A method and related equipment for identifying spurious subsequences in energy storage frequency regulation prediction

Abstract

This invention relates to the field of frequency modulation sequence prediction technology, and discloses a method and related equipment for energy storage frequency modulation prediction that identifies spurious subsequences. The method includes: decomposing a frequency modulation command sequence into multiple subsequences and determining the spurious parameter of each subsequence; determining the spuriousness degree of each subsequence based on the determined spurious parameter; identifying the subsequence corresponding to the maximum spuriousness degree based on the spuriousness degree of each subsequence, removing this subsequence as the one with the largest spurious component, and generating a compensation sequence for the removed subsequence; inputting the remaining subsequences and the compensation sequence into a GRU model for prediction to obtain multiple prediction results; and superimposing the multiple prediction results to generate a final prediction value. This invention comprehensively considers the effective data after removing spurious information and the compensation sequence, making the prediction results more accurate and reliable, and effectively improving prediction accuracy.

Description

Technical Field

[0001] This invention relates to the field of frequency modulation sequence prediction technology, specifically to a method and related equipment for identifying spurious subsequences in energy storage frequency modulation prediction. Background Technology

[0002] In a power system, a frequency regulation command sequence is a set of instructions used to adjust the power system frequency. These instructions are typically generated based on factors such as frequency deviation, load changes, and the operating status of generating equipment. The purpose of the frequency regulation command sequence is to maintain the frequency stability of the power system and ensure power quality by adjusting the output power of generating equipment (such as generator sets) or energy storage devices.

[0003] Traditional forecasting methods are often used for predictive analysis of various types of sequence data to predict future trends and support decision-making. Among them, the forecasting method based on Variational Mode Decomposition (VMD) is a relatively common and valuable technique.

[0004] The basic principle of the VMD decomposition method is that it can decompose the original complex sequence into a series of subsequences. By predicting these subsequences separately and then superimposing the prediction results of each subsequence, the final prediction result of the original sequence is obtained. This method of decomposition, separate prediction, and superposition can, to some extent, help to extract feature information of different frequency components in the original sequence, providing a possible way to improve prediction accuracy.

[0005] However, the VMD decomposition method has revealed significant shortcomings in practical applications. Specifically, this method may generate spurious subsequences during the decomposition process. These spurious subsequences do not effectively reflect the true characteristics of the original sequence; their presence not only fails to provide valuable information for the prediction process but also negatively impacts prediction accuracy. The interference of spurious subsequences may cause biases in the prediction model during training and prediction, resulting in significant errors between the final prediction results and the actual situation, thereby affecting the scientific validity and accuracy of decisions made based on the prediction results.

[0006] Therefore, in view of the above-mentioned problems of VMD decomposition method, it is necessary to develop a new prediction method or technical improvement scheme that can overcome the interference of spurious subsequences and improve prediction accuracy, so as to meet the demand for more accurate prediction results in practical applications. Summary of the Invention

[0007] In order to overcome the shortcomings of the existing technology, the purpose of this invention is to provide a method and related equipment for predicting energy storage frequency regulation by identifying spurious subsequences, so as to solve the technical problem of how to overcome spurious subsequence interference and improve prediction accuracy in the existing technology.

[0008] This invention is achieved through the following technical solution:

[0009] In a first aspect, the present invention provides a method for predicting energy storage frequency regulation by identifying spurious subsequences, comprising:

[0010] The frequency modulation command sequence is decomposed into multiple subsequences, and the spurious parameters of each subsequence are determined;

[0011] The degree of falsity for each subsequence is determined based on the identified falsity parameters;

[0012] Based on the falseness identification of each subsequence, the subsequence corresponding to the maximum falseness is obtained. The subsequence with the largest false component is removed, and a compensation sequence is generated for the removed subsequence.

[0013] The unremoved subsequence and the compensated sequence are respectively input into the GRU model for prediction and output multiple prediction results. The multiple prediction results are superimposed to generate the final prediction value.

[0014] Preferably, the process for determining the spurious parameters of each subsequence includes:

[0015] S1, set a sine function, and by adjusting the parameters of the sine function, record the data points that fall into the sine function and the number of corresponding data points in a subsequence of the frequency modulation command sequence according to the set landing conditions, and determine the data points that do not fall into the sine function as false parameters in the corresponding subsequence;

[0016] S2, determine the spurious parameters in the corresponding subsequences by sequentially determining the unremoved subsequences in the frequency modulation command sequence according to S1;

[0017] S3, form a set of dummy parameters by corresponding to the dummy parameters determined by each subsequence in the frequency modulation command sequence.

[0018] Furthermore, in S1, the landing point condition is to substitute the time series point into the sine function. If the calculated value of the sine function is equal to the data point of the subsequence corresponding to the time series point, then it is determined that the data point falls on the sine function; otherwise, the data point does not fall on the sine function.

[0019] The preferred formula for calculating the falsehood rate is as follows:

[0020]

[0021] in, , , , The dummy parameter representing the subsequence Indicates the number of landing points; || represents the absolute value.

[0022] Preferably, the process of generating the compensated sequence includes:

[0023] Based on the frequency parameters of the eliminated subsequences, phase parameters are randomly generated within a preset phase offset range to construct a sine function sequence with the same frequency characteristics but different phases.

[0024] Preferably, the calculation formula for compensating for sequence generation includes:

[0025]

[0026] in, Parameters representing the compensated sequence; The parameter representing the subsequence to be removed; Indicates the minimum degree of falsehood; ∣∣ represents the maximum falseness; |∣ represents the absolute value.

[0027] Preferably, multiple prediction results are fused and superimposed using a weighted summation method to generate the final prediction value.

[0028] Secondly, the present invention also provides a storage frequency regulation prediction system for identifying spurious subsequences, comprising:

[0029] The spurious parameter determination module is used to decompose the frequency modulation command sequence into multiple sub-sequences and determine the spurious parameters of each sub-sequence;

[0030] The spuriousness calculation module is used to determine the spuriousness of each subsequence based on the determined spuriousness parameters;

[0031] The sequence processing module is used to identify the subsequence corresponding to the maximum false score based on the false score of each subsequence, remove the subsequence as the subsequence with the largest false score, and generate a compensation sequence for the removed subsequence.

[0032] The sequence prediction module is used to input the unremoved subsequence and the compensated sequence into the GRU model for prediction and output multiple prediction results. The multiple prediction results are then superimposed to generate the final prediction value.

[0033] Thirdly, the present invention also provides a mobile terminal, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the energy storage frequency modulation prediction method for identifying spurious subsequences as described above.

[0034] Fourthly, the present invention also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the energy storage frequency modulation prediction method for identifying spurious subsequences as described above.

[0035] Compared with the prior art, the present invention has the following beneficial technical effects:

[0036] This invention provides a method for predicting energy storage frequency regulation by identifying spurious subsequences. By decomposing the frequency regulation command sequence into multiple subsequences, determining the spurious parameters and spuriousness, and identifying and removing the subsequence with the largest spurious component, it specifically addresses the interference of spurious subsequences on the prediction process, reducing the adverse impact of false information on prediction accuracy. After removing the subsequence with the largest spurious component, a compensation sequence is generated. The remaining subsequence and the compensation sequence are then input into a GRU model for prediction. The multiple prediction results are then superimposed to generate the final prediction value. This method comprehensively considers the effective data after removing spurious information and the compensation sequence, making the prediction results more accurate and reliable, effectively improving prediction precision.

[0037] Furthermore, by setting a sine function and recording the data points and their quantities according to the landing conditions, data points in the subsequence that do not conform to the sine function can be quantitatively identified, i.e., spurious parameters. By applying the same identification steps to each subsequence in the frequency modulation command sequence, the systematic nature and consistency of the entire process are ensured. By identifying and eliminating spurious data points corresponding to spurious parameters, the interference of this false information on the prediction model can be reduced, thereby improving prediction accuracy.

[0038] Furthermore, by substituting time series points into a sine function and comparing the calculated sine function value with the value of the corresponding subsequence data point, it is possible to accurately determine whether a data point falls on the sine function. If they are not equal, the data point is clearly a spurious parameter. This precise positioning method can effectively identify data that does not conform to the sine function, avoid fuzzy judgments, and improve the accuracy of spurious information identification.

[0039] Furthermore, constructing a sine function sequence based on the frequency parameters of the removed subsequences ensures that the compensation sequence maintains consistency with the original frequency modulation command sequence in terms of the key characteristic of frequency. Attached Figure Description

[0040] Figure 1 This is a flowchart of the energy storage frequency modulation prediction method for identifying spurious subsequences in an embodiment of the present invention;

[0041] Figure 2 This is a diagram illustrating the process of determining spurious parameters for each subsequence in an embodiment of the present invention.

[0042] Figure 3 This is a schematic diagram of the energy storage frequency modulation prediction system for identifying spurious subsequences in an embodiment of the present invention;

[0043] In the diagram: 1. False parameter determination module; 2. Falseness calculation module; 3. Sequence processing module; 4. Sequence prediction module. Detailed Implementation

[0044] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0045] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0046] The purpose of this invention is to provide a method and related equipment for predicting energy storage frequency regulation by identifying spurious subsequences, so as to solve the technical problem of how to overcome spurious subsequence interference and improve prediction accuracy in the prior art.

[0047] The terms used in this invention are explained as follows:

[0048] GRU model: Gated Recurrent Unit;

[0049] VMD decomposition: Variational Mode Decomposition;

[0050] IMF: Intrinsic Mode Function;

[0051] MAE: Mean Absolute Error;

[0052] SSE: Sum of Squared Errors, or residual sum of squares;

[0053] RMSE: Root Mean Squared Error.

[0054] MAPE: Mean Absolute Percentage Error.

[0055] The present invention will now be described in further detail with reference to the accompanying drawings:

[0056] Example 1

[0057] See Figure 1 In one embodiment of the present invention, a method for predicting energy storage frequency modulation by identifying spurious subsequences is provided, comprising:

[0058] Step 1: Decompose the frequency modulation command sequence into multiple sub-sequences and determine the spurious parameters of each sub-sequence;

[0059] Specifically, according to Figure 2 As shown, the process of determining the spurious parameters for each subsequence includes:

[0060] S1, set a sine function, and by adjusting the parameters of the sine function, record the data points that fall into the sine function and the number of corresponding data points in a subsequence of the frequency modulation command sequence according to the set landing conditions, and determine the data points that do not fall into the sine function as false parameters in the corresponding subsequence;

[0061] The landing point condition is that if the calculated value of the sine function is equal to the data point of the subsequence corresponding to the time series point, then the data point is determined to fall on the sine function; otherwise, the data point does not fall on the sine function.

[0062] S2, determine the spurious parameters in the corresponding subsequences by sequentially determining the unremoved subsequences in the frequency modulation command sequence according to S1;

[0063] S3, form a set of dummy parameters by corresponding to the dummy parameters determined by each subsequence in the frequency modulation command sequence.

[0064] Step 2: Determine the spuriousness of each subsequence based on the determined spurious parameters;

[0065] Specifically, the formula for calculating the degree of falsehood is as follows:

[0066]

[0067] in, , , , The dummy parameter representing the subsequence Indicates the number of landing points; || represents the absolute value.

[0068] Step 3: Based on the falseness of each subsequence, the subsequence corresponding to the maximum falseness is obtained. The subsequence with the largest false component is removed, and a compensation sequence is generated for the removed subsequence.

[0069] Specifically, the process of generating the compensated sequence includes:

[0070] Based on the frequency parameters of the eliminated subsequences, phase parameters are randomly generated within a preset phase offset range to construct a sine function sequence with the same frequency characteristics but different phases.

[0071] Specifically, the calculation formula for compensating for sequence generation includes:

[0072]

[0073] in, Parameters representing the compensated sequence; The parameter representing the subsequence to be removed; Indicates the minimum degree of falsehood; ∣∣ represents the maximum falseness; |∣ represents the absolute value.

[0074] Step 4: Input the unremoved subsequence and the compensated sequence into the GRU model for prediction and output multiple prediction results. Then, superimpose the multiple prediction results to generate the final prediction value.

[0075] Specifically, multiple prediction results are combined and superimposed using a weighted summation method to generate the final prediction value.

[0076] The specific operations implemented in this embodiment are as follows:

[0077] The frequency modulation command sequence is set to Pt, and after VMD decomposition, it becomes IMF1, IMF2, IMF3, ... IMF. i , ...,IMF K .

[0078] Set one of the subsequences IMF of VMD decomposition. i =[X i1 ,X i2 ,X i3 ,.X ii ..,X iN ];

[0079] And generate a sine function ;

[0080] Continuously adjust parameters , so that [X i1 ,X i2 ,X i3 ,.X ii ..,X iN As many as possible fall The final adjusted parameters are as follows: , falling There are Mi points on it.

[0081] The landing point condition is to substitute t=1 into... ,if This means X i1 fall into Up, otherwise it did not fall in.

[0082] The spurious parameters for each subsequence are compiled using the method described above:

[0083] as subsequence IMF i parameters For example, the formula for calculating the falseness Di is:

[0084]

[0085] When the subsequence falls within approximately The more points a function has, the stronger the regularity of the sequence and the lower its falsehood, and vice versa.

[0086] Calculated based on the above method and find The maximum value in.

[0087] set up The subsequence corresponding to the maximum value is the subsequence with the largest spurious component. This subsequence will have a negative impact on the prediction, so it is removed.

[0088] After removing the subsequence with the highest degree of falsehood, a set of compensating sequences is generated.

[0089] Let the subsequence to be removed be IMF. s =[X s1 ,X s2 ,X s3 ,.X si ..,X sN Let the generated subsequence be IMF. s 、 =[X s1 、 ,X s2 、 ,X s3、 ,.X si 、 ..,X sN 、 ].

[0090] With X si 、 For example, the formula for compensating for a sequence is as follows:

[0091]

[0092] As can be seen from the above, the compensation sequence is a range and is randomly generated.

[0093] After generating the compensated sequence, the subsequences are fed into the GRU for prediction, and then the final prediction results are superimposed.

[0094] Suppose the final sequence is IMF1, IMF2, IMF3, ... s 、 ..,IMF K These sequences are fed into a GRU for prediction, and the final predictions are summed to obtain the final result.

[0095] To further verify the advantages of the present invention, the present invention uses both the method of the present invention and the VMD-GRU (K=7) prediction method to predict frequency modulation sequences, and the results are shown in Table 1:

[0096] Frequency modulation sequence 1 is derived from frequency modulation data from a power plant in Hulunbuir from 0:00 to 20:00 on December 1, 2024.

[0097] Frequency modulation sequence 2 is derived from frequency modulation data from a power plant in Hulunbuir from 2:00 AM to 10:00 PM on December 2, 2024.

[0098] Table 1 shows the prediction results for frequency modulation sequences 1 and 2.

[0099]

[0100] Table 2 Four evaluation indicators

[0101]

[0102] In the table, N represents the sample size. and These represent the actual value and the predicted value at time n, respectively.

[0103] As shown in Table 2, the experimental results indicate that all four evaluation indicators have decreased, demonstrating that the decomposition method proposed in this invention can find the decomposition layer K with the fewest spurious components, which is equivalent to the traditional VMD-GRU prediction method. This avoids the inherent drawback of spurious components in VMD decomposition and further improves the prediction accuracy. This helps power plants improve their frequency regulation response capabilities and further increase their profitability.

[0104] In summary, the present invention provides a method for predicting energy storage frequency regulation by identifying spurious subsequences. This method decomposes the frequency regulation command sequence into multiple subsequences, determines the spurious parameters and spuriousness, and identifies and removes the subsequence with the largest spurious component. This specifically addresses the interference of spurious subsequences on the prediction process and reduces the adverse impact of false information on prediction accuracy. After removing the subsequence with the largest spurious component, a compensation sequence is generated. The remaining subsequences and the compensation sequence are then input into a GRU model for prediction. The multiple prediction results are then superimposed to generate the final prediction value. This method comprehensively considers the effective data after removing spurious information and the compensation sequence, making the prediction results more accurate and reliable, and effectively improving prediction precision.

[0105] Example 2

[0106] according to Figure 3 As shown, the present invention also provides a storage frequency regulation prediction system for identifying spurious subsequences, comprising:

[0107] The spurious parameter determination module 1 is used to decompose the frequency modulation command sequence into multiple sub-sequences and determine the spurious parameters of each sub-sequence;

[0108] The falsehood calculation module 2 is used to determine the falsehood degree of each subsequence based on the determined falsehood parameters;

[0109] The sequence processing module 3 is used to identify the subsequence corresponding to the maximum false score based on the false score of each subsequence, remove the subsequence as the subsequence with the largest false score, and generate a compensation sequence for the removed subsequence.

[0110] Sequence prediction module 4 is used to input the unremoved subsequence and the compensated sequence into the GRU model for prediction and output multiple prediction results. The multiple prediction results are then superimposed to generate the final prediction value.

[0111] Example 3

[0112] The present invention also provides a mobile terminal, including a memory, a processor, and a computer program stored in the memory and executable on the processor, such as an energy storage frequency modulation prediction program for identifying spurious subsequences.

[0113] When the processor executes the computer program, it implements the steps of the above-described method for identifying spurious subsequences in energy storage frequency modulation prediction, for example:

[0114] The frequency modulation command sequence is decomposed into multiple subsequences, and the spurious parameters of each subsequence are determined;

[0115] The degree of falsity for each subsequence is determined based on the identified falsity parameters;

[0116] Based on the falseness identification of each subsequence, the subsequence corresponding to the maximum falseness is obtained. The subsequence with the largest false component is removed, and a compensation sequence is generated for the removed subsequence.

[0117] The unremoved subsequence and the compensated sequence are respectively input into the GRU model for prediction and output multiple prediction results. The multiple prediction results are superimposed to generate the final prediction value.

[0118] Alternatively, when the processor executes the computer program, it implements the functions of each module in the above system, for example:

[0119] The spurious parameter determination module 1 is used to decompose the frequency modulation command sequence into multiple sub-sequences and determine the spurious parameters of each sub-sequence;

[0120] The falsehood calculation module 2 is used to determine the falsehood degree of each subsequence based on the determined falsehood parameters;

[0121] The sequence processing module 3 is used to identify the subsequence corresponding to the maximum false score based on the false score of each subsequence, remove the subsequence as the subsequence with the largest false score, and generate a compensation sequence for the removed subsequence.

[0122] Sequence prediction module 4 is used to input the unremoved subsequence and the compensated sequence into the GRU model for prediction and output multiple prediction results. The multiple prediction results are then superimposed to generate the final prediction value.

[0123] For example, the computer program may be divided into one or more modules / units, which are stored in the memory and executed by the processor to complete the present invention. The one or more modules / units may be a series of computer program instruction segments capable of performing a specific function, which describe the execution process of the computer program in the mobile terminal.

[0124] For example, the computer program can be divided into a spurious parameter determination module 1, a spuriousness calculation module 2, a sequence processing module 3, and a sequence prediction module 4;

[0125] The specific functions of each module are as follows:

[0126] The spurious parameter determination module 1 is used to decompose the frequency modulation command sequence into multiple sub-sequences and determine the spurious parameters of each sub-sequence;

[0127] The falsehood calculation module 2 is used to determine the falsehood degree of each subsequence based on the determined falsehood parameters;

[0128] The sequence processing module 3 is used to identify the subsequence corresponding to the maximum false score based on the false score of each subsequence, remove the subsequence as the subsequence with the largest false score, and generate a compensation sequence for the removed subsequence.

[0129] Sequence prediction module 4 is used to input the unremoved subsequence and the compensated sequence into the GRU model for prediction and output multiple prediction results. The multiple prediction results are then superimposed to generate the final prediction value.

[0130] The mobile terminal can be a computing device such as a desktop computer, laptop, handheld computer, or cloud server. The mobile terminal may include, but is not limited to, a processor and memory.

[0131] The processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor. The processor is the control center of the mobile terminal, connecting various parts of the mobile terminal via various interfaces and lines.

[0132] The memory can be used to store the computer program and / or module. The processor implements various functions of the mobile terminal by running or executing the computer program and / or module stored in the memory and calling the data stored in the memory.

[0133] The memory may primarily include a program storage area and a data storage area. The program storage area may store the operating system and at least one application program required for a function (such as sound playback, image playback, etc.); the data storage area may store data created based on the use of the mobile phone (such as audio data, phonebook, etc.). Furthermore, the memory may include high-speed random access memory, and may also include non-volatile memory, such as hard disks, RAM, plug-in hard disks, SmartMediaCards (SMC), Secure Digital (SD) cards, FlashCards, at least one disk storage device, flash memory device, or other volatile solid-state storage devices.

[0134] Example 4

[0135] The present invention also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the energy storage frequency modulation prediction method for identifying spurious subsequences.

[0136] If the modules / units integrated in the mobile terminal are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium.

[0137] Based on this understanding, all or part of the processes in the above method can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium. When executed by a processor, the computer program can implement the steps of the above-described aggregated reinforcement learning resource scheduling method. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or some intermediate form.

[0138] The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording media, USB flash drive, portable hard drive, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signal, telecommunication signal, and software distribution medium, etc.

[0139] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the protection scope of the present invention.

Claims

1. A method of identifying false subsequences in a stored energy frequency modulation prediction, comprising: The method comprises the following steps: decompose the frequency modulation instruction sequence into multiple subsequences, and determine the false parameters of each subsequence; determine the false degree of each subsequence according to the determined false parameters; identify the subsequence corresponding to the maximum false degree based on the false degree of each subsequence, remove the subsequence as the subsequence with the maximum false component, and generate a compensation sequence corresponding to the removed subsequence; input the unremoved subsequences and the compensation sequence into a GRU model respectively for prediction output to obtain multiple prediction results, and superimpose the multiple prediction results to generate a final prediction value; the determination process of the false parameters of each subsequence comprises: S1, set a sine function, adjust the sine function parameters, record the data points falling into the sine function and the number of falling points corresponding to the data points in one subsequence of the frequency modulation instruction sequence according to the set falling point condition, and determine the data points not falling into the sine function as the false parameters in the corresponding subsequence; S2, determine the false parameters in the corresponding subsequence of the unremoved subsequences in the frequency modulation instruction sequence according to S1; S3, form a false parameter set corresponding to the false parameters determined for each subsequence of the frequency modulation instruction sequence; in S1, the falling point condition is that the time sequence point is substituted into the sine function, if the calculated value of the sine function is equal to the data point of the subsequence corresponding to the time sequence point, it is judged that the data point falls on the sine function, otherwise the data point does not fall on the sine function; the calculation formula of the false degree is as follows: wherein , , , false parameters of the subsequence, denotes the number of hits; || denotes the absolute value.

2. The method of claim 1, wherein, the generation process of the compensation sequence comprises: generate phase parameters randomly in a preset phase offset range based on the frequency parameters of the removed subsequences, construct a sine function sequence with the same frequency characteristics and different phases, and the sine function sequence is the compensation sequence.

3. The method of claim 1, wherein the method further comprises: the calculation formula of the compensation sequence generation comprises: wherein denotes a parameter of the make-up sequence; denotes a parameter of the culled sub-sequence; denotes a minimum spurious degree; denotes a maximum spurious degree; denotes an absolute value.

4. The method of claim 1, wherein, the multiple prediction results are fused and superimposed to generate the final prediction value by weighted summation.

5. An energy storage frequency modulation prediction system that identifies spurious subsequences, characterized by, A method for identifying false subsequences in energy storage frequency modulation prediction, comprising: a false parameter determination module for decomposing the frequency modulation instruction sequence into multiple subsequences and determining the false parameters of each subsequence; a false degree calculation module for determining the false degree of each subsequence according to the determined false parameters; a sequence processing module for identifying the subsequence corresponding to the maximum false degree based on the false degree of each subsequence, removing the subsequence as the subsequence with the maximum false component, and generating a compensation sequence corresponding to the removed subsequence; a sequence prediction module for inputting the unremoved subsequences and the compensation sequence into a GRU model respectively for prediction output to obtain multiple prediction results, and superimposing the multiple prediction results to generate a final prediction value.

6. A mobile terminal comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, The processor executes the computer program to realize the steps of the method for identifying false subsequences in energy storage frequency modulation prediction according to any one of claims 1-4.

7. A computer-readable storage medium storing a computer program, wherein the computer program comprises the following steps of: receiving a request for a resource from a client; determining whether the client is authorized to access the resource; and if the client is authorized to access the resource, providing the resource to the client. The computer program is executed by the processor to realize the steps of the method for identifying false subsequences in energy storage frequency modulation prediction according to any one of claims 1-4.

Images