An improved supercapacitor energy storage frequency modulation and grid connection method and system

Abstract

The present invention discloses an improved supercapacitor energy storage frequency modulation grid connection method and system, belonging to the technical field of wind speed prediction. The method includes: decomposing the original wind speed sequence of the wind farm using variational mode decomposition according to a preset decomposition layer number, calculating the sequence correlation degrees of all subsequences in the original wind speed subsequence set, taking the reference subsequence as a reference, and adjusting the original subsequence set according to the sequence correlation degrees to obtain an adjusted subsequence set; fusing the adjusted subsequence set with the residual sequence to obtain a corrected subsequence set; calculating the sequence correlation of the corrected subsequence set to obtain the correlation at the current decomposition layer number; increasing the decomposition layer number of the variational mode decomposition, repeating the steps, taking the corrected subsequence set at the decomposition layer number corresponding to the minimum correlation as the transformed subsequence, comparing the maximum correlation degree of the residual sequence in the transformed subsequence with a threshold value to obtain the optimal wind speed subsequence; and using the optimal wind speed subsequence for wind speed prediction.

Description

Technical Field

[0001] The present invention belongs to the technical field of wind speed prediction, and particularly relates to an improved supercapacitor energy storage frequency modulation grid connection method and system. Background Art

[0002] Due to the randomness of wind speed changes, the grid connection of large-scale wind farms brings difficulties to power grid dispatching, thereby affecting the stability of the system. In order to improve the dispatchability of wind farms, the general method is to use the neural network method to predict the wind speed for the next day, calculate the power sent by the wind farm to the power grid through the predicted wind speed value, and submit the predicted value to the power grid dispatching agency. At the same time, a supercapacitor energy storage system is connected at the outlet of the wind farm to quickly respond to make up for the error between the actual power generated by the wind farm and the predicted power, thereby improving the credibility of the power grid to dispatch the power generation of the wind farm based on wind power prediction and improving the coordinated operation ability between the wind farm and the power system.

[0003] However, the traditional prediction method generally uses VMD (Variational Mode Decomposition) to decompose the original data and then puts it into the neural network for prediction. However, there is a problem that the subsequences after the existing variational mode decomposition are still correlated with each other, that is, the correlation between the subsequences after VMD decomposition is relatively high, and the subsequences are not independent enough from each other. Putting the subsequences that are not independent enough into the neural network for prediction will lead to inaccurate prediction results. Summary of the Invention

[0004] The purpose of the present invention is to overcome the problem that the subsequences after the existing variational mode decomposition have relatively high correlation, and proposes an improved supercapacitor energy storage frequency modulation grid connection method and system.

[0005] To achieve the above object, the present invention adopts the following technical solutions:

[0006] In the first aspect, the present invention provides an improved supercapacitor energy storage frequency modulation grid connection method, including the following steps:

[0007] S1. Obtain the original wind speed sequence of the wind farm;

[0008] S2. Perform variational mode decomposition on the original wind speed sequence of the wind farm based on a preset decomposition layer number to obtain a set of original wind speed subsequences;

[0009] S3. Calculate the sequence correlation degrees of all subsequences in the set of original wind speed subsequences, and compare the sequence correlation degrees of all subsequences to obtain the minimum correlation degree;

[0010] S4. Take the subsequence corresponding to the minimum correlation as the reference subsequence. Based on the reference subsequence, adjust the original subsequence set according to the sequence correlation to obtain the adjusted subsequence set;

[0011] S5. Compare the original subsequence set with the adjusted subsequence set to obtain the residual sequence, and fuse the adjusted subsequence set with the residual sequence to obtain the corrected subsequence set;

[0012] S6. Calculate the sequence correlation of the corrected subsequence set to obtain the correlation at the current decomposition level;

[0013] S7. Return to S2, increase the decomposition level of the variational mode decomposition, and repeat steps S2 - S6 until the decomposition level of the variational mode decomposition reaches the maximum, obtain the correlations at several decomposition levels, and compare the correlations at several decomposition levels to obtain the minimum correlation;

[0014] S8. Take the corrected subsequence set at the decomposition level corresponding to the minimum correlation as the transformed subsequence, compare the maximum correlation of the residual sequence in the transformed subsequence with the threshold. If the maximum correlation of the residual sequence in the transformed subsequence is less than the threshold, retain the residual sequence in the transformed subsequence as the optimal wind speed subsequence;

[0015] S9. Use the optimal wind speed subsequence to perform wind speed prediction to obtain the wind speed prediction sequence, and add up the wind speed prediction sequences to obtain the wind speed prediction result for the grid connection of the over-capacity energy storage.

[0016] Furthermore, the original wind speed sequence of the wind farm includes the historical wind speed data of the wind farm, and the number of subsequences in the original subsequence set in S2 is the decomposition level of the variational mode decomposition.

[0017] Furthermore, the calculation of the sequence correlation is carried out using the following formula:

[0018]

[0019] where, D is the sequence correlation, is the sequence correlation between the i -1th sequence and the i th sequence, j is the number of elements in the i -1th sequence that satisfy the correlation relation formula, N is the number of elements in the i -1th sequence and the i th sequence, i is a positive integer, 1 < i ≤ 30;

[0020] The correlation relation formula is b m=α a m , where a m is an element in the i -1th sequence, b m is an element in the i th sequence, α is the relevance relationship function, 0.5 < α < 1.5.

[0021] Furthermore, based on the reference subsequence, the original subsequence set is adjusted according to the sequence relevance to obtain the adjusted subsequence set, and the following adjustment formula is used:

[0022]

[0023]

[0024] where is the adjusted element in the subsequence to be adjusted, sigmoid( ) represents the activation function, is an element in the subsequence to be adjusted, represents the augmentation coefficient, K is the decomposition level; D min is the sequence relevance of the reference subsequence, D x is the sequence relevance of the subsequence to be adjusted, max( D 1 ,D 2 ,…,D K ) represents the maximum value of the sequence relevance of the original subsequence set, min( s 1 ,s 2 ,..,s i ,...,s N ) represents the minimum element in the reference subsequence, max( r 1 ,r 2 ,..,r i ,..., r N ) represents the maximum element in the subsequence to be adjusted, N represents the total number of elements in the sequence, avg( r 1 ,r 2 ,..,ri ,..., r N ) represents the average value of the elements in the subsequence to be adjusted, e is the natural constant, π is a constant, and i is an intermediate variable.

[0025] Further, the residual sequence obtained by comparing the original subsequence set and the adjusted subsequence set is performed using the following comparison formula:

[0026]

[0027] where, is the residual sequence, is the original subsequence set, is the adjusted subsequence set, K is the decomposition level, and i is an intermediate variable;

[0028] The sequence correlation of the corrected subsequence set in S6 is the average value of the correlation degrees of the adjusted subsequence set and the residual sequence.

[0029] Further, the maximum correlation degree of the residual sequence in the transformed subsequence is compared with the threshold. If the maximum correlation degree of the residual sequence in the transformed subsequence is greater than the threshold, the residual sequence in the transformed subsequence is decomposed using variational mode decomposition until the maximum correlation degree of the residual sequence in the transformed subsequence is less than the threshold, and the optimal wind speed subsequence is obtained.

[0030] Further, the wind speed prediction in S9 adopts the IPSO-SVR model, and the optimal wind speed subsequence is input into the IPSO-SVR model in sequence to obtain the wind speed prediction sequence.

[0031] In the second aspect, the present invention provides an improved supercapacitor energy storage frequency modulation grid-connected system, based on the improved supercapacitor energy storage frequency modulation grid-connected method described above, including:

[0032] The original wind speed sequence acquisition module is used to acquire the original wind speed sequence of the wind farm;

[0033] The variational mode decomposition module is used to perform variational mode decomposition on the original wind speed sequence of the wind farm based on a preset decomposition level to obtain the original wind speed subsequence set;

[0034] The sequence correlation calculation module is used to calculate the sequence correlation of all subsequences in the original wind speed subsequence set, and compare the sequence correlations of all subsequences to obtain the minimum correlation degree;

[0035] An original subsequence set adjustment module is used to take the subsequence corresponding to the minimum correlation degree as the reference subsequence, and based on the reference subsequence, adjust the original subsequence set according to the sequence correlation degree to obtain an adjusted subsequence set;

[0036] A residual sequence fusion module is used to compare the original subsequence set and the adjusted subsequence set to obtain a residual sequence, and fuse the adjusted subsequence set and the residual sequence to obtain a corrected subsequence set;

[0037] A sequence correlation calculation module is used to calculate the sequence correlation of the corrected subsequence set to obtain the correlation at the current decomposition level;

[0038] A repeated decomposition module is used to increase the decomposition level of the variational mode decomposition until the decomposition level of the variational mode decomposition reaches the maximum, obtain the correlations at several decomposition levels, and compare the correlations at several decomposition levels to obtain the minimum correlation value;

[0039] A module for comparing and obtaining the optimal wind speed subsequence is used to take the corrected subsequence set at the decomposition level corresponding to the minimum correlation value as the transformed subsequence, compare the maximum correlation degree of the residual sequence in the transformed subsequence with a threshold value. If the maximum correlation degree of the residual sequence in the transformed subsequence is less than the threshold value, then retain the residual sequence in the transformed subsequence as the optimal wind speed subsequence;

[0040] A module for obtaining the wind speed prediction result is used to perform wind speed prediction using the optimal wind speed subsequence to obtain a wind speed prediction sequence, and add the wind speed prediction sequences to obtain the wind speed prediction result for the supercapacitor energy storage grid connection;

[0041] In a third aspect, the present invention provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the improved supercapacitor energy storage frequency modulation grid connection method described above.

[0042] In a fourth aspect, the present invention provides a computer-readable storage medium. The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, it implements the improved supercapacitor energy storage frequency modulation grid connection method described above.

[0043] Compared with the prior art, the present invention has the following beneficial technical effects:

[0044] An improved supercapacitor energy storage frequency modulation grid connection method proposed by the present invention provides an improved variational mode decomposition (VMD) method, which can make the decomposed subsequences independent of each other, thereby improving the prediction accuracy, more effectively decomposing the original wind speed sequence, and obtaining a series of relatively stable intrinsic mode functions (IMFs) with different characteristic information. Compared with the original sequence, these components can better reflect the internal laws and characteristics of wind speed changes. By calculating the sequence correlation degree of the subsequences and adjusting the original subsequence set to reduce the correlation between the subsequences, making the subsequences more independent, the neural network can more accurately capture the characteristics of each subsequence, thereby improving the accuracy of wind speed prediction. By continuously increasing the decomposition layer number of variational mode decomposition and comparing the correlations under different decomposition layer numbers, the decomposition layer number with the smallest correlation can be found, which can determine the optimal decomposition layer number and further improve the prediction accuracy. By comparing the maximum correlation degree of the residual sequence with the threshold value to decide whether to retain the residual sequence, the residual information that has an important impact on the prediction result can be retained, while avoiding the introduction of irrelevant noise, and the multi-scale wind speed information can be fused. The neural network can more comprehensively understand the characteristics of wind speed data, improving the prediction robustness and accuracy. The wind speed prediction result obtained by the method of the present invention can more effectively guide the charge and discharge strategy of the supercapacitor energy storage system, thereby suppressing the power fluctuation in wind power generation and improving the overall performance and stability of the supercapacitor energy storage grid connection. The present invention improves the accuracy and robustness of wind speed prediction by optimizing the correlation problem between subsequences, providing strong support for the optimal operation of the supercapacitor energy storage system.

[0045] Further, the optimal subsequences obtained by decomposition are put into an Improved Particle Swarm Optimization - Support Vector Regression (IPSO-SVR) model for prediction. The IPSO-SVR model combines the advantages of particle swarm optimization and support vector regression, which can further improve the accuracy and generalization ability of the prediction model. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] The drawings described herein are for illustrative purposes only and are not intended to limit the scope of the present disclosure in any way. Additionally, the shapes and proportional dimensions of the components in the figures are only schematic for facilitating the understanding of the present invention and do not specifically limit the shapes and proportional dimensions of the components of the present invention. In the drawings:

[0047] Figure 1 is a flowchart of an improved supercapacitor energy storage frequency modulation grid connection method of the present invention.

[0048] Figure 2This is the structural diagram of an improved supercapacitor energy storage frequency modulation and grid connection system of the present invention.

[0049] Figure 3 This is the electronic device diagram of an improved supercapacitor energy storage frequency modulation and grid connection method of the present invention. Detailed implementation manners

[0050] In order to enable those skilled in the art to better understand the solution of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the 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.

[0051] Embodiment 1

[0052] See Figure 1 , an improved supercapacitor energy storage frequency modulation and grid connection method, including the following steps:

[0053] S1. Obtain the original wind speed sequence of the wind farm;

[0054] The original wind speed sequence in S1 includes the historical wind speed data of the wind farm;

[0055] S2. Decompose the original wind speed sequence of the wind farm using variational mode decomposition according to the preset decomposition layer number to obtain the original wind speed subsequence set;

[0056] The number of subsequences in the original subsequence set in S2 is the decomposition layer number of the variational mode decomposition;

[0057] S3. Calculate the sequence correlation degrees of all subsequences in the original wind speed subsequence set, and compare the sequence correlation degrees of all subsequences to obtain the minimum correlation degree;

[0058] The calculation of the sequence correlation degree in S3 is carried out using the following formula:

[0059]

[0060] Wherein, D is the sequence correlation degree, is the sequence correlation degree between the i -1th sequence and the i th sequence, j is the number of elements in the i -1th sequence that satisfy the correlation degree relationship, N is the number of elements in the i -1th sequence and the i th sequence, i is a positive integer, 1 < i≤30;

[0061] The relevance relationship formula is b m =α a m , where a m is the element in the i -1th sequence, b m is the element in the i th sequence, α is the relevance relationship function, 0.5 < α < 1.5;

[0062] The relevance calculation formula provided in this embodiment can quantify the similarity or relevance between two sequences, providing a reliable basis for subsequent sequence correction and optimal subsequence selection.

[0063] S4. Use the subsequence corresponding to the minimum relevance as the reference subsequence, and based on the reference subsequence, adjust the original subsequence set according to the sequence relevance to obtain the adjusted subsequence set;

[0064] In this embodiment, the selection of the reference subsequence is based on the minimum relevance, which means that this subsequence has the greatest difference from other subsequences. Therefore, it contains the most unique information or noise. By correcting other subsequences based on this reference subsequence, the noise and redundant information in these subsequences can be removed or reduced, thereby improving the quality; the corrected subsequences are closer to the true wind speed change characteristics, so the correlation between them will be enhanced. By calculating the correlation between the residual sequence of the corrected subsequence and the corresponding decomposition level, the quality of the subsequences under different decomposition levels can be evaluated, and the optimal decomposition level can be selected to balance the decomposition accuracy and calculation efficiency.

[0065] In S4, based on the reference subsequence, adjust the original subsequence set according to the sequence relevance to obtain the adjusted subsequence set, and use the following adjustment formula:

[0066]

[0067]

[0068] where is the adjusted element in the subsequence to be adjusted, sigmoid( ) represents the activation function, is the element in the subsequence to be adjusted, represents the augmentation coefficient, K is the decomposition level; D min is the sequence relevance of the reference subsequence, D x is the sequence relevance of the subsequence to be adjusted, max(D 1 ,D 2 ,…,D K ) represents the maximum value of the sequence correlation degree of the original subsequence set, min( s 1 ,s 2 ,..,s i ,...,s N ) represents the minimum element in the reference subsequence, max( r 1 ,r 2 ,.., r i ,...,r N ) represents the maximum element in the subsequence to be adjusted, N represents the total number of elements in the sequence, avg( r 1 ,r 2 ,.., r i ,...,r N ) represents the average value of the elements in the subsequence to be adjusted, e is the natural constant, π is a constant, i is an intermediate variable;

[0069] The adjustment formula provided in this embodiment uses the sigmoid activation function and the augmentation coefficient to precisely adjust the subsequence to be adjusted, and can optimize other subsequences based on the subsequence with the minimum correlation degree, that is, the reference subsequence, so as to reduce the influence of noise and outliers and improve the overall quality of the subsequence.

[0070] S5. Compare the original subsequence set with the adjusted subsequence set to obtain the residual sequence, and fuse the adjusted subsequence set with the residual sequence to obtain the corrected subsequence set;

[0071] In S5, comparing the original subsequence set with the adjusted subsequence set to obtain the residual sequence is performed using the following comparison formula:

[0072]

[0073] Among them, is the residual sequence, is the original subsequence set, is the adjusted subsequence set, K is the decomposition level,i is an intermediate variable;

[0074] S6. Calculate the sequence correlation of the corrected subsequence set to obtain the correlation at the current decomposition level;

[0075] The sequence correlation of the corrected subsequence set in S6 is the average of the correlation degrees of the adjusted subsequence set and the residual sequence;

[0076] The residual sequence calculation formula provided in this embodiment can quantify the difference between the adjusted subsequence and the original subsequence, which helps to evaluate the adjustment effect. At the same time, by calculating the average of the correlation degrees of the adjusted subsequence and the residual sequence, the quality of the subsequences at different decomposition levels can be further evaluated, providing a basis for selecting the optimal subsequence.

[0077] S7. Increase the decomposition level of the variational mode decomposition, repeat steps S2 - S6 until the decomposition level of the variational mode decomposition reaches the maximum, obtain the correlations at several decomposition levels, and compare the correlations at several decomposition levels to obtain the minimum correlation;

[0078] S8. Take the corrected subsequence set at the decomposition level corresponding to the minimum correlation as the transformed subsequence, compare the maximum correlation degree of the residual sequence in the transformed subsequence with the threshold. If the maximum correlation degree of the residual sequence in the transformed subsequence is less than the threshold, retain the residual sequence in the transformed subsequence to obtain the optimal wind speed subsequence;

[0079] The selection criterion for the optimal subsequence in this embodiment not only considers the correlation degree of the adjusted subsequence, but also considers the correlation degree of its residual sequence with the remaining sequences. This selection method can ensure that the selected subsequence is both representative and has a certain degree of independence, providing a more accurate and stable input for subsequent wind speed prediction. Through the above processing and the selection of the optimal subsequence, the noise and outlier interference in wind speed prediction can be significantly reduced, the prediction accuracy and stability can be improved, which is of great significance for optimizing the charge - discharge strategy of the over - capacity energy storage system and enhancing the stability and reliability of the grid - connected system.

[0080] In S8, compare the maximum correlation degree of the residual sequence in the transformed subsequence with the threshold. If the maximum correlation degree of the residual sequence in the transformed subsequence is greater than the threshold, decompose the residual sequence in the transformed subsequence using variational mode decomposition until the maximum correlation degree of the residual sequence in the transformed subsequence is less than the threshold to obtain the optimal wind speed subsequence;

[0081] In this embodiment, the improved VMD can more effectively decompose the original wind speed sequence, extract components with different frequencies and amplitudes, and these components reflect the internal laws and characteristics of wind speed changes. By presetting the decomposition level, a series of original subsequences can be obtained, and these subsequences are more stable in terms of frequency and amplitude, which is beneficial for subsequent prediction analysis. By adjusting the original subsequences with the benchmark subsequence, the influence of noise and outliers can be further reduced, and the quality of the subsequences can be improved. Calculating the correlation degree of the corrected subsequences and the correlation of the corresponding decomposition levels can evaluate the similarities and differences between different subsequences, so as to select the optimal subsequence for prediction.

[0082] S9. Use the optimal wind speed subsequence to perform wind speed prediction to obtain a wind speed prediction sequence, and add up the wind speed prediction sequences to obtain the wind speed prediction result for ultra-capacity energy storage grid connection;

[0083] Taking the optimal wind speed subsequence as the input of the prediction model can reduce the sensitivity of the model to noise and outliers, improve the generalization ability and robustness of the model; based on the prediction result of the optimal wind speed subsequence, the ultra-capacity energy storage system can more accurately judge when to charge and when to discharge to maximize the energy storage efficiency and reduce energy waste; accurate wind speed prediction helps the power grid dispatching department better plan power production and distribution, and reduce the problem of power supply-demand imbalance caused by wind speed changes.

[0084] In S9, the wind speed prediction adopts the IPSO-SVR model, and the optimal wind speed subsequence is sequentially input into the IPSO-SVR model to obtain a wind speed prediction sequence.

[0085] Putting the optimal subsequence into the IPSO-SVR for prediction can make full use of the advantages of the IPSO-SVR in terms of non-linear fitting and generalization ability, and improve the accuracy and stability of the prediction model. Since the optimal subsequence has removed some noise and outliers, the sensitivity of the prediction model to the input data is reduced, and the robustness of the model is improved.

[0086] The method in this embodiment that uses the improved VMD to decompose the original wind speed sequence to obtain the optimal subsequence and sequentially puts the optimal subsequence into the IPSO-SVR for prediction has significant benefits in improving the accuracy and stability of wind speed prediction, enhancing the reliability and economy of the wind power grid connection system, and promoting the innovation and development of wind power technology. Through the improved VMD algorithm, the original wind speed sequence is more effectively decomposed into components with different frequencies and amplitudes, and these components reflect the internal laws and characteristics of wind speed changes. This decomposition helps to remove noise and outliers, improve the quality of the subsequences, and thus improve the accuracy of wind speed prediction.

[0087] Embodiment 2

[0088] See Figure 2, an improved supercapacitor energy storage frequency modulation grid-connected system, using an improved supercapacitor energy storage frequency modulation grid-connected method described in Embodiment 1, including:

[0089] An original wind speed sequence acquisition module, configured to acquire the original wind speed sequence of the wind farm;

[0090] A variational mode decomposition module, configured to perform variational mode decomposition on the original wind speed sequence of the wind farm according to a preset decomposition layer number to obtain an original wind speed subsequence set;

[0091] A sequence correlation calculation module, configured to calculate the sequence correlation of all subsequences in the original wind speed subsequence set, and compare the sequence correlations of all subsequences to obtain the minimum correlation;

[0092] An original subsequence set adjustment module, configured to use the subsequence corresponding to the minimum correlation as the reference subsequence, and adjust the original subsequence set based on the reference subsequence according to the sequence correlation to obtain an adjusted subsequence set;

[0093] A residual sequence fusion module, configured to compare the original subsequence set with the adjusted subsequence set to obtain a residual sequence, and fuse the adjusted subsequence set with the residual sequence to obtain a corrected subsequence set;

[0094] A sequence correlation calculation module, configured to calculate the sequence correlation of the corrected subsequence set to obtain the correlation at the current decomposition layer number;

[0095] A repeated decomposition module, configured to increase the decomposition layer number of the variational mode decomposition until the decomposition layer number of the variational mode decomposition reaches the maximum, obtain the correlations at several decomposition layer numbers, and compare the correlations at several decomposition layer numbers to obtain the minimum correlation;

[0096] An optimal wind speed subsequence comparison and acquisition module, configured to use the corrected subsequence set at the decomposition layer number corresponding to the minimum correlation as the transformed subsequence, compare the maximum correlation of the residual sequence in the transformed subsequence with a threshold, and if the maximum correlation of the residual sequence in the transformed subsequence is less than the threshold, retain the residual sequence in the transformed subsequence to obtain the optimal wind speed subsequence;

[0097] A wind speed prediction result acquisition module, configured to perform wind speed prediction using the optimal wind speed subsequence to obtain a wind speed prediction sequence, and add the wind speed prediction sequences to obtain the wind speed prediction result for supercapacitor energy storage grid connection.

[0098] An improved supercapacitor energy storage frequency modulation and grid connection system provided by this embodiment can decompose the original wind speed sequence through improved VMD (Variational Mode Decomposition) to obtain more accurate and stable optimal subsequences. These subsequences can better reflect the volatility and periodicity of the wind speed, thereby improving the accuracy of wind speed prediction. Based on the high-precision wind speed prediction results, the supercapacitor energy storage system can more accurately determine when to charge and when to discharge to maximize the energy storage efficiency and reduce energy waste. Accurate wind speed prediction helps the power grid dispatching department better plan power production and distribution, reducing the problem of power supply-demand imbalance caused by wind speed changes. By optimizing the charge-discharge strategy of the supercapacitor energy storage system, the loss and maintenance cost of the energy storage equipment can be reduced. The combination of high-precision wind speed prediction and optimized supercapacitor energy storage system can improve the reliability and economy of new energy power generation.

[0099] Embodiment III

[0100] See Figure 3 , an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements an improved supercapacitor energy storage frequency modulation and grid connection method described in Embodiment I.

[0101] Embodiment IV

[0102] A computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, it implements an improved supercapacitor energy storage frequency modulation and grid connection method described in Embodiment I.

[0103] Embodiment V

[0104] Taking the preset decomposition layer number K = 6 as an example, this embodiment provides an improved supercapacitor energy storage frequency modulation and grid connection method, which specifically includes the following steps:

[0105] Step 1: Decompose the original wind speed time series by using the improved VMD decomposition of this embodiment to obtain the optimal wind speed subsequences. Specifically, the detailed process of Step 1 is as follows:

[0106] This embodiment proposes the concept of correlation:

[0107] Suppose there are two sequences: IMF a = and IMF b = , if there exists a function α (0.5 < α < 1.5) at this time, for the number of values judged in IMF a that satisfy the relationship b m = α am , assuming that there are j values satisfying this relationship, then the IMF a and the IMF b has a correlation of , , N , where m is an intermediate variable and N is the number of elements in the sequence.

[0108] Taking = [1, 2, 3, 4, 5, 6], = [2, 1.1, 7, 0, 2, 8] as an example to illustrate:

[0109] ;

[0110] It can be seen that among the 6 pairs of elements of the two sequences, 2 pairs of elements a 2 , b 2 and a 6 , b 6 satisfy the relationship b m = α a m , then the correlation between the IMF a and the IMF b is D aVb = 2 / 6 = 1 / 3.

[0111] Using VMD to decompose the original wind speed time series, assuming the initial decomposition layer K = 6, that is, the original wind speed time series x(t) is decomposed into 6 sub - sequences, IMF 1 , IMF 2 , IMF 3 ,..., IMF 6 ; among them, IMF is the intrinsic mode function, and IMF 1 represents the first sequence, IMF 2 represents the second sequence, IMF 3 represents the third sequence, IMF 6 represents the sixth sequence;

[0112] The correlation between these 6 sub - sequences and other sequences is calculated in turn and then an average value is obtained to get the sequence correlation;

[0113] For example, the correlation between the IMF 1 and the other 5 sub - sequences is obtained:

[0114] The correlation D 1 between the IMF 2 and the IMF 1V2 , the correlation D 1 between the IMF 3 and the IMF1V3 , the relevance D of IMF 1 and IMF 4 is D 1V4 , the relevance D of IMF 1 and IMF 5 is D 1V5 , the relevance D of IMF 1 and IMF 6 is D 1V6 . If so, the sequence relevance D of IMF 1 is D = (D 1 + D 1V2 + D 1V3 + D 1V4 + D 1V5 + D 1V6 ) / 5;

[0115] Find the subsequence IMF min corresponding to the minimum sequence relevance D min . The remaining 5 subsequences are adjusted in this direction, and the adjustment method is as follows:

[0116] Let the subsequence IMF min with the minimum relevance be [s 1 , s 2 ,.., s i ,..., s N and its relevance is D min .

[0117] The subsequence to be adjusted is IMF x = [r 1 , r 2 ,.., r i ,..., r N and its relevance is D x ;

[0118]

[0119] Among them, is the adjusted element in the subsequence to be adjusted, r i is the element in the subsequence to be adjusted, sigmoid( ) represents the activation function; D min is the sequence relevance of the reference subsequence, D x is the sequence relevance of the subsequence to be adjusted;

[0120] represents the augmentation coefficient, and the specific expression is:

[0121]

[0122] max( D 1 ,D 2 ,…,D 6 ) represents the maximum value of the sequence correlation degree of the original subsequence set, min( s 1 ,s 2 ,.., s i ,...,s N ) represents the minimum element in the reference subsequence, max( r 1 ,r 2 ,..,r i ,...,r N ) represents the maximum element in the subsequence to be adjusted, N represents the number of elements in the sequence; i is an intermediate variable.

[0123] Except for IMF min the remaining subsequences are adjusted according to the above rules, and the new subsequences IMF 1 、 , IMF 2 、 , IMF 3 、 ,..., IMF 6 、 .

[0124] Next, solve the residual sequence R s :

[0125]

[0126] Calculate the correlation degree between each pair of IMF 1 、 , IMF 2 、 , IMF 3 、 ,..., IMF 6 、 and R s respectively, and then solve the average value to obtain the correlation D k=6 under K = 6. i is an intermediate variable.

[0127] Let K = 7 and repeat the above steps to obtain D k=7 , and loop according to this logic until D is obtainedk=30 , output D k=6 , D k=7 , …, D k=30 The minimum value D min The corresponding transformed subsequence IMF 1 ,IMF 2 ,IMF 3 ,..,IMF min ,R S ;

[0128] For the minimum value D min The corresponding transformed subsequence is calculated as R S and [IMF 1 ,IMF 2 ,IMF 3 ,..,IMF min ];

[0129] If the maximum correlation is less than 0.6, then R S reserve;

[0130] Otherwise, for R S Continue to use VMD decomposition, the initial K 1 =5, iteration K 1 =K 1 +1 till R S Each subsequence of [IMF 1 ,IMF 2 ,IMF 3 ,..,IMF min ] are all less than 0.6.

[0131] Step 2: Put the subsequences into the GRU (Gated Recurrent Unit) network IPSO-SVR model for prediction, using the advantages of the GRU network in processing time series data, and using the IPSO-SVR model to improve the prediction accuracy, and get the prediction result sequence;

[0132] Step 3: Add the prediction results to get the final wind speed prediction result.

[0133] In order to further verify the advantages of this embodiment, the method of this embodiment and the general prediction method are used to predict the wind speed of the super-capacity energy storage grid using the test data set, and the MAPE (Mean Absolute Percentage Error) is compared. The comparison results are shown in Table 1:

[0134] Table 1 Comparison results of MAPE of wind speed series predicted by the method of this embodiment and general prediction methods

[0135]

[0136] It can be seen that the MAPE value generated by the optimal wind speed sequence for over-capacity energy storage grid connection obtained by the wind speed prediction method provided in this embodiment is significantly lower than that of the traditional wind speed prediction method. A lower MAPE value means that the wind speed prediction result is more accurate, thus being able to better guide the charge and discharge strategies of the over-capacity energy storage grid connection system.

[0137] Those skilled in the art should understand that the embodiments of the present invention can be provided as a method, a system, or a computer program product. Therefore, the present invention can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention 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, read-only optical disk, optical storage, etc.) containing computer-usable program code.

[0138] The present invention is described with reference to the flowcharts and / or block diagrams of methods, apparatuses (systems), and computer program products according to embodiments of the present invention. It should be understood that each flow and / or block in the flowchart and / or block diagram can be implemented by computer program instructions, and the combination of flows 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 means for implementing the specified functions in one process Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0139] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable memory generate a manufactured article including instruction means, and the instruction means implements the specified functions in one process Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0140] These computer program instructions can also be loaded onto a computer or other programmable data processing device, so that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process. Thus, the instructions executed on the computer or other programmable device provide means for implementing the specified functions in one process Figure 1 one process or multiple processes and / or blocks Figure 1Steps of the functions specified in one or more boxes. Finally, it should be noted that: The above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to the above embodiments, those of ordinary skill in the art should understand that: Modifications or equivalent substitutions can still be made to the specific implementation manners of the present invention, and any modification or equivalent substitution that does not depart from the spirit and scope of the present invention should be covered within the protection scope of the present invention.

Claims

1. An improved supercapacitor energy storage frequency modulation grid-connected method, characterized in that: The following steps are involved: S1. Obtain the original wind speed sequence of the wind farm; S2. performing variational modal decomposition on the original wind speed sequence of the wind farm based on a preset number of decomposition layers to obtain a set of original wind speed subsequences; S3, calculating the sequence correlation of all subsequences in the original wind speed subsequence set, and comparing the sequence correlations of all subsequences to obtain the minimum correlation; S4, taking the subsequence corresponding to the minimum correlation as a reference subsequence, and adjusting the original subsequence set according to the sequence correlation to obtain an adjusted subsequence set based on the reference subsequence; S5, comparing the original subsequence set with the adjusted subsequence set to obtain a residual sequence, and fusing the adjusted subsequence set with the residual sequence to obtain a modified subsequence set; S6. Calculate the sequence correlation of the modified subsequence set to obtain the correlation at the current decomposition level; S7, return to S2, and increase the number of decomposition levels of variational mode decomposition, repeat steps S2-S6 until the number of decomposition levels of variational mode decomposition reaches the maximum, obtain the correlations under several decomposition levels, and compare the correlations under several decomposition levels to obtain the minimum correlation value; S8, taking the modified subsequence set under the decomposition level corresponding to the minimum correlation value as the transformed subsequence, comparing the maximum correlation of the residual sequence in the transformed subsequence with the threshold value, and if the maximum correlation of the residual sequence in the transformed subsequence is less than the threshold value, retaining the residual sequence in the transformed subsequence as the optimal wind speed subsequence; S9. Use the optimal wind speed subsequence to predict the wind speed to obtain a wind speed prediction sequence, and add the wind speed prediction sequences to obtain the wind speed prediction result of the super-capacity energy storage grid connection.

2. The improved supercapacitor energy storage frequency modulation grid-connected method according to claim 1, characterized in that: The original wind speed sequence of the wind farm includes historical wind speed data of the wind farm, and the number of subsequences in the original subsequence set in S2 is the number of decomposition levels of variational mode decomposition.

3. The improved supercapacitor energy storage frequency modulation grid-connected method according to claim 2, characterized in that: The calculation of the sequence correlation is performed using the following formula: in, D is the sequence correlation, For the i -1 sequence and i The sequence correlation of the sequences, j For the i -1 is the number of elements in the sequence that satisfy the correlation relationship, N is the i -1 sequence with the i The number of elements in the sequence, i is a positive integer, 1< i ≤30; The correlation relationship is b m =α a m ,in, a m For the i -1 element in the sequence, b m For the i elements in a sequence, α is the correlation function, 0.5<α<1.

5.

4. The improved supercapacitor energy storage frequency modulation grid-connected method according to claim 1, characterized in that: The reference subsequence is used as a reference, and the original subsequence set is adjusted according to the sequence correlation to obtain the adjusted subsequence set, using the following adjustment formula: in, is the adjusted element in the subsequence that needs to be adjusted, sigmoid( ) represents the activation function, is the element in the subsequence that needs to be adjusted, represents the supplementary coefficient; D min is the sequence correlation of the benchmark subsequence, D x is the sequence correlation of the subsequence that needs to be adjusted, max( D 1 ,D 2 ,…,D K ) represents the maximum value of the sequence correlation of the original subsequence set, K is the number of decomposition levels, min( s 1 ,s 2 ,..,s i ,...,s N ) represents the minimum element in the base subsequence, max( r 1 ,r 2 ,..,r i ,...,r N ) represents the maximum element in the subsequence that needs to be adjusted. N Represents the total number of elements in the sequence, avg( r 1 ,r 2 ,..,r i ,...,r N ) represents the average value of the elements in the subsequence that needs to be adjusted. e is a natural constant, π is a constant, i is an intermediate variable.

5. The improved supercapacitor energy storage frequency modulation grid-connected method according to claim 4, characterized in that: The comparison between the original subsequence set and the adjusted subsequence set to obtain a residual sequence is performed using the following comparison formula: in, is the residual sequence, is the original subsequence set, is the adjusted subsequence set, K is the number of decomposition layers, i is an intermediate variable; The sequence correlation of the modified subsequence set in S6 is the average value of the correlation between the adjusted subsequence set and the residual sequence.

6. The improved supercapacitor energy storage frequency modulation grid-connected method according to claim 1, characterized in that: The maximum correlation of the residual sequence in the transformed subsequence is compared with a threshold. If the maximum correlation of the residual sequence in the transformed subsequence is greater than the threshold, the residual sequence in the transformed subsequence is decomposed using variational mode decomposition until the maximum correlation of the residual sequence in the transformed subsequence is less than the threshold, thereby obtaining the optimal wind speed subsequence.

7. The improved supercapacitor energy storage frequency modulation grid-connected method according to claim 1, characterized in that: The wind speed prediction described in S9 adopts the IPSO-SVR model, and the optimal wind speed subsequences are sequentially input into the IPSO-SVR model to obtain the wind speed prediction sequence.

8. An improved supercapacitor energy storage frequency modulation grid-connected system, based on an improved supercapacitor energy storage frequency modulation grid-connected method as claimed in any one of claims 1 to 7, characterized in that: include: The module for obtaining the original wind speed sequence is used to obtain the original wind speed sequence of the wind farm; A variational mode decomposition module is used to perform variational mode decomposition on the original wind speed sequence of the wind farm based on a preset number of decomposition layers to obtain a set of original wind speed subsequences; The sequence correlation calculation module is used to calculate the sequence correlation of all subsequences in the original wind speed subsequence set, and compare the sequence correlations of all subsequences to obtain the minimum correlation; An original subsequence set adjustment module is used to take the subsequence corresponding to the minimum correlation as a reference subsequence, and adjust the original subsequence set according to the sequence correlation to obtain an adjusted subsequence set based on the reference subsequence; A fusion residual sequence module is used to compare the original subsequence set with the adjusted subsequence set to obtain a residual sequence, and fuse the adjusted subsequence set with the residual sequence to obtain a modified subsequence set; A module for calculating sequence correlation is used to calculate the sequence correlation of the modified subsequence set to obtain the correlation at the current decomposition level; A repeated decomposition module is used to increase the number of decomposition layers of the variational mode decomposition, repeat the variational mode decomposition until the number of decomposition layers of the variational mode decomposition reaches a maximum, obtain the correlations under several decomposition layers, and compare the correlations under several decomposition layers to obtain the minimum value of the correlation; The wind speed optimal subsequence module is used to compare and obtain the modified subsequence set under the decomposition layer number corresponding to the minimum correlation value as the transformed subsequence, and compare the maximum correlation of the residual sequence in the transformed subsequence with the threshold value. If the maximum correlation of the residual sequence in the transformed subsequence is less than the threshold value, the residual sequence in the transformed subsequence is retained as the wind speed optimal subsequence; The wind speed prediction result obtaining module is used to use the wind speed optimal subsequence to perform wind speed prediction to obtain a wind speed prediction sequence, and the wind speed prediction sequence is added to obtain the wind speed prediction result of the super-capacity energy storage grid connection.

9. An electronic device, characterized in that: The invention comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, an improved supercapacitor energy storage frequency modulation and grid-connected method as described in any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, an improved supercapacitor energy storage frequency modulation and grid-connected method as described in any one of claims 1 to 7 is implemented.

Images