An adaptive optimization method for supercapacitor energy storage frequency modulation and related products

Abstract

The present invention discloses an adaptive optimization method for supercapacitor energy storage frequency modulation and related products, belonging to the technical field of wind power generation. The adaptive optimization method for supercapacitor energy storage frequency modulation provided by the present invention collects wind speed data and establishes an original wind speed sequence, divides the original wind speed sequence into a prediction set and a verification set, and adjusts and corrects the original data of the wind speed sequence based on the actual value and the predicted value in the verification set wind speed sequence; by calculating the difference error and the proportional error, and calculating the correction parameter accordingly, the adaptive adjustment of the prediction model is realized, and by iteratively optimizing the prediction results, the inherent laws and characteristics of the original data of the wind speed sequence are more effectively mined and utilized, thereby improving the accuracy and reliability of wind speed prediction, ensuring the accuracy of wind speed prediction, and providing a more accurate data basis for the supercapacitor energy storage system, reducing the energy waste caused by wind energy fluctuations.

Description

Technical Field

[0001] The present invention relates to the technical field of wind power generation, and in particular to an adaptive optimization method for supercapacitor energy storage frequency modulation and related products. Background Art

[0002] Wind power generation, as a clean, environmentally friendly and sustainable renewable energy technology, has attracted much attention and has developed rapidly in recent years. With the improvement of wind power generation efficiency and the continuous decline in costs, it has gradually become a key link in the transformation of energy structures in many countries, making important contributions to achieving energy diversification and reducing greenhouse gas emissions.

[0003] Although wind power generation has many advantages, its inherent challenges cannot be ignored. Among them, the natural randomness of wind speed is a particularly prominent problem. The unpredictability of wind speed causes the output power of wind farms to be highly volatile and uncertain, which not only increases the complexity of grid dispatching, but also poses severe challenges to the stable operation of the grid. In order to effectively meet this challenge, wind speed prediction technology has emerged and has become one of the core technologies in the field of wind power generation. Traditional wind speed prediction methods have obvious limitations. The original wind speed sequence is directly input into the neural network for prediction, while ignoring the inherent laws and characteristics of the original data of the wind speed sequence. This method has a large prediction error and is difficult to meet the actual needs of grid dispatching and wind farm dispatching.

[0004] Therefore, how to more effectively utilize the inherent laws and characteristics of the original data of the wind speed sequence to improve the accuracy and reliability of the prediction has become a key technical problem that needs to be urgently solved by technical personnel in this field. Summary of the invention

[0005] The purpose of the present invention is to provide an adaptive optimization method for supercapacitor energy storage frequency modulation and related products, aiming to fundamentally solve the problem of insufficient wind speed prediction accuracy and reliability in traditional methods.

[0006] The present invention solves the above technical problems through the following technical solutions:

[0007] The present invention provides an adaptive optimization method for supercapacitor energy storage frequency modulation, comprising the following steps:

[0008] Step 1: Collect wind speed data, establish the original wind speed sequence, and divide the original wind speed sequence into a prediction set wind speed sequence and a verification set wind speed sequence according to the order and preset ratio; input the prediction set wind speed sequence into the trained neural network model in order for prediction, and obtain the predicted value wind speed sequence Y i , i is the number of cycles; based on the predicted wind speed sequence Y iThe difference error W of the i-th cycle is calculated by using the predicted value in the validation set and the actual value in the wind speed sequence of the validation set. 1i and proportional error W 2i ; Based on the difference error W of the i-th cycle 1i and proportional error W 2i , calculate the correction parameter Q i ;

[0009] Step 2: Determine the correction parameter Q i Is it greater than 1? If so, rearrange the correction parameter Q i The corresponding wind speed sequence is input to the trained neural network model for prediction, and the rearranged wind speed sequence is obtained; if the judgment is no, the correction parameter Q is corrected i The corresponding wind speed sequence is input into the trained neural network model for prediction to obtain the corrected wind speed sequence;

[0010] Step 3: Let i=i+1, and input the rearranged wind speed sequence or the corrected wind speed sequence obtained in step 2 into the trained neural network model in order for prediction, and obtain the predicted wind speed sequence Y i , based on the predicted wind speed series Y i and the actual value in the wind speed sequence of the validation set, and calculate the difference error W of the i-th cycle 1i and proportional error W 2i , and execute step 4;

[0011] Step 4: Determine the difference error W 1i and proportional error W 2i Whether conditions one and two are met at the same time, the condition one being:

[0012]

[0013] The second condition is:

[0014]

[0015] If the judgment is no, then based on the difference error W 1i and proportional error W 2i , calculate the correction parameter Q corresponding to the current number of cycles i And return to step 2; if the judgment is yes, the predicted value wind speed sequence Y corresponding to the current cycle number i is the final wind speed prediction result;

[0016] Step 5. Calculate the predicted power sent by the wind farm to the grid based on the final wind speed prediction result, obtain the actual power sent by the wind farm to the grid, and calculate the difference between the predicted power and the actual power. When the difference is positive, adjust the super-capacity energy storage system connected to the wind farm exit to the charging state; when the difference is 0, the super-capacity energy storage system connected to the wind farm exit maintains the current state unchanged; when the difference is negative, adjust the super-capacity energy storage system connected to the wind farm exit to the discharging state.

[0017] A further improvement of the present invention is that: the correction parameter Q i for:

[0018] .

[0019] A further improvement of the present invention is that in step 2, the correction parameter Q i The corresponding wind speed sequence input to the trained neural network model for prediction is X(t), X(t)=X1,X2,X3,...,X N , where N is the number of subsequences in X(t), X1,X2,X3,...,X N are the 1st, 2nd, 3rd, …, Nth subsequences in X(t) respectively;

[0020] The rearrangement correction parameter Q i The corresponding wind speed sequence input to the trained neural network model for prediction is as follows: the two subsequences in X(t) whose order conforms to the first relationship are swapped. The first relationship is:

[0021]

[0022] Among them, sigmoid() is the activation function; tanh() is the activation function; sin() is the sine function; arccos() is the inverse cosine function; rand() is the random function; e is a natural constant; , b are the orders of the two subsequences that conform to the first relation; x b is the b-th subsequence in X(t); [ ] is the rounding function.

[0023] A further improvement of the present invention is that: When the values ​​correspond to multiple subsequences in X(t), they will have the same The first relation of multiple subsequences of values ​​is replaced by:

[0024]

[0025] and select the calculated The closest to the original Subsequence of values ​​and subsequence Xa swap positions;

[0026] in, To have the same Multiple subsequences of values ​​are replaced by the first relation after calculation value.

[0027] A further improvement of the present invention is that: the correction parameter Q i The corresponding wind speed sequence input to the trained neural network model for prediction is specifically: by compensating each subsequence in X(t) with the second relationship, the second relationship is:

[0028]

[0029] Among them, x s is the s-th subsequence in X(t), and the value of s ranges from 1 to N; S is the sth subsequence in X(t) after compensation; max() is the maximum value function; W 21 , W 22 , W 23 , …, W 2i They are the proportional errors corresponding to the 1st, 2nd, 3rd, …, i-th cycles respectively.

[0030] A further improvement of the present invention is that: assuming that the original wind speed sequence contains T subsequences, the preset ratio is : , then the wind speed sequence of the forecast set contains subsequences, the wind speed sequence of the validation set contains Actual value;

[0031] The difference error W 1i Specifically:

[0032]

[0033] Among them, y n is the nth actual value in the wind speed sequence of the validation set; y in is the predicted wind speed sequence Y in the i-th cycle i The nth predicted value in .

[0034] A further improvement of the present invention is that: the proportional error W 2i Specifically:

[0035] .

[0036] A computer device comprises a memory and a processor, wherein the memory stores a computer program and the processor implements the steps of the above-mentioned self-adaptive optimization method for supercapacitor energy storage frequency modulation when executing the computer program.

[0037] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of the above-mentioned adaptive optimization method for supercapacitor energy storage frequency modulation.

[0038] A computer program product includes a computer program, which implements the steps of the above-mentioned self-adaptive optimization method for supercapacitor energy storage frequency modulation when executed by a processor.

[0039] Compared with the prior art, the positive and progressive effects of the present invention are:

[0040] The adaptive optimization method for supercapacitor energy storage frequency modulation provided by the present invention can quickly adapt to and make accurate predictions when facing complex and changeable wind speed environments, thereby providing strong guarantees for the stable operation of wind farms. Specifically, wind speed data is collected and an original wind speed sequence is established, the original wind speed sequence is divided into a prediction set and a verification set, and the original data of the wind speed sequence is adjusted and corrected based on the actual value and the predicted value in the verification set wind speed sequence; the difference error and the proportional error are calculated, and the correction parameters are calculated accordingly to achieve adaptive adjustment of the prediction model, and the prediction results are iteratively optimized to achieve more effective mining and utilization of the inherent laws and characteristics of the original data of the wind speed sequence, thereby improving the accuracy and reliability of wind speed prediction, ensuring the accuracy of wind speed prediction, providing a more accurate data basis for the supercapacitor energy storage system, and reducing energy waste caused by wind energy fluctuations. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] The drawings in the specification are used to provide further understanding of the present invention and constitute a part of the present invention. The schematic embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute improper limitations on the present invention.

[0042] Figure 1 It is a schematic diagram of the process of the present invention. DETAILED DESCRIPTION

[0043] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0044] In the description of the present invention, it should be understood that the terms “include” and “comprises” indicate the presence of described features, wholes, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components and / or collections thereof.

[0045] It should also be understood that the terms used in the present specification are only for the purpose of describing specific embodiments and are not intended to limit the present invention. As used in the present specification and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an" and "the" are intended to include plural forms.

[0046] It should be understood that, although the terms first, second, third, etc. may be used to describe preset ranges, etc. in the embodiments of the present invention, these preset ranges should not be limited to these terms. These terms are only used to distinguish preset ranges from each other. For example, without departing from the scope of the embodiments of the present invention, the first preset range may also be referred to as the second preset range, and similarly, the second preset range may also be referred to as the first preset range.

[0047] The word "if" as used herein may be interpreted as "at the time of" or "when" or "in response to determining" or "in response to detecting", depending on the context. Similarly, the phrases "if it is determined" or "if (stated condition or event) is detected" may be interpreted as "when it is determined" or "in response to determining" or "when detecting (stated condition or event)" or "in response to detecting (stated condition or event)", depending on the context.

[0048] Various structural schematic diagrams of the embodiments disclosed in the present invention are shown in the accompanying drawings. These figures are not drawn to scale, and some details are magnified and some details may be omitted for the purpose of clear expression. The shapes of various regions and layers shown in the figures and the relative sizes and positional relationships therebetween are only exemplary, and may deviate in practice due to manufacturing tolerances or technical limitations, and those skilled in the art may additionally design regions / layers with different shapes, sizes, and relative positions according to actual needs.

[0049] The present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments, which are intended to explain the present invention rather than to limit it.

[0050] See also Figure 1 The present invention provides an adaptive optimization method for supercapacitor energy storage frequency modulation, comprising the following steps:

[0051] An adaptive optimization method for supercapacitor energy storage frequency modulation comprises the following steps:

[0052] Step 1: Collect wind speed data, establish the original wind speed sequence, and divide the original wind speed sequence into a prediction set wind speed sequence and a verification set wind speed sequence according to the order and preset ratio; input the prediction set wind speed sequence into the trained neural network model in order for prediction, and obtain the predicted value wind speed sequence Y i , i is the number of cycles; based on the predicted wind speed sequence Y i The difference error W of the i-th cycle is calculated by using the predicted value in the validation set and the actual value in the wind speed sequence of the validation set. 1i and proportional error W 2i ; Based on the difference error W of the i-th cycle 1i and proportional error W 2i , calculate the correction parameter Q i ;

[0053] Step 2: Determine the correction parameter Q i Is it greater than 1? If so, rearrange the correction parameter Q i The corresponding wind speed sequence is input to the trained neural network model for prediction, and the rearranged wind speed sequence is obtained; if the judgment is no, the correction parameter Q is corrected i The corresponding wind speed sequence is input into the trained neural network model for prediction to obtain the corrected wind speed sequence;

[0054] Step 3: Let i=i+1, and input the rearranged wind speed sequence or the corrected wind speed sequence obtained in step 2 into the trained neural network model in order for prediction, and obtain the predicted wind speed sequence Y i , based on the predicted wind speed series Y i and the actual value in the wind speed sequence of the validation set, and calculate the difference error W of the i-th cycle 1i and proportional error W 2i , and execute step 4;

[0055] Step 4: Determine the difference error W 1i and proportional error W 2i Whether conditions one and two are met at the same time, the condition one being:

[0056]

[0057] The second condition is:

[0058]

[0059] If the judgment is no, then based on the difference error W 1i and proportional error W 2i , calculate the correction parameter Q corresponding to the current number of cycles i And return to step 2; if the judgment is yes, the predicted value wind speed sequence Y corresponding to the current cycle numberi is the final wind speed prediction result;

[0060] Step 5. Calculate the predicted power sent by the wind farm to the grid based on the final wind speed prediction result, obtain the actual power sent by the wind farm to the grid, and calculate the difference between the predicted power and the actual power. When the difference is positive, adjust the super-capacity energy storage system connected to the wind farm exit to the charging state; when the difference is 0, the super-capacity energy storage system connected to the wind farm exit maintains the current state unchanged; when the difference is negative, adjust the super-capacity energy storage system connected to the wind farm exit to the discharging state.

[0061] The adaptive optimization method for supercapacitor energy storage frequency modulation provided by the present invention can quickly adapt to and make accurate predictions when facing complex and changeable wind speed environments, thereby providing strong guarantees for the stable operation of wind farms. Specifically, wind speed data is collected and an original wind speed sequence is established, the original wind speed sequence is divided into a prediction set and a verification set, and the original data of the wind speed sequence is adjusted and corrected based on the actual value and the predicted value in the verification set wind speed sequence; the difference error and the proportional error are calculated, and the correction parameters are calculated accordingly to achieve adaptive adjustment of the prediction model, and the prediction results are iteratively optimized to achieve more effective mining and utilization of the inherent laws and characteristics of the original data of the wind speed sequence, thereby improving the accuracy and reliability of wind speed prediction, ensuring the accuracy of wind speed prediction, providing a more accurate data basis for the supercapacitor energy storage system, and reducing energy waste caused by wind energy fluctuations.

[0062] Specific: Correction parameter Q i for:

[0063] .

[0064] Specifically: In step 2, correct the parameter Q i The corresponding wind speed sequence input to the trained neural network model for prediction is X(t), X(t)=X1,X2,X3,...,X N , where N is the number of subsequences in X(t), X1,X2,X3,...,X N are the 1st, 2nd, 3rd, …, Nth subsequences in X(t) respectively;

[0065] The rearrangement correction parameter Q i The corresponding wind speed sequence input to the trained neural network model for prediction is as follows: the two subsequences in X(t) whose order conforms to the first relationship are swapped. The first relationship is:

[0066]

[0067] Among them, sigmoid() is the activation function; tanh() is the activation function; sin() is the sine function; arccos() is the inverse cosine function; rand() is the random function; e is a natural constant; , b are the orders of the two subsequences that conform to the first relation; x b is the b-th subsequence in X(t); [ ] is the rounding function.

[0068] Specific: When the same When the values ​​correspond to multiple subsequences in X(t), they will have the same The first relation of multiple subsequences of values ​​is replaced by:

[0069]

[0070] and select the calculated The closest to the original Subsequence of values ​​and subsequence X a swap positions;

[0071] in, To have the same Multiple subsequences of values ​​are replaced by the first relation after calculation value.

[0072] Specifically: the correction parameter Q i The corresponding wind speed sequence input to the trained neural network model for prediction is specifically: by compensating each subsequence in X(t) with the second relationship, the second relationship is:

[0073]

[0074] Among them, x s is the s-th subsequence in X(t), and the value of s ranges from 1 to N; S is the sth subsequence in X(t) after compensation; max() is the maximum value function; W 21 , W 22 , W 23 , …, W 2i They are the proportional errors corresponding to the 1st, 2nd, 3rd, …, i-th cycles respectively.

[0075] Specifically: Assume that the original wind speed sequence contains T subsequences, and the preset ratio is : , then the wind speed sequence of the forecast set contains subsequences, the wind speed sequence of the validation set contains Actual value;

[0076] The difference error W 1i Specifically:

[0077]

[0078] Among them, y n is the nth actual value in the wind speed sequence of the validation set; y in is the predicted wind speed sequence Y in the i-th cycle i The nth predicted value in .

[0079] Specific: Proportional error W 2i Specifically:

[0080] .

[0081] Embodiment 1

[0082] An adaptive optimization method for supercapacitor energy storage frequency modulation comprises the following steps:

[0083] Step 1: Collect wind speed data, establish an original wind speed sequence, and divide the original wind speed sequence into a prediction set wind speed sequence and a verification set wind speed sequence according to the order and preset ratio;

[0084] The original wind speed sequence X0(t) contains T subsequences, T=10; X0(t)==X1,X2,X3,...,X 10 ,X1,X2,X3,...,X 10 are the 1st, 2nd, 3rd, ..., 10th subsequences in the original wind speed sequence respectively; set the preset ratio to q1:q2= : , then the wind speed sequence of the prediction set includes X1, X2, X3, X4, X5, X6, X7, X8, X9, and the wind speed sequence of the verification set includes X 10 The actual value of a subsequence.

[0085] The wind speed sequence of the prediction set is input into the trained neural network model in order for prediction, and the predicted wind speed sequence Y is obtained. i , i is the number of cycles; at this time i=1, the predicted wind speed sequence is Y1, and the number of predicted values ​​in the predicted wind speed sequence is equal to the number of actual values ​​in the verification set wind speed sequence. Since there is only one actual value X 10 , so there is only one predicted value.

[0086] Based on the predicted values ​​in the predicted wind speed sequence Y1 and the actual values ​​in the verification set wind speed sequence, the difference error W of the first cycle is calculated. 11 and proportional error W 21 ;

[0087] Since i=1, q2=1 at this time, the difference error W 11 =|y1-y 11 |, proportional error W 21 =| |;

[0088] Among them, y1 is the first actual value in the wind speed sequence of the validation set, that is, X 10 ;y 11 It is the first predicted value in the predicted wind speed sequence Y1 in the first cycle.

[0089] Based on the difference error W of the first cycle 11 and proportional error W 21 , calculate and obtain the correction parameter Q1;

[0090] Step 2: Determine whether the correction parameter Q1 is greater than 1. If it is, rearrange the wind speed sequence corresponding to the correction parameter Q1 input to the trained neural network model for prediction (when i=1, it is the prediction set wind speed sequence) to obtain the rearranged wind speed sequence; if it is not, correct the wind speed sequence corresponding to the correction parameter Q1 input to the trained neural network model for prediction (when i=1, it is the prediction set wind speed sequence) to obtain the corrected wind speed sequence;

[0091] In this embodiment, the correction parameter Q1 is less than 1, and the prediction set wind speed sequence corresponding to the correction parameter Q1 should be corrected and input to the trained neural network model for prediction. The prediction set wind speed sequence includes: X1, X2, X3, X4, X5, X6, X7, X8, X9, and the corrected X(t)* is obtained;

[0092] Step 3: Let i=i+1=2, and input X(t)* obtained in step 2 into the trained neural network model in sequence for prediction to obtain the predicted wind speed sequence Y2. Based on the predicted wind speed sequence Y2 and the actual values ​​in the verification set wind speed sequence, calculate the difference error W of the second cycle. 12 and proportional error W 22 , and execute step 4;

[0093] Step 4: Determine the difference error W of the second cycle 12 and proportional error W 22 Whether conditions 1 and 2 are met at the same time, condition 1 is:

[0094]

[0095] Condition 2 is:

[0096]

[0097] If the judgment is no, then based on the difference error W of the second cycle 12 and proportional error W 22 , calculate the correction parameter Q2 of the second cycle and return to step 2 to judge the correction parameter Q2;

[0098] If the answer is yes, the predicted wind speed sequence Y2 corresponding to the second cycle is the final wind speed prediction result;

[0099] In this embodiment, assuming that the judgment result is yes, the predicted value wind speed sequence Y2 is the final wind speed prediction result;

[0100] Step 5. Calculate the predicted power sent by the wind farm to the grid based on the final wind speed prediction result, obtain the actual power sent by the wind farm to the grid, and calculate the difference between the predicted power and the actual power. When the difference is positive, adjust the super-capacity energy storage system connected to the wind farm exit to the charging state; when the difference is 0, the super-capacity energy storage system connected to the wind farm exit maintains the current state unchanged; when the difference is negative, adjust the super-capacity energy storage system connected to the wind farm exit to the discharging state.

[0101] Embodiment 2

[0102] An adaptive optimization method for supercapacitor energy storage frequency modulation comprises the following steps:

[0103] Step 1: Collect wind speed data, establish an original wind speed sequence, and divide the original wind speed sequence into a prediction set wind speed sequence and a verification set wind speed sequence according to the order and preset ratio;

[0104] The original wind speed sequence X0(t) contains T subsequences, T=10; X0(t)==X1,X2,X3,...,X 10 ,X1,X2,X3,...,X 10 are the 1st, 2nd, 3rd, ..., 10th subsequences in the original wind speed sequence respectively; set the preset ratio to q1:q2= : , then the wind speed sequence of the prediction set includes X1, X2, X3, X4, X5, X6, X7, X8, X9, and the wind speed sequence of the verification set includes X 10 The actual value of a subsequence.

[0105] The wind speed sequence of the prediction set is input into the trained neural network model in order for prediction, and the predicted wind speed sequence Y is obtained. i , i is the number of cycles; at this time i=1, the predicted wind speed sequence is Y1, and the number of predicted values ​​in the predicted wind speed sequence is equal to the number of actual values ​​in the verification set wind speed sequence. Since there is only one actual value X 10 , so there is only one predicted value.

[0106] Based on the predicted values ​​in the predicted wind speed sequence Y1 and the actual values ​​in the verification set wind speed sequence, the difference error W of the first cycle is calculated. 11 and proportional error W 21 ;

[0107] Since i=1, q2=1 at this time, the difference error W 11 =|y1-y 11 |, proportional error W 21 =| |;

[0108] Among them, y1 is the first actual value in the wind speed sequence of the validation set, that is, X 10 ;y 11 It is the first predicted value in the predicted wind speed sequence Y1 in the first cycle.

[0109] Based on the difference error W of the first cycle 11 and proportional error W 21 , calculate and obtain the correction parameter Q1;

[0110] Step 2: Determine whether the correction parameter Q1 is greater than 1. If it is, rearrange the wind speed sequence corresponding to the correction parameter Q1 input to the trained neural network model for prediction (when i=1, it is the prediction set wind speed sequence) to obtain the rearranged wind speed sequence X(t)**; if it is not, correct the wind speed sequence corresponding to the correction parameter Q1 input to the trained neural network model for prediction (when i=1, it is the prediction set wind speed sequence) to obtain the corrected wind speed sequence;

[0111] In this embodiment, the correction parameter Q1 is greater than 1, and the wind speed sequence of the prediction set is rearranged. The wind speed sequence of the prediction set includes: X1, X2, X3, X4, X5, X6, X7, X8, X9, specifically:

[0112] Substitute X1, X2, X3, X4, X5, X6, X7, X8, and X9 into the first relationship to obtain the corresponding a value. In this embodiment, it is assumed that the a values ​​obtained by X1, X2, X3, X4, X5, X6, X7, X8, and X9 are 9, 8, 7, 6, 5, 4, 3, 2, and 1, respectively, that is, there are no multiple subsequences in X(t) corresponding to the same a value. The subsequences of the order of a values ​​in X(t) are swapped with the subsequences corresponding to the a values ​​obtained by substituting them, and the rearranged wind speed sequence X(t)**=X9, X8, X7, X6, X5, X4, X3, X2, X1 is obtained;

[0113] Step 3: Let i=i+1=2, and input X(t)** obtained in step 2 into the trained neural network model in sequence for prediction to obtain the predicted wind speed sequence Y2. Based on the predicted wind speed sequence Y2 and the actual values ​​in the verification set wind speed sequence, calculate the difference error W of the second cycle. 12 and proportional error W 22 ;

[0114] Step 4: Determine the difference error W of the second cycle 12 and proportional error W 22 Whether both condition 1 and condition 2 are satisfied; if not, then based on the difference error W of the second cycle 12 and proportional error W 22 , calculate the correction parameter Q2 of the second cycle and return to step 2 to judge the correction parameter Q2;

[0115] If the answer is yes, the predicted wind speed sequence Y2 corresponding to the second cycle is the final wind speed prediction result;

[0116] In this embodiment, assuming that the judgment result is yes, the predicted value wind speed sequence Y2 is the final wind speed prediction result;

[0117] Step 5. Calculate the predicted power sent by the wind farm to the grid based on the final wind speed prediction result, obtain the actual power sent by the wind farm to the grid, and calculate the difference between the predicted power and the actual power. When the difference is positive, adjust the super-capacity energy storage system connected to the wind farm exit to the charging state; when the difference is 0, the super-capacity energy storage system connected to the wind farm exit maintains the current state unchanged; when the difference is negative, adjust the super-capacity energy storage system connected to the wind farm exit to the discharging state.

[0118] In order to further verify the advantages of this method, the wind speed is predicted by using the method of this application and the traditional prediction method VMD (Variational Mode Decomposition). In the experiment of this application, MAPE (Mean Absolute Percentage Error) is selected as the evaluation standard of each model. The prediction results (i.e., MAPE values) are shown in Table 1:

[0119] Table 1 Prediction results

[0120]

[0121] It can be seen that the wind speed prediction using this method is more accurate and reliable.

[0122] Based on the same inventive concept, an embodiment of the present application provides a computer 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, the steps of the adaptive optimization method of supercapacitor energy storage frequency modulation are implemented, wherein the memory may include a memory, such as a high-speed random access memory, and may also include a non-volatile memory, such as at least one disk memory, etc.; the processor, the network interface, and the memory are interconnected through an internal bus, and the internal bus may be an industrial standard architecture bus, a peripheral component interconnection standard bus, an extended industrial standard structure bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc. The memory is used to store programs. Specifically, the program may include a program code, and the program code includes computer operation instructions. The memory may include a memory and a non-volatile memory, and provide instructions and data to the processor.

[0123] Based on the same inventive concept, an embodiment of the present application provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, the steps of the adaptive optimization method of supercapacitor energy storage frequency modulation are implemented. Specifically, the computer-readable storage medium includes, but is not limited to, for example, volatile memory and / or non-volatile memory. The volatile memory may include RAM (Random Access Memory) and / or cache memory (cache), etc. The non-volatile memory may include ROM (Read-Only Memory), hard disk, flash memory, optical disk, magnetic disk, etc.

[0124] Based on the same inventive concept, an embodiment of the present application provides a computer program product, which includes a computer program stored on a computer-readable storage medium, and the computer program includes program instructions. When the program instructions are executed by a computer device, the computer device executes the steps of the above-mentioned adaptive optimization method for supercapacitor energy storage frequency modulation.

[0125] It should be understood by those skilled in the art that the embodiments of the present invention may be provided as methods or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM (Compact Disc Read-Only Memory), optical storage, etc.) containing computer-usable program codes.

[0126] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram and the combination of processes and / or blocks in the flowchart and / or block diagram can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer device or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0127] These computer program instructions may also be stored in a computer readable memory capable of directing a computer device or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer readable memory produce a manufactured product including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0128] These computer program instructions can also be loaded onto a computer device or other programmable data processing device so that a series of operation steps are executed on the computer device or other programmable device to produce a process implemented by the computer device, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0129] Although the preferred embodiments of the present invention have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

[0130] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. An adaptive optimization method for supercapacitor energy storage frequency modulation, characterized in that: The following steps are involved: Step 1: Collect wind speed data, establish the original wind speed sequence, and divide the original wind speed sequence into a prediction set wind speed sequence and a verification set wind speed sequence according to the order and preset ratio; input the prediction set wind speed sequence into the trained neural network model in order for prediction, and obtain the predicted value wind speed sequence Y i , i is the number of cycles; Based on the predicted wind speed series Y i The difference error W of the i-th cycle is calculated by using the predicted value in the validation set and the actual value in the wind speed sequence of the validation set. 1i and proportional error W 2i ; Based on the difference error W of the i-th cycle 1i and proportional error W 2i , calculate the correction parameter Q i ; Step 2: Determine the correction parameter Q i Is it greater than 1? If so, rearrange the correction parameter Q i The corresponding wind speed sequence is input to the trained neural network model for prediction, and the rearranged wind speed sequence is obtained; if the judgment is no, the correction parameter Q is corrected i The corresponding wind speed sequence is input into the trained neural network model for prediction to obtain the corrected wind speed sequence; Step 3: Let i=i+1, and input the rearranged wind speed sequence or the corrected wind speed sequence obtained in step 2 into the trained neural network model in order for prediction, and obtain the predicted wind speed sequence Y i , based on the predicted wind speed series Y i and the actual value in the wind speed sequence of the validation set, and calculate the difference error W of the i-th cycle 1i and proportional error W 2i , and execute step 4; Step 4: Determine the difference error W 1i and proportional error W 2i Whether conditions 1 and 2 are met at the same time, where condition 1 is: The second condition is: If the judgment is no, then based on the difference error W 1i and proportional error W 2i , calculate the correction parameter Q corresponding to the current number of cycles i And return to step 2; If the judgment is yes, then the predicted wind speed sequence Y corresponding to the current cycle number i is the final wind speed prediction result; Step 5. Calculate the predicted power sent by the wind farm to the grid based on the final wind speed prediction result, obtain the actual power sent by the wind farm to the grid, and calculate the difference between the predicted power and the actual power. When the difference is positive, adjust the super-capacity energy storage system connected to the wind farm exit to the charging state; when the difference is 0, the super-capacity energy storage system connected to the wind farm exit maintains the current state unchanged; when the difference is negative, adjust the super-capacity energy storage system connected to the wind farm exit to the discharging state.

2. The adaptive optimization method for supercapacitor energy storage frequency modulation according to claim 1, characterized in that: The correction parameter Q i for: 。 3. The adaptive optimization method for supercapacitor energy storage frequency modulation according to claim 1, characterized in that: In step 2, correct the parameter Q i The corresponding wind speed sequence input to the trained neural network model for prediction is X(t), X(t)=X1,X2,X3,...,X N , where N is the number of subsequences in X(t), X1,X2,X3,...,X N are the 1st, 2nd, 3rd, …, Nth subsequences in X(t) respectively; The rearrangement correction parameter Q i The corresponding wind speed sequence input to the trained neural network model for prediction is as follows: the two subsequences in X(t) whose order conforms to the first relationship are swapped. The first relationship is: Among them, sigmoid() is the activation function; tanh() is the activation function; sin() is the sine function; arccos() is the inverse cosine function; rand() is the random function; e is a natural constant; , b are the orders of the two subsequences that conform to the first relation; x b is the b-th subsequence in X(t); [ ] is the rounding function.

4. The adaptive optimization method for supercapacitor energy storage frequency modulation according to claim 3 is characterized in that: When the same When the values ​​correspond to multiple subsequences in X(t), they will have the same The first relation of multiple subsequences of values ​​is replaced by: and select the calculated The closest to the original Subsequence of values ​​and subsequence X a swap positions; in, To have the same Multiple subsequences of values ​​are replaced by the first relation after calculation value.

5. The adaptive optimization method for supercapacitor energy storage frequency modulation according to claim 3 is characterized in that: The modified correction parameter Q i The corresponding wind speed sequence input to the trained neural network model for prediction is specifically: by compensating each subsequence in X(t) with the second relationship, the second relationship is: Among them, x s is the s-th subsequence in X(t), and the value of s ranges from 1 to N; S is the sth subsequence in X(t) after compensation; max() is the maximum value function; W 21 , W 22 , W 23 , …, W 2i They are the proportional errors corresponding to the 1st, 2nd, 3rd, …, i-th cycles respectively.

6. The adaptive optimization method for supercapacitor energy storage frequency modulation according to claim 1 is characterized in that The original wind speed sequence contains T subsequences, and the preset ratio is : , then the wind speed sequence of the forecast set contains subsequences, the wind speed sequence of the validation set contains Actual value; The difference error W 1i Specifically: Among them, y n is the nth actual value in the wind speed sequence of the validation set; y in is the predicted wind speed sequence Y in the i-th cycle i The nth predicted value in .

7. The adaptive optimization method for supercapacitor energy storage frequency modulation according to claim 6, characterized in that: The proportional error W 2i Specifically: 。 8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the adaptive optimization method for supercapacitor energy storage frequency modulation described in any one of claims 1 to 7 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the adaptive optimization method for supercapacitor energy storage frequency modulation described in any one of claims 1 to 7 are implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the adaptive optimization method for supercapacitor energy storage frequency modulation as claimed in any one of claims 1 to 7 are implemented.

Images