Energy storage frequency modulation instruction prediction method and system of gravitational method

Abstract

The invention discloses an energy storage frequency modulation instruction prediction method and system based on a gravitational method, and relates to the technical field of energy storage frequency modulation prediction. Optimal parameter optimization is carried out on the VMD through an improved sparrow search algorithm; and outputting each optimal sub-sequence to obtain a final prediction result. According to the energy storage frequency modulation instruction prediction method based on the gravitational method provided by the invention, the independence of each sub-sequence after sequence decomposition is improved by evaluating the similarity between the sub-sequences, the decomposition effect is improved through the improved sparrow search algorithm and adaptive optimization of VMD decomposition parameters, each IMF sub-sequence is independently predicted through the GRU network, and the prediction efficiency is improved. The method improves the reliability and accuracy of overall prediction, and achieves better effects in the aspects of decomposition effect, reliability and accuracy.

Description

Technical Field

[0001] The invention relates to the technical field of energy storage frequency modulation prediction, and in particular to an energy storage frequency modulation instruction prediction method using a gravity method. Background Art

[0002] As an important means of stabilizing the frequency of power systems, energy storage frequency regulation technology has made significant progress in recent years. Traditional frequency regulation methods mainly rely on the rotational inertia of thermal power units. However, with the increase in the proportion of renewable energy, thermal power units are gradually withdrawing from the stage of history. This has prompted the widespread application of hybrid energy storage technology (such as the combination of supercapacitors and lithium batteries) in the field of frequency regulation. By dynamically allocating frequency regulation instructions, supercapacitors take on the high-frequency part and lithium batteries handle the low-frequency part. This model has played a certain role in improving the response speed and stability of the system. At the same time, signal processing technology and prediction algorithms have also been applied to optimize the performance of energy storage frequency regulation. For example, the variational mode decomposition (VMD) method improves prediction accuracy by decomposing sequences. However, despite the progress made in these technologies, the lag problem of frequency regulation instruction response has not been fundamentally solved, which has brought potential losses to system benefits.

[0003] Existing hybrid energy storage frequency modulation methods have many shortcomings, which limit their potential in high-precision prediction. First, traditional frequency modulation prediction methods rely on simple sequence decomposition (such as VMD), but the strong similarity between subsequences leads to high error accumulation in their prediction results, which is difficult to meet actual needs. In addition, there is a significant time delay in the signal transmission and energy storage device response process, which further affects the optimization of the benefit value. Secondly, existing similarity calculation methods are usually based on linear models, which are difficult to accurately capture the inherent correlation in complex dynamic sequences, resulting in insufficient robustness and accuracy of the prediction. Summary of the invention

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

[0005] Therefore, the technical problems solved by the present invention are: the existing frequency modulation instruction prediction method has insufficient prediction accuracy due to subsequence similarity, loss of profit caused by signal response lag, VMD parameter optimization is difficult to achieve the best decomposition effect, and how to improve the optimization problem of prediction accuracy and profit value.

[0006] To solve the above technical problems, the present invention provides the following technical solutions: a method for predicting energy storage frequency modulation instructions based on the gravity method, comprising calculating the similarity between sequences through a gravity algorithm; optimizing the optimal parameters of VMD through an improved sparrow search algorithm; outputting each optimal subsequence to obtain a final prediction result.

[0007] As a preferred solution of the energy storage frequency modulation instruction prediction method of the gravity method described in the present invention, wherein: the similarity between sequences calculated by the gravity algorithm includes gravity calculation between sequences, and the subsequences after the original sequence is decomposed by VMD and the sequence change rate sequence are expressed as: , , in, For the The intrinsic mode functions, are the subsequences after VMD decomposition, is a change rate sequence, when there are two subsequences and , the gravitational force between subsequences is expressed as: , in, is the cosine similarity between two subsequences, and They are and The experimental group error, Compare the value of the previous sequence with the value of the next sequence and then add them together. is the number of values.

[0008] As a preferred solution of the energy storage frequency modulation instruction prediction method of the gravity method described in the present invention, wherein: the similarity between the sequences calculated by the gravity algorithm also includes the total gravity calculation, when there is IMF , different subsequences share The total gravitational force is expressed as: , , in, is the total gravitational force, is the gravitational value, is the gravitational coefficient, and are the gravity coefficient and gravity value of the last set of subsequence pairs.

[0009] As a preferred solution of the energy storage frequency modulation instruction prediction method of the gravity method described in the present invention, wherein: the optimal parameter optimization of VMD by improving the sparrow search algorithm includes updating the position of the finder, for In the problem of dimensional solution space, the position of each sparrow is expressed as: , in, Reason The initial population consists of sparrows. is the number of populations, is the dimension of the solution space, and the fitness is expressed as: , in, is the fitness value, and the discoverer position update in the sparrow is expressed as: , in, is the updated finder location, is the current iteration number, For the A sparrow in The location of the dimension, for A random number between is the maximum number of iterations of the population, is a random number that follows a normal distribution, is the warning value in the population, , is the warning value, .

[0010] As a preferred solution of the energy storage frequency modulation instruction prediction method of the gravity method described in the present invention, wherein: the optimal parameter optimization of VMD by improving the sparrow search algorithm also includes follower position update, and the follower in the sparrow observes the finder position and updates the position, which is expressed as: , in, is the updated follower position, is the worst position of sparrows in the current iteration sparrow population, is the optimal position for the discoverer in the sparrow population, for The matrix has random values ​​of 1 or -1. For the matrix The transpose of .

[0011] As a preferred solution of the energy storage frequency modulation instruction prediction method of the gravity method described in the present invention, wherein: the optimal parameter optimization of VMD by improving the sparrow search algorithm also includes updating the position of the early warning behavior, and the position update of the sparrow with early warning behavior in the sparrow population is expressed as: , in, The updated position of the sparrow with warning behavior, is the optimal position of the sparrow in the current environment, is the step size control parameter, for A uniform random number between is the denominator control parameter, is the current sparrow fitness value, is the best fitness value in the current sparrow population, is the worst fitness value of sparrow.

[0012] As a preferred solution of the energy storage frequency modulation instruction prediction method of the gravity method described in the present invention, wherein: the final prediction result is obtained by optimizing the optimal The predicted value of each subsequence is compared with the actual value, and the error is calculated. The predicted results of each subsequence are added together to obtain the final prediction result.

[0013] Another object of the present invention is to provide a storage frequency modulation instruction prediction system based on the gravity method, which can evaluate the independence and intrinsic correlation between each subsequence through a similarity calculation module, thereby solving the problem that the current prediction results are interfered by strong correlations and the accuracy is reduced.

[0014] As a preferred solution of the energy storage frequency modulation instruction prediction system of the gravity method described in the present invention, it includes: a similarity calculation module, a parameter optimization module, and an output prediction module; the similarity calculation module is used to calculate the similarity between sequences through the gravity algorithm; the parameter optimization module is used to optimize the optimal parameters of VMD through the improved sparrow search algorithm; the output prediction module is used to output each optimal subsequence to obtain the final prediction result.

[0015] A computer device includes a memory and a processor, wherein the memory stores a computer program, and is characterized in that the processor executes the computer program to implement the steps of a method for predicting energy storage frequency modulation instructions using a gravity method.

[0016] 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 a method for predicting energy storage frequency modulation instructions using a gravity method are implemented.

[0017] Beneficial effects of the present invention: The energy storage frequency modulation instruction prediction method of the gravity method provided by the present invention improves the independence of each subsequence after sequence decomposition by evaluating the similarity between subsequences, adaptively optimizes VMD decomposition parameters through an improved sparrow search algorithm, improves the decomposition effect, and predicts each IMF subsequence separately through a GRU network, thereby improving the reliability and accuracy of the overall prediction. The present invention achieves better results in terms of decomposition effect, reliability and accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative labor. Among them: Figure 1 An overall flow chart of a method for predicting energy storage frequency modulation instructions using a gravity method provided in the first embodiment of the present invention.

[0019] Figure 2 A prediction flow chart of a method for predicting energy storage frequency modulation instructions using a gravity method provided in the first embodiment of the present invention.

[0020] Figure 3 An overall module diagram of a gravity method energy storage frequency modulation instruction prediction system provided for the third embodiment of the present invention. DETAILED DESCRIPTION

[0021] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in the art without creative work should fall within the scope of protection of the present invention.

[0022] Example 1, reference Figure 1-Figure 2 , which is an embodiment of the present invention, provides a method for predicting energy storage frequency modulation instructions using a gravity method, comprising: S1: Calculate the similarity between sequences using the gravitational algorithm.

[0023] Furthermore, calculating the similarity between sequences by the gravitational algorithm includes gravitational calculation between sequences.

[0024] It should be noted that the subsequences of the original sequence after VMD decomposition and the sequence change rate sequence are expressed as: , , in, For the The intrinsic mode functions, are the subsequences after VMD decomposition, is a change rate sequence, when there are two subsequences and , the gravitational force between subsequences is expressed as: , in, is the cosine similarity between two subsequences, and They are and The experimental group error, Compare the value of the previous sequence with the value of the next sequence and then add them together. is the number of values.

[0025] Furthermore, calculating the similarity between sequences by the gravitational algorithm also includes total gravitational calculation.

[0026] It should be noted that when there is IMF , different subsequences share The total gravitational force is expressed as: , , in, is the total gravitational force, is the gravitational value, is the gravitational coefficient, and are the gravity coefficient and gravity value of the last set of subsequence pairs.

[0027] It should also be noted that, assuming the frequency modulation instruction is , the front length is The sequence is used as input to predict the experimental group and unknown group where the actual and predicted values ​​for the test group produce the total error (MAPE) of the invention, and the invention predicts the value of the unknown group.

[0028] S2: Optimize the optimal parameters of VMD by improving the sparrow search algorithm.

[0029] Furthermore, the optimal parameters of VMD are optimized by improving the sparrow search algorithm, including the update of the discoverer position.

[0030] It should be noted that for In the problem of dimensional solution space, the position of each sparrow is expressed as: , in, Reason The initial population consists of sparrows. is the number of populations, is the dimension of the solution space, and the fitness is expressed as: , in, is the fitness value, and the discoverer position update in the sparrow is expressed as: , in, is the updated finder location, is the current iteration number, For the A sparrow in The location of the dimension, for A random number between is the maximum number of iterations of the population, is a random number that follows a normal distribution, is the warning value in the population, , is the warning value, .

[0031] It should also be noted that when When there are no natural enemies nearby, the discoverer can conduct a wide search for food. When the bird is spotted, it means some sparrows have discovered the danger and the current population needs to fly to other areas to forage for food.

[0032] Furthermore, the optimal parameter optimization of VMD by improving the sparrow search algorithm also includes follower position update.

[0033] It should be noted that the follower in the sparrow observes the discoverer's position and updates the position, which is expressed as: , in, is the updated follower position, is the worst position of sparrows in the current iteration sparrow population, is the optimal position for the discoverer in the sparrow population, for The matrix has random values ​​of 1 or -1. For the matrix The transpose of .

[0034] It should also be noted that when When , it means that the current location of the sparrow is poor, the food obtained is very little, and it needs to go to other places to forage. , it means to randomly find a place near the optimal position to forage.

[0035] Furthermore, the optimal parameter search for VMD by improving the sparrow search algorithm also includes the update of the warning behavior position.

[0036] It should be noted that the position update of the sparrows with early warning behavior in the sparrow population is expressed as: , in, The updated position of the sparrow with warning behavior, is the optimal position of the sparrow in the current environment, is the step size control parameter, for A uniform random number between is the denominator control parameter, is the current sparrow fitness value, is the best fitness value in the current sparrow population, is the worst fitness value of sparrow.

[0037] It should also be noted that when When , it means that the current sparrow position is very vulnerable to predators and needs to be moved to the global optimal position to avoid damage. , it means that the sparrow is already in the best position globally, but in order to avoid being hunted by predators, it needs to move to a position close to the current position.

[0038] S3: Output each optimal subsequence to obtain the final prediction result.

[0039] Furthermore, obtaining the final prediction result includes the optimal Generates a subsequence of values.

[0040] It should be noted that each subsequence is predicted separately through the GRU model, the predicted value output by the GRU is compared with the actual value, and the error is calculated, and the prediction results of each subsequence are added together to obtain the final prediction result.

[0041] Example 2, an embodiment of the present invention, provides a method for predicting energy storage frequency modulation instructions based on the gravity method. In order to verify the beneficial effects of the present invention, scientific demonstration is carried out through economic benefit calculation and simulation experiments.

[0042] Prediction tests were conducted on different frequency modulation sequences. The data used in the tests included the historical sequences of frequency modulation instructions of real power systems. All data were standardized after collection to more accurately evaluate the performance of the algorithm. To ensure the scientificity and fairness of the test, the data set was divided into a training set and a test set, of which 90% of the data was used for model training and 10% of the data was used to test the prediction performance.

[0043] During the experiment, the original FM sequence was first preprocessed, and the VMD method was used to decompose the sequence into multiple intrinsic mode functions (IMFs). To ensure the optimal parameters of VMD decomposition, the improved sparrow search algorithm (ISSA) was used to optimize the key parameters such as the number of decomposition layers and the penalty factor. The optimization process used the minimum total gravity and the minimum error of the test group as the objective function to determine the optimal decomposition parameters. The original FM sequence was decomposed into multiple independent subsequences according to the optimized parameters, and the gravity algorithm was used to calculate the similarity between the subsequences to ensure that the decomposition results had high independence and accuracy. For each decomposed subsequence, the ISSA algorithm was used to optimize the decomposition layer number and the penalty factor. The GRU network model is used for prediction. The training parameters of the GRU network include learning rate, number of hidden layer units and number of training steps. These parameters are tuned through experiments to adapt to the characteristics of different sequences. The prediction results of each subsequence are superimposed to generate the final frequency modulation instruction prediction value. The prediction result is compared with the actual value and evaluated by multiple indicators, including mean square error (MAE), residual sum of squares (SSE), root mean square error (RMSE) and mean absolute percentage error (MAPE). The experiment compares the method of the present invention with the traditional VMD-GRU method and predicts with the same data set and parameter configuration.

[0044] As shown in Table 1, the method of the present invention is superior to the traditional VMD-GRU method in all evaluation indicators. Taking FM sequence 1 as an example, the error of the present invention is reduced from 5.33 to 0.17 in MAE, and the error reduction rate is 96.8%. In terms of SSE indicator, the present invention is reduced by 80.3% compared with the traditional method, while RMSE and MAPE are also reduced by 54.1% and 52.9% respectively. A similar trend is also shown in FM sequence 2, especially in SSE, where the error is reduced by 97.7%. These differences fully illustrate the performance of the present invention in prediction accuracy and error control.

[0045] Table 1 Experimental comparison table , Example 3, reference Figure 3 , which is an embodiment of the present invention, provides a gravity method energy storage frequency modulation instruction prediction system, including: a similarity calculation module, a parameter optimization module, and an output prediction module.

[0046] The similarity calculation module is used to calculate the similarity between sequences through the gravity algorithm; the parameter optimization module is used to optimize the optimal parameters of VMD through the improved sparrow search algorithm; the output prediction module is used to output each optimal subsequence to obtain the final prediction result.

[0047] If the function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods of each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc. Various media that can store program codes.

[0048] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in conjunction with such instruction execution systems, devices or apparatuses. For the purposes of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in conjunction with such instruction execution systems, devices or apparatuses.

[0049] More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic devices), a portable computer disk case (magnetic devices), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be a paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering or, if necessary, processing in another suitable manner, and then stored in a computer memory.

[0050] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logical function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc. It should be noted that the above embodiments are only used to illustrate the technical solution of the present invention and are not limited. Although the present invention is described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that the technical solution of the present invention can be modified or replaced by equivalents without departing from the spirit and scope of the technical solution of the present invention, which should be included in the scope of the claims of the present invention.

[0051] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A method for predicting energy storage frequency modulation instructions using a gravity method, characterized in that: include: The similarity between sequences is calculated by the gravitational algorithm; The optimal parameters of VMD are optimized by improving the sparrow search algorithm; Output each optimal subsequence to obtain the final prediction result.

2. The energy storage frequency modulation instruction prediction method of the gravity method as claimed in claim 1, characterized in that: The calculation of similarity between sequences by the gravitational algorithm includes gravitational calculation between sequences. The subsequences after VMD decomposition of the original sequence and the sequence change rate sequence are expressed as: ， ， in, For the The intrinsic mode functions, are the subsequences after VMD decomposition, is a change rate sequence, when there are two subsequences and , the gravitational force between subsequences is expressed as: ， in, is the cosine similarity between two subsequences, and They are and The experimental group error, Compare the value of the previous sequence with the value of the next sequence and then add them together. is the number of values.

3. The energy storage frequency modulation instruction prediction method of the gravity method as claimed in claim 2, characterized in that: The calculation of similarity between sequences by gravity algorithm also includes total gravity calculation. IMF , different subsequences share The total gravitational force is expressed as: ， ， in, is the total gravitational force, is the gravitational value, is the gravitational coefficient, and are the gravity coefficient and gravity value of the last set of subsequence pairs.

4. The energy storage frequency modulation instruction prediction method of the gravity method as claimed in claim 3 is characterized in that: The method of optimizing the optimal parameters of VMD by improving the sparrow search algorithm includes updating the position of the discoverer. In the problem of dimensional solution space, the position of each sparrow is expressed as: ， in, Reason The initial population consists of sparrows. is the number of populations, is the dimension of the solution space, and the fitness is expressed as: ， in, is the fitness value, and the discoverer position update in the sparrow is expressed as: ， in, is the updated finder location, is the current iteration number, For the A sparrow in the The location of the dimension, for A random number between is the maximum number of iterations of the population, is a random number that follows a normal distribution, is the warning value in the population, , is the warning value, .

5. The energy storage frequency modulation instruction prediction method of the gravity method as claimed in claim 4, characterized in that: The optimization of VMD parameters by improving the sparrow search algorithm also includes follower position update. The follower in the sparrow observes the finder position and updates the position, which is expressed as: ， in, is the updated follower position, is the worst position of sparrows in the current iteration sparrow population, is the optimal position for the discoverer in the sparrow population, for The matrix has random values ​​of 1 or -1. For the matrix The transpose of .

6. The energy storage frequency modulation instruction prediction method of the gravity method as claimed in claim 5, characterized in that: The optimization of the optimal parameters of VMD by improving the sparrow search algorithm also includes updating the position of the early warning behavior. The position update of the sparrows with early warning behavior in the sparrow population is expressed as: ， in, The updated position of the sparrow with warning behavior, is the optimal position of the sparrow in the current environment, is the step size control parameter, for A uniform random number between is the denominator control parameter, is the current sparrow fitness value, is the best fitness value in the current sparrow population, is the worst fitness value of sparrows.

7. The energy storage frequency modulation instruction prediction method of the gravity method as claimed in claim 6, characterized in that: The final prediction result is obtained by The predicted value of each subsequence is compared with the actual value, and the error is calculated. The predicted results of each subsequence are added together to obtain the final prediction result.

8. A gravity method energy storage frequency modulation instruction prediction system, using any method as claimed in claim 1 to 7, characterized in that: Including similarity calculation module, parameter optimization module, output prediction module; The similarity calculation module is used to calculate the similarity between sequences through a gravitational algorithm; The parameter optimization module is used to optimize the optimal parameters of VMD by improving the sparrow search algorithm; The output prediction module is used to output each optimal subsequence to obtain a final prediction result.

9. An electronic device comprising a memory and a processor, characterized in that : The memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions. When the computer-executable instructions are executed by the processor, the steps of the method as described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having computer-executable instructions stored thereon, characterized in that: When the computer executable instructions are executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images