

[0023] In the disclosed embodiment, the first prediction error without adding the penalty factor and the second prediction error with adding the penalty factor stored in the historical data in each cycle are used to obtain the corresponding first average error coefficient and second average error coefficient to judge the correction effect of the added penalty factor on the model prediction result, and the penalty factor is continuously corrected and adjusted through multiple cycles, so that the prediction result of the model, that is, the second prediction decomposition purity is constantly close to the actual decomposition accuracy, so that the prediction effect of the model is better. Among them, the penalty factor is a common parameter of VMD, and the value can be set manually.

[0024] In another embodiment provided by the present disclosure, a pre-trained prediction model takes the corresponding frequency modulation instructions and decomposition layer numbers in the historical data as input, predicts the decomposition purity as output, and trains the model based on the true value of the decomposition purity.

[0025] In the disclosed embodiment, the above-mentioned pre-trained prediction model uses the corresponding frequency modulation instructions and decomposition layer number in the historical data as input during the training process, predicts the decomposition purity as output and uses the actual decomposition purity as the true value.

[0026] In another embodiment provided by the present disclosure, the frequency modulation instruction and the number of decomposition layers are input into a pre-trained prediction model to obtain a first prediction decomposition purity and a corresponding first prediction error, including: Input the frequency modulation instruction and the number of decomposition layers into a pre-trained prediction model to obtain a first prediction decomposition purity; The actual decomposition purity is calculated based on the frequency modulation instruction and the number of decomposition layers; The first prediction error is obtained by subtracting the first predicted decomposition purity from the actual decomposition purity.

[0027] In the disclosed embodiment, there will be an error between the first predicted decomposition purity obtained by the prediction model and the actual decomposition purity. The actual decomposition purity corresponding to the frequency modulation instruction and the decomposition layer number can be obtained by calculation, and the error of the prediction result can be obtained by subtracting the predicted decomposition purity from the actual decomposition purity, which is convenient for subsequent use of the error to correct the penalty factor added to the model, and then correct the prediction result.

[0028] In another embodiment provided by the present disclosure, the actual decomposition purity is calculated by the following formula: , in, Characterizes the actual decomposition purity, Characterizes the number of decomposition levels, Characterize the decomposition The subsequences are mapped to the peak frequencies in the frequency domain, Characterize the linear independence analysis function in the time domain, The subsequence generated by the characterization decomposition.

[0029] In another embodiment provided by the present disclosure, determining a first average error coefficient corresponding to a first prediction error and a second average error coefficient corresponding to a second prediction error based on historical data includes: Taking the average of the first prediction errors stored in the historical data in multiple cycles to obtain the first average error coefficient; Taking the average of the second prediction errors stored in the historical data in multiple cycles to obtain the second average error coefficient.

[0030] In the embodiment of the present disclosure, the first prediction error and the second prediction error stored in the historical data in each cycle before the current cycle are respectively averaged to obtain the first average error coefficient and the second average error coefficient. By comparing between the first average error coefficient and the second average error coefficient, it can effectively characterize the correction effect on the model prediction result while continuously correcting the punishment factor.

[0031] In another embodiment provided by the present disclosure, correcting the punishment factor based on the first average error coefficient and the second average error coefficient; including: If the first average error coefficient is greater than the second average error coefficient, correct the punishment factor α, otherwise directly use the punishment factor as the corrected punishment factor; The correction formula is as follows: , where, represents the punishment factor, represents the first average error coefficient, represents the second average error coefficient, represents the number of correction times, represents the natural constant, represents the Padé approximation.

[0032] In the embodiment of the present disclosure, if W1 < W2, it means that the improvement effect on the prediction accuracy of the model after adding the punishment factor is not good, and then the punishment factor needs to be corrected.

[0033] Based on the same inventive concept, the embodiment of the present disclosure also provides an apparatus for determining the capacity allocation accuracy rate of an ultra-capacitor energy storage system. Since the principle of the problem solved by this apparatus is similar to that of the foregoing method for determining the capacity allocation accuracy rate of an ultra-capacitor energy storage system, the implementation of this apparatus can refer to the implementation of the foregoing method, and the repeated parts will not be elaborated.

[0034] For ease of understanding, the specific process of the present disclosure solution is as Figure 2As shown, the decomposition purity is predicted by the prediction model, and the penalty factor is added in the process to continuously explore and correct the penalty factor to further improve the accuracy of the model prediction results. Finally, the capacity allocation accuracy of the system is determined by the final predicted decomposition purity, that is, the second predicted decomposition purity evaluation.

[0035] With the above Figure 1 Corresponding to the method shown, the embodiment of the present disclosure also provides a device for determining the capacity allocation accuracy of a super-capacity energy storage system, such as Figure 3 As shown, including: The data acquisition module 301 is used to acquire the frequency modulation instruction and the number of decomposition layers used for capacity allocation; The loop correction module 302 is used to loop multiple times, and in each loop, the frequency modulation instruction and the number of decomposition layers are input into the pre-trained prediction model to obtain the first prediction decomposition purity and the corresponding first prediction error and store them in the historical data; the frequency modulation instruction and the number of decomposition layers are input into the pre-trained prediction model and the penalty factor of the model is set to obtain the second prediction decomposition purity and the corresponding second prediction error and store them in the historical data; based on the historical data, a first average error coefficient corresponding to the first prediction error and a second average error coefficient corresponding to the second prediction error are determined; based on the first average error coefficient and the second average error coefficient, the penalty factor is corrected; until the second prediction error corresponding to the second prediction decomposition purity is less than the second prediction error obtained in the previous loop, the loop is terminated; The accuracy determination module 303 is used to determine the second predicted decomposition purity obtained in the last cycle to characterize the accuracy of capacity allocation of the super-capacity energy storage system. The smaller the second predicted decomposition purity, the higher the capacity allocation accuracy.

[0036] In another embodiment provided by the present disclosure, in the above-mentioned loop correction module 302, the pre-trained prediction model takes the corresponding frequency modulation instructions and decomposition layer numbers in the historical data as input, predicts the decomposition purity as output, and trains the model based on the true value of the decomposition purity.

[0037] In another embodiment provided by the present disclosure, in the above-mentioned cycle correction module 302, the frequency modulation instruction and the number of decomposition layers are input into a pre-trained prediction model to obtain a first prediction decomposition purity and a corresponding first prediction error, including: Input the frequency modulation instruction and the number of decomposition layers into a pre-trained prediction model to obtain a first prediction decomposition purity; The actual decomposition purity is calculated based on the frequency modulation instruction and the number of decomposition layers; The first prediction error is obtained by subtracting the first predicted decomposition purity from the actual decomposition purity.

[0038] In another embodiment provided by the present disclosure, in the above-mentioned cycle correction module 302, the actual decomposition purity is calculated by the following formula: , in, Characterizes the actual decomposition purity, Characterizes the number of decomposition levels, Characterize the decomposition The subsequences are mapped to the peak frequencies in the frequency domain, Characterize the linear independence analysis function in the time domain, Characterize the decomposition subsequences.

[0039] In another embodiment provided by the present disclosure, in the above-mentioned cyclic correction module 302, determining a first average error coefficient corresponding to the first prediction error and a second average error coefficient corresponding to the second prediction error based on historical data includes: The first prediction errors stored in the historical data for multiple cycles are averaged to obtain a first average error coefficient; The second prediction errors stored in the historical data for multiple cycles are averaged to obtain a second average error coefficient.

[0040] In another embodiment provided by the present disclosure, in the above-mentioned cycle correction module 302, the penalty factor is corrected based on the first average error coefficient and the second average error coefficient; including: If the first average error coefficient is greater than the second average error coefficient, the penalty factor α is modified; otherwise, the penalty factor is directly used as the modified penalty factor; The revised formula is as follows: , in, Characterizes the punishment factor, Characterizes the first average error coefficient, Characterizes the second mean error coefficient, Characterize the number of corrections, Characterizes the natural constants, Characterizing the PADE approximation.

[0041] An embodiment of the present disclosure provides a computer device, including: a processor, a memory and a bus, wherein the memory stores machine-readable instructions executable by the processor, and when the computer device is running, the processor communicates with the memory via the bus, and when the machine-readable instructions are executed by the processor, the steps of the method for determining the capacity allocation accuracy of the ultra-capacity energy storage system provided in any embodiment of the present disclosure are performed.

[0042] The computer device provided by the embodiment of the present disclosure includes a processor, a memory, and a bus. The memory is used to store execution instructions, including internal memory and external memory; the internal memory here is also called internal memory, which is used to temporarily store the operation data in the processor and the data exchanged with the external memory such as the hard disk. The processor exchanges data with the external memory through the internal memory. When the electronic device is running, the processor and the memory communicate through the bus, so that the processor executes the following instructions: Obtain frequency modulation instructions and decomposition layers used for capacity allocation; The method repeats the loop multiple times, and in each loop, the frequency modulation instruction and the number of decomposition layers are input into a pre-trained prediction model to obtain a first prediction decomposition purity and a corresponding first prediction error and store them in historical data; the frequency modulation instruction and the number of decomposition layers are input into a pre-trained prediction model and a penalty factor of the model is set to obtain a second prediction decomposition purity and a corresponding second prediction error and store them in historical data; based on the historical data, a first average error coefficient corresponding to the first prediction error and a second average error coefficient corresponding to the second prediction error are determined; based on the first average error coefficient and the second average error coefficient, the penalty factor is corrected; and the loop ends when the second prediction error corresponding to the second prediction decomposition purity is less than the second prediction error obtained in the previous loop; Determine that the second prediction decomposition purity obtained in the last cycle represents the capacity allocation accuracy of the super-capacity energy storage system, and the smaller the second prediction decomposition purity is, the higher the capacity allocation accuracy is.

[0043] The embodiment of the present disclosure provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the steps of the method for determining the capacity allocation accuracy of the ultra-capacity energy storage system provided in any embodiment of the present disclosure are executed. The storage medium may be a volatile or non-volatile computer-readable storage medium.

[0044] Through the description of the above implementation methods, those skilled in the art can clearly understand that the embodiments of the present disclosure can be implemented by hardware, or by means of software plus a necessary general hardware platform. Based on such an understanding, the technical solution of the embodiments of the present disclosure can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.), including a number of instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment of the present disclosure.

[0045] Those skilled in the art will appreciate that the accompanying drawings are merely schematic diagrams of a preferred embodiment, and the modules or processes in the accompanying drawings are not necessarily required for implementing the present disclosure.

[0046] Those skilled in the art can understand that the modules in the device in the embodiment can be distributed in the device in the embodiment according to the description of the embodiment, or can be changed accordingly and located in one or more devices different from the present embodiment. The modules in the above embodiment can be combined into one module, or can be further divided into multiple sub-modules.

[0047] The serial numbers of the above-mentioned embodiments of the present disclosure are only for description and do not represent the advantages or disadvantages of the embodiments.

[0048] Obviously, those skilled in the art can make various changes and modifications to the present disclosure without departing from the spirit and scope of the present disclosure. Thus, if these modifications and variations of the present disclosure fall within the scope of the claims of the present disclosure and their equivalents, the present disclosure is also intended to include these modifications and variations.

Claims

1. A method for determining the capacity allocation accuracy of an ultra-capacity energy storage system, characterized in that: include: Obtain frequency modulation instructions and decomposition layers used for capacity allocation; The method repeats the loop multiple times, and in each loop, the frequency modulation instruction and the number of decomposition layers are input into a pre-trained prediction model to obtain a first prediction decomposition purity and a corresponding first prediction error and store them in historical data; the frequency modulation instruction and the number of decomposition layers are input into a pre-trained prediction model and a penalty factor of the model is set to obtain a second prediction decomposition purity and a corresponding second prediction error and store them in historical data; based on the historical data, a first average error coefficient corresponding to the first prediction error and a second average error coefficient corresponding to the second prediction error are determined; based on the first average error coefficient and the second average error coefficient, the penalty factor is corrected; and the loop ends when the second prediction error corresponding to the second prediction decomposition purity is less than the second prediction error obtained in the previous loop; Determine that the second prediction decomposition purity obtained in the last cycle represents the capacity allocation accuracy of the super-capacity energy storage system, and the smaller the second prediction decomposition purity is, the higher the capacity allocation accuracy is.

2. The method according to claim 1, characterized in that The pre-trained prediction model takes the corresponding frequency modulation instructions and decomposition layer number in the historical data as input, predicts the decomposition purity as output, and trains the model based on the true value of the decomposition purity.

3. The method according to claim 1, characterized in that The step of inputting the frequency modulation instruction and the number of decomposition layers into a pre-trained prediction model to obtain a first prediction decomposition purity and a corresponding first prediction error includes: Inputting the frequency modulation instruction and the number of decomposition layers into a pre-trained prediction model to obtain a first prediction decomposition purity; The actual decomposition purity is calculated based on the frequency modulation instruction and the number of decomposition layers; The first prediction error is obtained by subtracting the first predicted decomposition purity from the actual decomposition purity.

4. The method according to claim 3, characterized in that The actual decomposition purity is calculated by the following formula: ， in, Characterizes the actual decomposition purity, Characterizes the number of decomposition levels, Characterize the decomposition The subsequences are mapped to the peak frequencies in the frequency domain, Characterize the linear independence analysis function in the time domain, Characterize the decomposition subsequences.

5. The method according to claim 1, characterized in that The determining, based on historical data, a first average error coefficient corresponding to the first prediction error and a second average error coefficient corresponding to the second prediction error includes: Averaging the first prediction errors stored in the historical data for multiple cycles to obtain the first average error coefficient; The second prediction error stored in the historical data in multiple cycles is averaged to obtain the second average error coefficient.

6. The method according to claim 1, characterized in that The step of modifying the penalty factor based on the first average error coefficient and the second average error coefficient comprises: If the first average error coefficient is greater than the second average error coefficient, the penalty factor α is modified; otherwise, the penalty factor is directly used as the modified penalty factor; The revised formula is as follows: ， in, Characterizes the punishment factor, Characterizes the first mean error coefficient, Characterizes the second mean error coefficient, Characterize the number of corrections, Characterizes the natural constants, Characterizing the PADE approximation.

7. A device for determining the capacity allocation accuracy of an ultra-capacity energy storage system, characterized in that: include: A data acquisition module, used to obtain frequency modulation instructions and decomposition layers used for capacity allocation; A loop correction module, for looping multiple times, each loop inputting the frequency modulation instruction and the number of decomposition layers into a pre-trained prediction model, obtaining a first prediction decomposition purity and a corresponding first prediction error and storing them in historical data; inputting the frequency modulation instruction and the number of decomposition layers into a pre-trained prediction model and setting a penalty factor of the model, obtaining a second prediction decomposition purity and a corresponding second prediction error and storing them in historical data; determining a first average error coefficient corresponding to the first prediction error and a second average error coefficient corresponding to the second prediction error based on historical data; correcting the penalty factor based on the first average error coefficient and the second average error coefficient; until the second prediction error corresponding to the second prediction decomposition purity is less than the second prediction error obtained in the previous loop, the loop is terminated; The accuracy determination module is used to determine that the second predicted decomposition purity obtained in the last cycle represents the accuracy of capacity allocation of the super-capacity energy storage system, and the smaller the second predicted decomposition purity is, the higher the capacity allocation accuracy is.

8. The device according to claim 7, characterized in that In the cycle correction module, the pre-trained prediction model takes the corresponding frequency modulation instructions and decomposition layer number in the historical data as input, predicts the decomposition purity as output, and trains the model based on the true value of the decomposition purity.

9. A computer device, characterized in that: include: A processor, a memory and a bus, wherein the memory stores machine-readable instructions executable by the processor, and when the computer device is running, the processor and the memory communicate via the bus, and when the machine-readable instructions are executed by the processor, the steps of the method for determining the capacity allocation accuracy of the ultra-capacity energy storage system according to any one of claims 1 to 6 are performed.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, which, when executed by a processor, executes the steps of the method for determining the capacity allocation accuracy of the ultra-capacity energy storage system according to any one of claims 1 to 6.

Images