A prediction method and system for energy storage frequency modulation commands based on a two-stage method

Abstract

The present invention relates to the field of energy storage frequency modulation prediction, and discloses a method and system for predicting energy storage frequency modulation commands based on a two-stage method. The method includes: obtaining energy storage frequency modulation commands and performing variational mode decomposition; judging whether subsequences attract or repel each other according to the results of variational mode decomposition; if it is judged that two subsequences attract each other, performing sequence fusion operation, and if it is judged that two subsequences repel each other, performing sequence generation operation; obtaining a new set of subsequences through sequence fusion operation and sequence generation operation; respectively inputting the sequences in the new set of subsequences into a recurrent neural network to obtain the final prediction result of the energy storage frequency modulation commands. By optimizing the frequency modulation commands in advance, the present invention enables the energy storage system to respond to the frequency change requirements more quickly and accurately, significantly improves the frequency modulation efficiency and economic benefits of the power system, and at the same time enhances the stability of the system and the operation efficiency of the power plant.

Description

Technical Field

[0001] The present invention relates to the technical field of energy storage frequency modulation prediction, and particularly to a method and system for predicting energy storage frequency modulation commands based on a two-stage method. Background Art

[0002] In a power system, frequency regulation is a key link to ensure the stable operation of the power grid. The response speed of traditional thermal power units is slow, and it is difficult to quickly adapt to load changes and maintain the stability of the power grid frequency. To improve this situation, a hybrid energy storage system (such as a combination of supercapacitor and lithium battery) is usually used to assist the thermal power unit in frequency regulation. In this mode, the hybrid energy storage system is responsible for handling the difference between the frequency modulation command and the output of the thermal power unit. The high-frequency component is borne by the supercapacitor with a fast response speed, while the low-frequency component is handled by the lithium battery with a higher energy density. However, in the actual operation process, the transmission of signals (frequency modulation commands transmitted to the supercapacitor / lithium battery) takes time, and the response of the supercapacitor or lithium battery itself also takes time, which will inevitably lead to time delay, affecting not only the immediate response ability of the system but also having a negative impact on the economic benefits of the power plant.

[0003] To solve this problem, researchers have developed prediction algorithms to predict the change trend of frequency modulation commands in advance, so that the energy storage system can act in advance, thereby improving the effect and economic benefits of frequency regulation. One of the traditional prediction methods is to use variational mode decomposition (VMD) to decompose the original frequency modulation command sequence into multiple subsequences, predict these subsequences separately, and finally synthesize the prediction results. Although this method helps to improve the prediction accuracy, it also has some limitations: for example, similar subsequences may lead to repeated calculations; at the same time, there may be mutually exclusive situations between some subsequences, generating false components, and new subsequences need to be generated additionally for compensation. Summary of the Invention

[0004] In view of the above existing problems, the present invention is proposed.

[0005] Therefore, the present invention provides a method and system for predicting energy storage frequency modulation commands based on a two-stage method to solve the problem that the similarity and exclusivity between subsequences are not effectively processed in the traditional decomposition and prediction process, which may lead to repeated calculations and false components, affecting the prediction accuracy and timeliness.

[0006] To solve the above technical problems, the present invention provides the following technical solutions:

[0007] In the first aspect, the present invention provides a method for predicting energy storage frequency modulation commands based on a two-stage method, including:

[0008] Obtain the energy storage frequency modulation command and perform variational mode decomposition;

[0009] Judgment is made on whether the subsequences attract or repel each other according to the results of the variational mode decomposition;

[0010] If it is determined that two subsequences attract each other, a sequence fusion operation is performed to fuse the two subsequences into a new sequence. If it is determined that two subsequences repel each other, a sequence generation operation is performed to generate a new sequence based on the two mutually repulsive subsequences;

[0011] A new set of subsequences is obtained through the sequence fusion operation and the sequence generation operation;

[0012] The sequences in the new set of subsequences are respectively input into a recurrent neural network to obtain the final prediction result of the energy storage frequency modulation command.

[0013] As a preferred solution of the energy storage frequency modulation command prediction method based on the two-stage method of the present invention, wherein: the judgment on whether the subsequences attract or repel each other includes:

[0014] If the subsequence IMF r = [X r1 , X r2 , X r3 ,.. X ri , X rN and the subsequence IMF j = [X j1 , X j2 , X j3 ,.. X ji , X jN satisfy the determination conditions for sequence mutual attraction, it is determined that the subsequences IMF r and IMF j attract each other; if the determination conditions for sequence mutual repulsion are satisfied, it is determined that the subsequences IMF r and IMF j repel each other;

[0015] The determination conditions for sequence mutual attraction are:

[0016] and

[0017] ,

[0018] wherein, N represents the total length of the energy storage frequency modulation sequence, represents the experimental group error of the subsequence IMF j , represents the experimental group error of the subsequence IMF r ;

[0019] The determination conditions for sequence mutual repulsion are:

[0020] and

[0021] 。

[0022] As a preferred embodiment of the energy storage frequency modulation command prediction method based on the two - stage method of the present invention, wherein: the judgment of mutual attraction or mutual repulsion between the subsequences further includes:

[0023] If a certain subsequence IMF q is both mutually attractive to the subsequence IMF j and mutually repulsive to the subsequence IMF r then compare the magnitude of the attraction between the subsequence IMF q and the subsequence IMF j with the magnitude of the repulsion between the subsequence IMF q and the subsequence IMF r ;

[0024] If the attraction is greater than the repulsion , then it is determined that the subsequence IMF q is mutually attractive to the subsequence IMF j , otherwise it is determined that the subsequence IMF q is mutually repulsive to the subsequence IMF r .

[0025] As a preferred embodiment of the energy storage frequency modulation command prediction method based on the two - stage method of the present invention, wherein: the sequence fusion operation includes:

[0026] If it is determined that the subsequence IMF r and IMF j are mutually attractive, then perform a sequence fusion operation to fuse the subsequence IMF r and IMF j into a new sequence IMF rj , wherein the element value at the corresponding position i in the new sequence IMF rj is expressed as: :

[0027] ,

[0028] wherein, represents the front - term coefficient, represents the back - term coefficient.

[0029] As a preferred embodiment of the energy storage frequency modulation command prediction method based on the two - stage method of the present invention, wherein: the sequence generation operation includes:

[0030] If it is determined that the subsequence IMF r and IMFj If they are mutually exclusive, a sequence generation operation is performed, and according to the sub-sequence IMF r and IMF j generate a new sequence IMF r-j , where the element value at the corresponding position i in the new sequence IMF r-j is expressed as: as follows:

[0031] .

[0032] As a preferred solution of the energy storage frequency modulation command prediction method based on the two-stage method described in the present invention, wherein: the process of obtaining the new sub-sequence set includes:

[0033] Using the two-stage method to traverse the sequences IMF1, IMF2, IMF3,..., IMF after variational mode decomposition K , and judge whether the sub-sequences attract or repel each other. If the sub-sequences attract each other, a sequence fusion operation is performed. If the sub-sequences repel each other, a sequence generation operation is performed; continuously loop and judge until the final sequences IMF1 、 , IMF2 、 , IMF3 、 ,..., IMF S 、 neither have sequences that attract each other nor sequences that repel each other.

[0034] As a preferred solution of the energy storage frequency modulation command prediction method based on the two-stage method described in the present invention, wherein: the process of obtaining the final prediction result of the energy storage frequency modulation command includes:

[0035] Input each sequence in the new sub-sequence set into a recurrent neural network respectively, output the values of the corresponding unknown groups, and then add up the values of the unknown groups predicted by each sub-sequence to obtain the final prediction result of the energy storage frequency modulation command.

[0036] In a second aspect, the present invention provides an energy storage frequency modulation command prediction system based on a two-stage method, including:

[0037] An instruction decomposition module, configured to obtain an energy storage frequency modulation command and perform variational mode decomposition;

[0038] A sequence processing module, configured to judge whether the sub-sequences attract or repel each other according to the result of the variational mode decomposition. If the sub-sequences attract each other, a sequence fusion operation is performed. If the sub-sequences repel each other, a sequence generation operation is performed; a new sub-sequence set is obtained through the sequence fusion operation and the sequence generation operation;

[0039] An instruction prediction module, configured to respectively input the sequences in the new subsequence set into a recurrent neural network to obtain the final prediction result of the energy storage frequency modulation instruction.

[0040] In a third aspect, the present invention provides an electronic device, including:

[0041] A memory and a processor;

[0042] The memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions. When the computer-executable instructions are executed by the processor, the steps of the energy storage frequency modulation instruction prediction method based on the two-stage method are implemented.

[0043] In a fourth aspect, the present invention provides a computer-readable storage medium, which stores computer-executable instructions. When the computer-executable instructions are executed by a processor, the steps of the energy storage frequency modulation instruction prediction method based on the two-stage method are implemented.

[0044] Compared with the prior art, the beneficial effects of the present invention are as follows: The present invention provides an energy storage frequency modulation instruction prediction method and system based on the two-stage method. By introducing an innovative processing method combining variational mode decomposition with the two-stage method, intelligent preprocessing of the energy storage frequency modulation instruction is performed. For subsequences that attract each other, a fusion strategy is implemented to reduce redundancy, and for subsequences that repel each other, new independent sequences are created to eliminate the influence of false components. Finally, a recurrent neural network is used to achieve accurate prediction. The present invention can not only effectively identify and process the similarity and repulsion between subsequences, reduce redundant calculations and eliminate false components, thereby improving the accuracy and stability of the prediction model, but also optimize the frequency modulation instruction in advance, enabling the energy storage system to respond more quickly and accurately to the frequency change requirements, significantly improving the frequency modulation efficiency and economic benefits of the power system, while enhancing the stability of the system and the operation efficiency of the power plant. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0046] Figure 1 It is a schematic diagram of the overall process logic of the energy storage frequency modulation instruction prediction method based on the two-stage method according to an embodiment of the present invention;

[0047] Figure 2 It is a schematic diagram of the prediction process of the energy storage frequency modulation instruction prediction method based on the two-stage method according to an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0048] To make the above objects, features, and advantages of the present invention more obvious and understandable, the following provides a detailed description of the specific embodiments of the present invention in conjunction with the accompanying drawings of the specification. Obviously, the described embodiments are part of the 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.

[0049] Embodiment 1

[0050] Refer to Figure 1 - Figure 2 For an embodiment of the present invention, a prediction method for energy storage frequency modulation commands based on a two-stage method is provided, aiming to more effectively utilize the energy storage system, improve the accuracy of frequency modulation command prediction, and thereby enhance the stability of the power system and the operation efficiency of power plants. As Figure 1 shown, it specifically includes the following steps:

[0051] S100: Obtain the energy storage frequency modulation command and perform variational mode decomposition;

[0052] S200: According to the results of variational mode decomposition, judge whether the subsequences attract or repel each other. If the subsequences attract each other, perform sequence fusion operation; if the subsequences repel each other, perform sequence generation operation; obtain a new set of subsequences through sequence fusion operation and sequence generation operation;

[0053] S300: Input the sequences in the new set of subsequences into a recurrent neural network respectively to obtain the final prediction result of the energy storage frequency modulation command.

[0054] It should be noted that the present invention provides a prediction method and system for energy storage frequency modulation commands based on a two-stage method. Through an innovative processing method that combines variational mode decomposition with a two-stage method, intelligent preprocessing is performed on the energy storage frequency modulation commands. For subsequences that attract each other, a fusion strategy is implemented to reduce redundancy; for subsequences that repel each other, new independent sequences are created to eliminate the influence of false components. Finally, a recurrent neural network is used to achieve accurate prediction. The present invention can not only effectively identify and process the similarity and repulsion between subsequences, reduce redundant calculations and eliminate false components, thereby improving the accuracy and stability of the prediction model; but also by optimizing the frequency modulation commands in advance, the energy storage system can respond to frequency change requirements more quickly and accurately, significantly improving the frequency modulation efficiency and economic benefits of the power system, while enhancing the stability of the system and the operation efficiency of power plants.

[0055] In the embodiment of the present application, the above step S100 of obtaining the energy storage frequency modulation command and performing variational mode decomposition includes:

[0056] Set the energy storage frequency modulation command to be represented as Pt = [X1, X2, X3,...X i .., XN ;

[0057] Use the sequence with a length of 0.9N at the front in the energy storage frequency modulation instruction [X1, X2, X3....., X 0.9N as the input to pre-test the values of the experimental group [X 0.9N+1 , X 0.9N+2 , X 0.9N+3 ,..., X N and the [X N+1 , X N+2 , X N+3 ,..., X N+0.1N of the unknown group;

[0058] Perform variational mode decomposition on the energy storage frequency modulation instruction Pt to obtain the sequences IMF1, IMF2, IMF3,..., IMF after variational mode decomposition K ;

[0059] It should be noted that the actual values and predicted values of the experimental group are used to generate all the errors (MAPE) in the following of this embodiment. What the present invention truly predicts is the value of the unknown group.

[0060] It should be noted that the above step S100 can effectively decompose the complex original frequency modulation instruction sequence into multiple subsequences with different characteristic frequencies by obtaining the energy storage frequency modulation instruction and performing variational mode decomposition. As an adaptive signal processing technology, variational mode decomposition can more accurately capture the essential features of the signal compared with traditional decomposition methods, ensuring that each subsequence carries the effective information of the original signal, providing a high-quality data basis for the subsequent two-level rule judgment, sequence fusion or generation operations, thereby improving the performance and reliability of the entire prediction system.

[0061] In the embodiment of the present application, the above step S200 determines whether the subsequences attract or repel each other according to the results of variational mode decomposition. If the subsequences attract each other, sequence fusion operations are performed; if the subsequences repel each other, sequence generation operations are performed; a new set of subsequences is obtained through sequence fusion operations and sequence generation operations, including:

[0062] Specifically, the determination of whether the subsequences attract or repel each other includes:

[0063] If the subsequence IMF r = [X r1 , X r2 , X r3 ,..X ri ,.., X rN and the subsequence IMF j = [X j1 , X j2 , X j3 ,..Xji ..,X jN If the determination conditions for the sequences to attract each other are met, then the subsequence IMF is determined. r and IMF j attract each other; if the subsequence IMF r =[X r1 ,X r2 ,X r3 ,..X ri ..,X rN and the subsequence IMF j =[X j1 ,X j2 ,X j3 ,..X ji ..,X jN meet the determination conditions for the sequences to repel each other, then the subsequence IMF r and IMF j repel each other;

[0064] Among them, the determination conditions for the sequences to attract each other are:

[0065] and

[0066] ,

[0067] where N represents the total length of the energy storage frequency modulation sequence, represents the experimental group error of the subsequence IMF j ; represents the experimental group error of the subsequence IMF r ;

[0068] Among them, the determination conditions for the sequences to repel each other are:

[0069] and

[0070] ,

[0071] Specifically, the judgment of the subsequences attracting or repelling each other also includes:

[0072] If a certain subsequence IMF q is both attracted to the subsequence IMF j and repelled by the subsequence IMF r , then compare the magnitude of the attraction between the subsequence IMF q and the subsequence IMF j with the magnitude of the repulsion between the subsequence IMF q and the subsequence IMF r ;

[0073] If the attraction Greater than the repulsive force , then the subsequence IMF is determined q and the subsequence IMF j attract each other;

[0074] If the attractive force is less than the repulsive force , then the subsequence IMF is determined q and the subsequence IMF r repel each other.

[0075] Specifically, the sequence fusion operation includes:

[0076] If it is determined that the subsequence IMF r and IMF j attract each other, then perform the sequence fusion operation to fuse the subsequence IMF r and IMF j into a new sequence IMF rj ; where the element value at the corresponding position i in the new sequence IMF rj is expressed as: :

[0077] ,

[0078] wherein, represents the front-term coefficient, represents the back-term coefficient.

[0079] Specifically, the sequence generation operation includes:

[0080] If it is determined that the subsequence IMF r and IMF j repel each other, then perform the sequence generation operation to generate a new sequence IMF r and IMF j based on the subsequence IMF r-j ; where the element value at the corresponding position i in the new sequence IMF r-j is expressed as: :

[0081] ,

[0082] Specifically, use the two-level method to traverse the sequences IMF1, IMF2, IMF3,..., IMF after variational mode decomposition K , and judge whether the subsequences attract or repel each other. If the subsequences attract each other, perform the sequence fusion operation; if the subsequences repel each other, perform the sequence generation operation; continuously loop and judge until the final sequences IMF1 、 , IMF2 、 , IMF3、 ,...,IMF S 、 There is neither an attracting sequence nor a repelling sequence.

[0083] It should be noted that the above step S200 makes an intelligent judgment of attraction or repulsion for the subsequences after variational mode decomposition by introducing a two - level rule, and accordingly implements sequence fusion or generation operations. This not only effectively reduces the redundant calculation caused by similar subsequences, avoids resource waste, but also eliminates the influence of false components by creating new independent sequences, improving the data input quality of the prediction model. This step enhances the accuracy and stability of the subsequent recurrent neural network prediction, enabling the energy storage system to respond more precisely to the frequency change requirements, thereby improving the frequency modulation efficiency and economic benefits of the power system, and at the same time ensuring the stability and reliability of the power grid operation.

[0084] In the embodiment of the present application, the above step S300 inputs the sequences in the new subsequence set into the recurrent neural network respectively, and the final prediction result of the energy storage frequency modulation command includes:

[0085] As Figure 2 shown, each sequence in the new subsequence set is input into the recurrent neural network respectively, the value of the corresponding unknown group is output, and then the values of the unknown groups predicted by each subsequence are added to obtain the final prediction result of the energy storage frequency modulation command.

[0086] It should be noted that the above step S300 can make full use of the powerful non - linear modeling ability of the neural network by inputting the optimized new subsequence set into the recurrent neural network respectively to accurately predict the energy storage frequency modulation command. The recurrent neural network is particularly good at processing data with time - series characteristics, and can capture the long - term dependence relationship in the subsequence, ensuring that the prediction result is not only accurate but also stable and reliable, greatly improving the operation efficiency and economic benefits of the power plant, and at the same time enhancing the overall stability and flexibility of the power system.

[0087] Embodiment 2

[0088] Based on the previous embodiment, this embodiment provides an application example of a method and system for predicting energy storage frequency modulation commands based on a two - level method to verify and illustrate the technical effects adopted in this method.

[0089] In order to further verify the advantages of the present invention, this embodiment respectively uses the method for predicting energy storage frequency modulation commands provided by the present invention and the prediction method of VMD - GRU to predict the frequency modulation sequence. The prediction results are shown in Table 1, and the definitions of 4 evaluation indexes are shown in Table 2.

[0090] Table 1: Prediction results of the method of the present invention and the prediction method of VMD - GRU for the frequency modulation sequence.

[0091] ,

[0092] Table 2: Four evaluation metrics.

[0093] ,

[0094] Where N represents the sample size, and represent the actual value and the predicted value at time n, respectively.

[0095] From the above comparison of the prediction results, it can be seen that the energy storage frequency modulation command prediction method provided by the present invention performs well in all four evaluation metrics, and its performance is better than that of the VMD-GRU prediction method.

[0096] Therefore, the present invention introduces an innovative processing method combining variational mode decomposition with a two-stage rule to perform intelligent preprocessing on the energy storage frequency modulation command. For subsequences that attract each other, a fusion strategy is implemented to reduce redundancy, and for subsequences that repel each other, new independent sequences are created to eliminate the influence of false components. Finally, a recurrent neural network is used to achieve accurate prediction. The present invention can not only effectively identify and process the similarity and repulsion between subsequences, reduce redundant calculations and eliminate false components, thereby improving the accuracy and stability of the prediction model, but also optimize the frequency modulation command in advance, enabling the energy storage system to respond more quickly and accurately to the frequency change requirements, significantly improving the frequency modulation efficiency and economic benefits of the power system, while enhancing the stability of the system and the operation efficiency of the power plant.

[0097] Embodiment 3

[0098] In this embodiment, an energy storage frequency modulation command prediction system based on a two-stage method is provided, including a command decomposition module, a sequence processing module, and a command prediction module;

[0099] Specifically, the command decomposition module is used to obtain the energy storage frequency modulation command and perform variational mode decomposition;

[0100] Specifically, the sequence processing module is used to judge whether the subsequences attract or repel each other according to the result of variational mode decomposition. If the subsequences attract each other, a sequence fusion operation is performed; if the subsequences repel each other, a sequence generation operation is performed; a new subsequence set is obtained through the sequence fusion operation and the sequence generation operation;

[0101] Specifically, the command prediction module is used to input the sequences in the new subsequence set into the recurrent neural network respectively to obtain the final prediction result of the energy storage frequency modulation command.

[0102] It should be noted that the technical solution of the energy storage frequency modulation command prediction system based on the two - level method and the technical solution of the above - mentioned energy storage frequency modulation command prediction method based on the two - level method belong to the same concept. For the details not described in detail in the technical solution of the energy storage frequency modulation command prediction system based on the two - level method in this embodiment, reference can be made to the description of the technical solution of the energy storage frequency modulation command prediction method based on the two - level method above.

[0103] The above - mentioned each unit module can be embedded in the processor in the computer device in the form of hardware or be independent of the processor, or can be stored in the memory in the computer device in the form of software, so as to facilitate the processor to call and execute the operations corresponding to each above - mentioned module.

[0104] This embodiment also provides an electronic device, which includes a processor, a memory, a communication interface, a display screen and an input device connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non - volatile storage medium and an internal memory. The non - volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non - volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be realized through WIFI, a carrier network, NFC (Near Field Communication) or other technologies. When the computer program is executed by the processor, it realizes an energy storage frequency modulation command prediction method based on the two - level method. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covering the display screen, or can be a key, a trackball or a touchpad set on the shell of the computer device, or can also be an external keyboard, a touchpad or a mouse, etc.

[0105] This embodiment also provides a computer - readable storage medium, on which a computer program is stored. When the program is executed by the processor, it realizes the method proposed in the above - mentioned embodiment.

[0106] The storage medium proposed in this embodiment and the method proposed in the above - mentioned embodiment belong to the same inventive concept. The technical details not described in detail in this embodiment can be referred to the above - mentioned embodiment, and this embodiment has the same beneficial effects as the above - mentioned embodiment.

[0107] From the above description of the embodiments, those skilled in the art can clearly understand that the present invention can be implemented by means of software and necessary general-purpose hardware. Of course, it can also be implemented by hardware, but in many cases the former is a better implementation. Based on this understanding, the technical solution 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. This computer software product can be stored in a computer-readable storage medium, such as a floppy disk, read-only memory (ROM), random access memory (RAM), flash memory, hard disk, or optical disc of a computer, etc., including several instructions for causing a computer device (which can be a personal computer, server, or network device, etc.) to execute the method of the embodiments of the present invention.

[0108] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not restrictive. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present invention, and they should all be covered by the scope of the claims of the present invention.

[0109] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code. The solutions in the embodiments of the present application can be implemented in various computer languages.

[0110] The present application is described with reference to the flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or block in the flowchart and / or block diagram can be implemented by computer program instructions, and the combination of the processes and / or blocks in the flowchart and / or block diagram can also be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices generate a device for realizing the functions specified in Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0111] These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means that implement the function specified in one or more of the flows Figure 1 and / or boxes Figure 1 specified in one or more of the boxes.

[0112] These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process, whereby the instructions executed on the computer or other programmable apparatus provide steps for implementing the function specified in one or more of the flows Figure 1 and / or boxes Figure 1 specified in one or more of the boxes.

[0113] Although the preferred embodiments of the present application have been described, additional changes and modifications can be made by those skilled in the art once they learn of the basic inventive concept. Therefore, the appended claims are intended to be construed to cover the preferred embodiments as well as all changes and modifications that fall within the scope of the present application.

[0114] It is apparent that those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalent technologies, the present application is also intended to include these modifications and variations.

Claims

1. A prediction method for energy storage frequency modulation commands based on a two-stage method, characterized in that Comprising: Obtaining a frequency modulation instruction for energy storage and performing variational mode decomposition; Judging the mutual attraction and mutual repulsion of subsequences according to the result of the variational mode decomposition; If it is judged that there is mutual attraction between two subsequences, a sequence fusion operation is performed to fuse the two subsequences into a new sequence. If it is judged that there is mutual repulsion between two subsequences, a sequence generation operation is performed to generate a new sequence according to the two mutually repulsive subsequences; Obtaining a new set of subsequences through the sequence fusion operation and the sequence generation operation; Respectively inputting the sequences in the new set of subsequences into a recurrent neural network to obtain the final prediction result of the energy storage frequency modulation instruction; The judgment of the mutual attraction and mutual repulsion of the subsequences includes: If the subsequences and the subsequences meet the determination conditions for sequence mutual attraction, then it is determined that the subsequences and attract each other; if they meet the determination conditions for sequence mutual repulsion, then it is determined that the subsequences and repel each other; The determination condition for the mutual attraction of the sequences is: and Among them, N represents the total length of the energy storage frequency modulation sequence, represents the subsequence experimental group error of, represents the subsequence experimental group error of; The determination condition for the mutual repulsion of the sequences is: and The judgment of the mutual attraction and mutual repulsion of the subsequences further includes: If a certain subsequence is both attracted to the subsequence and repelled by the subsequence , then compare the magnitude of the attraction between the subsequence and the subsequence with the magnitude of the repulsion between the subsequence and the subsequence ; If the attraction force is greater than the repulsive force , it is determined that the subsequence and the subsequence attract each other; otherwise, it is determined that the subsequence and the subsequence repel each other; The sequence fusion operation includes: If it is determined that the subsequences and attract each other, then a sequence fusion operation is performed to fuse the subsequences and into a new sequence , where the element value at the corresponding position i in the new sequence is expressed as: Among them, represents the coefficient of the previous term, represents the coefficient of the subsequent term; The sequence generation operation includes: If it is determined that the subsequences and are mutually exclusive, then a sequence generation operation is performed to generate a new sequence and according to the subsequences , where the element value at the corresponding position i in the new sequence is expressed as: The process of obtaining the new set of subsequences includes: Traverse the sequence after variational mode decomposition using a two-level method , and judge the mutual attraction and mutual repulsion of subsequences. If the subsequences attract each other, perform sequence fusion operation; if the subsequences repel each other, perform sequence generation operation; continuously loop and judge until the final sequence has neither sequences that attract each other nor sequences that repel each other.

2. The energy storage frequency modulation command prediction method based on a two-stage method according to claim 1, wherein, The process of obtaining the final prediction result of the energy storage frequency modulation instruction includes: Respectively inputting each sequence in the new set of subsequences into a recurrent neural network, outputting the value of the corresponding unknown group, and then adding the values of the unknown groups predicted by each subsequence to obtain the final prediction result of the energy storage frequency modulation instruction.

3. A prediction system for energy storage frequency modulation commands based on a two-stage method, which applies a prediction method for energy storage frequency modulation commands based on a two-stage method as described in any one of claims 1 to 2, characterized in that, Comprising: An instruction decomposition module, configured to obtain an energy storage frequency modulation instruction and perform variational mode decomposition; A sequence processing module, configured to judge the mutual attraction and mutual repulsion of subsequences according to the result of the variational mode decomposition. If the subsequences are mutually attractive, a sequence fusion operation is performed. If the subsequences are mutually repulsive, a sequence generation operation is performed; Obtaining a new set of subsequences through the sequence fusion operation and the sequence generation operation; An instruction prediction module, configured to respectively input the sequences in the new set of subsequences into a recurrent neural network to obtain the final prediction result of the energy storage frequency modulation instruction.

4. An electronic device, comprising a memory and a processor, characterized in that: The memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions. When the computer-executable instructions are executed by the processor, the steps of the method according to any one of claims 1 to 2 are implemented.

5. A computer-readable storage medium having computer-executable instructions stored thereon, characterized in that: When the computer-executable instructions are executed by the processor, the steps of the method according to any one of claims 1 to 2 are implemented.

Images