Energy storage frequency modulation instruction prediction method and system based on VMD decomposition and exchange method

Abstract

The invention discloses an energy storage frequency modulation instruction prediction method and system based on a VMD decomposition and exchange method, and relates to the technical field of time sequence prediction, and the method comprises the following steps: decomposing a frequency modulation instruction Pt through VMD to obtain sub-sequences IMF; calculating a resistance value of each data point of each sub-sequence IMF to obtain a resistance sequence; and exchanging the data points of the IMFs of the different subsequences, and predicting the exchanged result to obtain a final frequency modulation instruction predicted value. According to the method, the data points are adjusted across the subsequences through the exchange method, on the premise that the combined resistance condition is met, local feature distribution of the subsequences is optimized, interference of high-resistance points on a prediction model is reduced, and adaptability of the model to a complex mode is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of time series prediction, and particularly to a prediction method and system for energy storage frequency modulation commands based on VMD decomposition and exchange method. Background Art

[0002] In the traditional frequency modulation method of hybrid energy storage (supercapacitor + lithium battery) assisting thermal power units, the difference between the frequency modulation command and the thermal power unit is transmitted to the hybrid energy storage, where the low-frequency part is borne by the battery and the high-frequency part is borne by the supercapacitor. However, the transmission of the signal (transmission of the frequency modulation command to the supercapacitor / lithium battery) takes time, and the supercapacitor or the lithium battery itself also takes time to respond, which will cause a certain response time difference, further affecting the revenue value and further affecting the revenue of the power plant. Therefore, the present invention proposes a prediction method to predict the magnitude of the frequency modulation command in advance, enabling the supercapacitor / battery to act in advance to increase the revenue value. In traditional prediction methods, the decomposed subsequences are more difficult to predict, and the non-linear degree of some subsequences is even higher than that of the original sequence, resulting in greater prediction difficulty. Summary of the Invention

[0003] In view of the problems existing in the existing energy storage frequency modulation command prediction and system based on VMD decomposition and exchange method, the present invention is proposed.

[0004] Therefore, the problem to be solved by the present invention is that in traditional prediction methods, the decomposed subsequences are more difficult to predict, and the non-linear degree of some subsequences is even higher than that of the original sequence, resulting in greater prediction difficulty.

[0005] To solve the above technical problems, the present invention provides the following technical solutions: In a first aspect, an embodiment of the present invention provides a prediction method for energy storage frequency modulation commands based on VMD decomposition and exchange method, which includes the following steps: Decompose the frequency modulation command Pt by VMD to obtain subsequences IMF; Calculate the resistance value for each data point of each subsequence IMF to obtain a resistance sequence; Exchange the data points of different subsequences IMF, and predict the result after the exchange to obtain the final predicted value of the frequency modulation command.

[0006] As a preferred solution of the prediction method for energy storage frequency modulation commands based on VMD decomposition and exchange method of the present invention, the calculation of the resistance value includes the following steps: Project the subsequence IMF onto the Y-X coordinate system, where the Y-axis is the value and the X-axis is the data point subscript; Generate a reference curve , where Z is the median of the subsequence, J is the average value, [•] represents rounding, and N represents the length of the subsequence; For each data point X ii , draw a vertical line downward to intersect the reference curve at point O, and calculate the area of the triangle formed by the tangent at point O and the X ii vertical line as the resistance value.

[0007] As a preferred scheme of the energy storage frequency modulation command prediction method based on VMD decomposition and exchange method of the present invention, wherein: when data points are exchanged, the following conditions are satisfied simultaneously, The exchanged data points come from different subsequences; The combined resistance value satisfies: And , where, is the resistance value of the m-th data point in the i-th subsequence, is the resistance value of the (m - 1)-th data point in the i-th subsequence, is the resistance value of the n-th data point in the j-th subsequence, is the resistance value of the (n - 1)-th data point in the j-th subsequence.

[0008] As a preferred scheme of the energy storage frequency modulation command prediction method based on VMD decomposition and exchange method of the present invention, wherein: when predicting the result after exchange, the specific steps are as follows, Input the exchanged subsequences into the GRU network for independent prediction respectively, and add the prediction results of each subsequence to obtain the final predicted value of the frequency modulation command.

[0009] As a preferred scheme of the energy storage frequency modulation command prediction method based on VMD decomposition and exchange method of the present invention, wherein: when the data points are exchanged, when there are multiple data points that meet the conditions within the same subsequence, the following judgment is made: Obtain the absolute values of all data points that meet the conditions; Judge the maximum value among the absolute values of all data points, and arrange the maximum value in the first echelon.

[0010] As a preferred scheme of the energy storage frequency modulation command prediction method based on VMD decomposition and exchange method of the present invention, wherein: when the reference curve is generated, when the median Z or the average value J of the subsequence is zero, it is replaced by the standard deviation of the subsequence.

[0011] As a preferred scheme of the energy storage frequency modulation command prediction method based on VMD decomposition and exchange method of the present invention, wherein: when decomposing through VMD and obtaining the subsequence IMF, normalize the subsequence IMF.

[0012] Second aspect, an embodiment of the present invention provides an energy storage frequency modulation command prediction system based on VMD decomposition and exchange method, which includes a VMD decomposition module, a resistance calculation module, a data exchange module, and a prediction module; The VMD decomposition module is responsible for decomposing the frequency modulation command Pt into multiple subsequences IMF; The resistance calculation module calculates the data points in each subsequence IMF; The data exchange module determines whether the data points between different subsequence IMFs meet the exchange condition according to the calculation result of the resistance value; The prediction module inputs the subsequences after exchange processing into the prediction model to obtain the final predicted value of the frequency modulation command.

[0013] Third aspect, an embodiment of the present invention provides a computer device, including a memory and a processor, where the memory stores a computer program, and: when the processor executes the computer program, any step of the above-mentioned energy storage frequency modulation command prediction method based on VMD decomposition and exchange method is implemented.

[0014] Fourth aspect, an embodiment of the present invention provides a computer-readable storage medium, on which a computer program is stored, and: when the computer program is executed by a processor, any step of the above-mentioned energy storage frequency modulation command prediction method based on VMD decomposition and exchange method is implemented.

[0015] The beneficial effects of the present invention are: Ensure the mathematical rigor of decomposition through constrained variational optimization, and effectively avoid mode mixing; combine the resistance method to dynamically evaluate the prediction difficulty (resistance value) of each subsequence data point, and provide a quantitative basis for subsequent optimization.

[0016] Adjust the data points across subsequences through the exchange method. On the premise of meeting the combined resistance condition, optimize the local feature distribution of the subsequences, reduce the interference of high-resistance points on the prediction model, and improve the adaptability of the model to complex patterns.

[0017] Input the optimized subsequences into the GRU network for independent prediction, make full use of its gating mechanism to capture long-term dependence relationships, and fuse the prediction advantages of each subsequence through result superposition, ultimately achieving error dispersion and accuracy improvement.

[0018] Introduce a fault tolerance mechanism and normalization preprocessing to ensure the stability of the resistance value calculation and enhance the applicability of the method in different noise environments. Description of the Drawings

[0019] 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 drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0020] Figure 1 It is a flowchart of a method for predicting energy storage frequency modulation commands based on VMD decomposition and exchange method.

[0021] Figure 2 It is a schematic diagram of the YX coordinate axes of a method for predicting energy storage frequency modulation commands based on VMD decomposition and exchange method.

[0022] Figure 3 It is a schematic diagram of a reference curve of a method for predicting energy storage frequency modulation commands based on VMD decomposition and exchange method.

[0023] Figure 4 It is a VMD decomposition flowchart of a method for predicting energy storage frequency modulation commands based on VMD decomposition and exchange method. Specific Embodiments

[0024] To make the above objects, features, and advantages of the present invention more obvious and understandable, the following will describe the specific embodiments of the present invention in detail with reference to the 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.

[0025] In the following description, many specific details are set forth to fully understand the present invention. However, the present invention can also be implemented in other ways different from those described herein. Those skilled in the art can make similar generalizations without departing from the connotation of the present invention. Therefore, the present invention is not limited by the specific embodiments disclosed below.

[0026] It should be noted that: IMF is the Intrinsic Mode Function, VMD is the Variational Mode Decomposition, and IMF (Intrinsic Mode Function) and VMD (Variational Mode Decomposition) are two important concepts in the field of signal processing.

[0027] Embodiment 1, referring to Figures 1 to 4 , which is the first embodiment of the present invention. This embodiment provides a method for predicting energy storage frequency modulation commands based on VMD decomposition and exchange method, including the following steps. S1. Decompose the frequency modulation command Pt through VMD to obtain the subsequence IMF.

[0028] Refer to Figure 4 As shown, it should be noted that when decomposing through VMD to obtain the subsequence IMFs, the subsequence IMFs are normalized.

[0029] In this embodiment, the input signal is: the frequency modulation command Pt (time series data with a length of N).

[0030] Moreover, the decomposition parameter settings include the number of modes K: selected according to the signal characteristics (for example, set to K = 5 through spectral analysis or experience); the penalty factor α: default value is 2000, used to balance the bandwidth constraint and data fidelity; the convergence tolerance: set to 1e - 6 to ensure the decomposition accuracy; the maximum number of iterations: 1000 times to avoid infinite loops.

[0031] Call the VMD algorithm library (such as vmdpy in Python or the MATLAB toolbox), input Pt and the above parameters, and obtain the K decomposed subsequences IMF1, IMF2,..., IMF K ; When outputting, the set output requirement is that the length of each IMF is the same as the original signal (N points), and it satisfies orthogonality and band separation characteristics.

[0032] Suppose Pt becomes IMF1, IMF2, IMF3, IMF i ,..., IMF K , and their subsequences are: .

[0033] As Figure 2 shown, project this subsequence onto the Y - X coordinate axes, where the Y - axis represents the magnitude of the value and the horizontal axis is the subscript.

[0034] Perform independent normalization on each IMF subsequence to ensure the numerical stability of the subsequence, provide reliable input for subsequent resistance value calculation and exchange optimization, and at the same time retain the characteristics of the original signal to avoid information loss.

[0035] S2. Calculate the resistance value for each data point of each subsequence IMF to obtain the resistance sequence.

[0036] The corresponding resistance sequence is: ; K represents the number of subsequences, i represents the number of the i - th subsequence, N is the length of the subsequence, and S represents the resistance value.

[0037] Refer to Figure 3 shown, Project the subsequence IMF onto the Y - X coordinate system, with the Y - axis being the value and the X - axis being the data point subscript; Generate a reference curve , where Z is the median of the subsequence, J is the average value, [•] represents rounding, and N represents the length of the subsequence; For each data point X ii , draw a vertical line downward to intersect the reference curve at point O, and calculate the area of the triangle formed by the tangent at point O and the X ii vertical line as the resistance value, and the positive and negative signs are determined by whether X ii is above or below the reference curve. If it is above the reference curve, it is a positive sign; if it is below the reference curve, it is a negative sign. That is, the point is positive above, and vice versa negative. When generating the reference curve, when the median Z or the average value J of the subsequence is zero, it is replaced by the standard deviation of the subsequence.

[0038] Assume that for a certain subsequence of IMF i , the parameters are: Z = 0.5, J = 0.3, N = 10, and the current calculation point i = 2, X ii = 1.2; Reference curve Y value: ; Tangent slope: ; Vertical line length: ; Resistance value: ; Because , take a positive value.

[0039] When the data points are exchanged, the following conditions are satisfied simultaneously: The exchanged data points come from different subsequences; The combined resistance value satisfies: and , where is the resistance value of the mth data point in the ith subsequence, is the resistance value of the (m - 1)th data point in the ith subsequence, is the resistance value of the nth data point in the jth subsequence, is the resistance value of the (n - 1)th data point in the jth subsequence.

[0040] S3. Exchange the data points of the IMF of different subsequences, and predict the result after the exchange to obtain the final predicted value of the frequency modulation command.

[0041] When predicting the result after the exchange, the specific steps are as follows: Input the exchanged subsequences into the GRU network for independent prediction respectively, and add the prediction results of each subsequence to obtain the final predicted value of the frequency modulation command.

[0042] When the data points are exchanged, if there are multiple data points that meet the conditions within the same subsequence, the following judgment is made: Obtain the absolute values of all data points that meet the conditions; Judge the maximum value among the absolute values of all data points, and arrange the maximum value in the first echelon.

[0043] In this embodiment, the GRU network structure includes an input layer: univariate time series (each subsequence is input independently); a hidden layer: 2 layers of GRU units, with 64 neurons in each layer; an output layer: a fully connected layer, outputting the predicted values for the next T steps; Training parameters: Loss function: mean squared error (MSE); Optimizer: Adam (learning rate = 0.001); Batch size: 32; Early stopping mechanism: If the validation loss does not decrease for 5 consecutive rounds, terminate the training.

[0044] Input each subsequence into an independent GRU model to obtain the prediction results, sum the prediction results of all subsequences, and obtain the final frequency modulation command prediction value: ; Among them, is the prediction result, is the final frequency modulation command prediction value.

[0045] If there is no exchange point that meets the conditions in a certain subsequence, retain the original subsequence and directly predict.

[0046] Exchange conflict: If multiple subsequences compete for the same external point, give priority to the subsequence with a greater absolute value of resistance. When verifying the prediction results, calculate the prediction errors (such as MAE, RMSE); compare the errors before and after the exchange to verify the optimization effect.

[0047] This embodiment also provides an example calculation: Original subsequences: IMF1 = [0.2, -0.5, 1.1, -0.3], IMF2 = [0.8, -1.2, 0.4, 0.6]; Resistance sequences: S1 = [0.3, -0.7, 1.3, -0.4], S2 = [0.9, -1.5, 0.5, 0.7]; Candidate exchange pairs: (IMF1[2], IMF2[1]): ; Comprehensive resistance value: 1.3 + (-1.5) - 1 = -1.2 (does not meet the conditions, discard); (IMF1[3], IMF2[4]): ; Comprehensive resistance value: -0.4 + 0.7 - 1 = -0.7 (meets the conditions).

[0048] Swap IMF1[3] and IMF2[4] to obtain: IMF1' = [0.2, -0.5, 1.1, 0.6], IMF2' = [0.8, -1.2, 0.4, -0.3]; Input IMF1' and IMF2', and add the prediction results.

[0049] In this embodiment, through the priority sorting of the absolute value of resistance, the priority exchange of high-resistance points is ensured, effectively reducing the local interference of the prediction model; combined with the independent prediction of GRU and result fusion, the accuracy and stability of the frequency modulation command prediction are significantly improved.

[0050] In summary, by constraining variational optimization to ensure the mathematical rigor of decomposition, modal aliasing is effectively avoided; combined with the resistance method to dynamically evaluate the prediction difficulty (resistance value) of each subsequence data point, a quantitative basis is provided for subsequent optimization. Through the exchange method, data points are adjusted across subsequences. On the premise of meeting the combined resistance condition, the local feature distribution of the subsequence is optimized, the interference of high-resistance points on the prediction model is reduced, and the adaptability of the model to complex patterns is improved. The optimized subsequences are input into the GRU network for independent prediction, making full use of its gating mechanism to capture long-term dependence relationships, and the prediction advantages of each subsequence are fused through result superposition, ultimately achieving error dispersion and accuracy improvement. The introduction of a fault tolerance mechanism and normalization preprocessing ensures the stability of resistance value calculation and enhances the applicability of the method in different noise environments.

[0051] Embodiment 2, based on the first embodiment, this embodiment further provides an energy storage frequency modulation command prediction system based on VMD decomposition and the exchange method, including a VMD decomposition module, a resistance calculation module, a data exchange module, and a prediction module; The VMD decomposition module is responsible for decomposing the frequency modulation command Pt into multiple subsequences IMF; The resistance calculation module calculates the data points in each subsequence IMF; The data exchange module determines whether the data points between different subsequence IMFs meet the exchange conditions according to the calculation results of the resistance values; The prediction module inputs the subsequences after exchange processing into the prediction model to obtain the final predicted value of the frequency modulation command.

[0052] This embodiment also provides a computer device applicable to the case of the energy storage frequency modulation command prediction method based on VMD decomposition and the exchange method, including a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement the energy storage frequency modulation command prediction method based on VMD decomposition and the exchange method as proposed in the above embodiment.

[0053] The computer device may be a terminal, 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 computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The communication interface of the computer device is used to communicate with external terminals in a wired or wireless manner, and the wireless manner can be implemented through WIFI, a carrier network, NFC (Near Field Communication), or other technologies. 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 a button, a trackball, or a touchpad provided on the housing of the computer device, or an external keyboard, touchpad, or mouse, etc.

[0054] This embodiment also provides a storage medium, on which a computer program is stored. When the program is executed by a processor, it implements the method for predicting energy storage frequency modulation instructions based on VMD decomposition and exchange method as proposed in the above embodiment.

[0055] The storage medium proposed in this embodiment and the data storage method proposed in the above embodiment belong to the same inventive concept. Technical details not described in detail in this embodiment can be referred to the above embodiment, and this embodiment has the same beneficial effects as the above embodiment.

[0056] Embodiment 3, based on the first two embodiments, this embodiment provides a method for predicting energy storage frequency modulation instructions based on VMD decomposition and exchange method.

[0057] In order to further verify the advantages of the present invention, the present invention uses the method of the present invention and the method of GRU prediction to predict the frequency modulation sequence respectively. The results are shown in Table 1. Table 1: Schematic table of frequency modulation sequence prediction results , There are also four evaluation indexes, and the results are shown in Table 2. Table 2: Schematic table of evaluation indexes , In this embodiment, N represents the sample size. and respectively represent the actual value and the predicted value at time n. It can be seen from the experimental results that all four evaluation indexes have decreased, indicating that the proposed model can well improve the prediction accuracy.

[0058] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit them. 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 all of them should be covered by the scope of the claims of the present invention.

Claims

1. A prediction method for energy storage frequency modulation instructions based on VMD decomposition and exchange method, characterized in that: It includes the following steps: Decompose the frequency modulation command Pt through VMD to obtain subsequences IMF; Calculate the resistance value for each data point of each subsequence IMF to obtain a resistance sequence; Exchange the data points of different subsequences IMF, and predict the result after the exchange to obtain the final predicted value of the frequency modulation command.

2. The energy storage frequency modulation command prediction method based on VMD decomposition and exchange method according to claim 1, wherein: The calculation of the resistance value includes the following steps: Project the subsequence IMF onto the Y-X coordinate system, with the Y-axis being the value and the X-axis being the data point subscript; Generate a reference curve , where Z is the median of the subsequence, J is the average value, [•] represents rounding, and N represents the length of the subsequence; For each data point X ii , draw a vertical line downward to intersect the reference curve at point O, and calculate the area of the triangle formed by the tangent at point O and the X ii vertical line as the resistance value, where the positive and negative signs are determined by whether X ii is above or below the reference curve.

3. The energy storage frequency modulation command prediction method based on VMD decomposition and exchange method according to claim 2, wherein: When exchanging the data points, the following conditions are satisfied simultaneously: The exchanged data points come from different subsequences; The resistance value satisfies: and , where is the resistance value of the m-th data point in the i-th subsequence, is the resistance value of the (m - 1)-th data point in the i-th subsequence, is the resistance value of the n-th data point in the j-th subsequence, is the resistance value of the (n - 1)-th data point in the j-th subsequence.

4. The energy storage frequency modulation command prediction method based on VMD decomposition and exchange method according to claim 3, characterized in that: When predicting the result after the exchange, the specific steps are as follows: Input the exchanged subsequences into the GRU network for independent prediction respectively, and add the prediction results of each subsequence to obtain the final predicted value of the frequency modulation command.

5. A prediction method for energy storage frequency modulation instructions based on VMD decomposition and exchange method, characterized in that: When exchanging the data points, when there are multiple data points that meet the conditions within the same subsequence, the following judgment is made: Obtain the absolute values of all data points that meet the conditions; Judge the maximum value among the absolute values of all data points, and arrange the maximum value in the first echelon.

6. The energy storage frequency modulation command prediction method based on VMD decomposition and exchange method according to claim 5, characterized in that: When generating the reference curve, when the median Z or the average value J of the subsequence is zero, it is replaced by the standard deviation of the subsequence.

7. A prediction method for energy storage frequency modulation command based on VMD decomposition and exchange method as claimed in claim 6, characterized in that: When decomposing through VMD to obtain the subsequence IMF, normalize the subsequence IMF.

8. A prediction system for energy storage frequency modulation commands based on VMD decomposition and exchange method, based on the prediction method for energy storage frequency modulation commands based on VMD decomposition and exchange method according to any one of claims 1 to 7, characterized in that: It includes a VMD decomposition module, a resistance calculation module, a data exchange module, and a prediction module; The VMD decomposition module is responsible for decomposing the frequency modulation command Pt into multiple subsequences IMF; The resistance calculation module calculates for the data points in each subsequence IMF; The data exchange module judges whether the data points between different subsequences IMF meet the exchange conditions according to the calculation result of the resistance value; The prediction module inputs the subsequence after the exchange process into the prediction model to obtain the final predicted value of the frequency modulation command.

9. A computer device, comprising a memory and a processor, the memory storing a computer program, characterized in that: When the processor executes the computer program, it implements the steps of a method for predicting energy storage frequency modulation commands based on VMD decomposition and exchange method according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by the processor, it implements the steps of a method for predicting energy storage frequency modulation commands based on VMD decomposition and exchange method according to any one of claims 1 to 7.

Images