Energy storage frequency modulation instruction prediction method and system of centripetal force method

Abstract

The invention discloses an energy storage frequency modulation instruction prediction method and system based on a centripetal force method, and relates to the technical field of artificial intelligence in a power system, and the method comprises the steps: decomposing an original frequency modulation sequence through VMD, decomposing the original frequency modulation sequence into a plurality of subsequences, setting the number of decomposition layers as K, recognizing convex points and concave points in the sequence, and carrying out the prediction of an energy storage frequency modulation instruction. Obtaining a resultant force of each sub-sequence to evaluate features; evaluating the centripetal force direction and magnitude of each subsequence decomposed by the VMD by adopting a centripetal force method, and determining the maximum centripetal force as a target function; and an improved frost ice optimization algorithm is applied to optimize the decomposition layer number K of the VMD so as to maximize the objective function and obtain the optimal decomposition effect. According to the invention, the processing efficiency and accuracy of the frequency modulation signal are effectively improved, and important support is provided for real-time prediction of a frequency modulation instruction in an energy management system.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence in power systems, and in particular to a method and system for predicting energy storage frequency modulation instructions using a centripetal force method. Background Art

[0002] The traditional frequency modulation method of hybrid energy storage (supercapacitor + lithium battery) to assist thermal power units is to transfer the difference between the frequency modulation command and the thermal power unit to the hybrid energy storage, where the low-frequency part is borne by the battery and the high-frequency part is borne by the supercapacitor. However, the transmission of the signal (frequency modulation command to supercapacitor / lithium battery) takes time, and the response of the supercapacitor or lithium battery itself also takes time, which will cause a certain response time difference, further affecting the profit value, and further affecting the profit of the power plant. For this reason, the present invention proposes a prediction method to predict the size of the frequency modulation command in advance, so that the supercapacitor / battery can act in advance to increase the profit value.

[0003] However, traditional prediction methods have two disadvantages: 1. Lack of assessment of the difficulty of predicting the original data: The original sequence of frequency modulation instructions is highly nonlinear and irregular. The stronger the irregularity, the more difficult it is to predict. Assessment of the difficulty of prediction is conducive to the adjustment of subsequent predictions.

[0004] 2. The prediction network lacks personalized customization for different sequences. Different sequences have different complexities. If the same parameters are used, the prediction results will not be ideal.

[0005] In view of these two points, the present invention proposes an original sequence prediction difficulty analysis based on the centripetal force method and proposes a selection of the coefficient of VMD decomposition K that is most suitable for the centripetal force method based on the centripetal force method. Summary of the invention

[0006] In view of the above existing problems, the present invention adopts VMD decomposition for the original sequence. Different decomposition levels of VMD decomposition have different corresponding prediction accuracies. Therefore, the present invention adopts the centripetal force method to evaluate the strength of the centripetal force of each subsequence of VMD decomposition, and then takes the maximum centripetal force as the objective function, and adopts the improved frost ice optimization algorithm (IRIME) to optimize the parameters of VMD.

[0007] In order to solve the above technical problems, a centripetal force method for predicting energy storage frequency modulation instructions is proposed, including: The original FM sequence is decomposed into multiple subsequences through VMD decomposition. The decomposition layer number is set to K. The convex points and concave points in the sequence are identified, and the resultant force of each subsequence is obtained to evaluate the characteristics. The centripetal force method is used to evaluate the direction and magnitude of the centripetal force of each subsequence decomposed by VMD, and the maximum centripetal force is determined as the objective function. The improved frost optimization algorithm is applied to optimize the decomposition layer number K of VMD to maximize the objective function and obtain the best decomposition effect.

[0008] As a preferred solution of the energy storage frequency modulation instruction prediction method of the centripetal force method described in the present invention, wherein: the VMD decomposition of the original frequency modulation sequence includes decomposing the original frequency modulation sequence, assuming the number of decomposition layers is K, into K subsequences , the resultant force of each subsequence is , the total force is ; The centripetal force method evaluation includes setting any subsequence of any original sequence after VMD decomposition to , each group of three consecutive points is read through the entire sequence. When the middle point among the three consecutive points is larger than the two adjacent points, the current point is determined to be a convex point. When the middle point among the three consecutive points is the smallest, the current point is determined to be a concave point. When a convex or concave point is determined in the sequence, it is allowed to find the point of centripetal force.

[0009] As a preferred solution of the energy storage frequency modulation instruction prediction method of the centripetal force method described in the present invention, wherein: the direction and magnitude of the centripetal force include: If it is a convex point, the centripetal force direction of the current convex point determines the center position of the circle, and the sequence The mean of is A, and the point closest to A ,sequence The median of is B, and the point closest to B , then the center of the circle is , and Belongs to M, then the direction of the current convex point is: , when If it is a concave point, the direction of the current concave point is: , in, Indicates the direction of the centripetal force of a convex or concave point, i represents the i-th value, s represents the subscript of the value closest to A, and q represents the subscript of the value closest to B.

[0010] As a preferred solution of the energy storage frequency modulation instruction prediction method of the centripetal force method described in the present invention, wherein: the centripetal force direction and magnitude also include, when is a sequence of convex or concave points The mean of is A, the sequence The median of is B, then the centripetal force is: , Among them, M is The length of the subsequence, is the magnitude of the centripetal force.

[0011] As a preferred solution of the energy storage frequency modulation instruction prediction method of the centripetal force method described in the present invention, wherein: the determination of the maximum centripetal force as the objective function includes solving the centripetal force and the direction of the corresponding force corresponding to all convex points or concave points in the entire sequence, and the magnitude of the final resultant force is: , Among them, c is the number of all concave and convex points set .

[0012] As a preferred solution of the energy storage frequency modulation instruction prediction method of the centripetal force method described in the present invention, the improved frost and ice optimization algorithm includes optimizing the K value of VMD by the improved frost and ice optimization algorithm and performing the soft frost search strategy before the improvement: , in, is the updated position of the particle, ab represents the bth particle of the ath rime-agent; is the best b-th particle in the population, is a parameter representing a random number in the range (-1, 1), and r controls the direction of particle movement. Will change with the number of iterations: , is the environmental factor; h is the adhesion, which is a random number in the range of (0, 1); and are the upper and lower bounds of the escape space, respectively, and E is the adhesion coefficient, which increases with the number of iterations; For parameters A random number in the range of (0, 1) determines whether the particle position is updated; For the soft cream search strategy before improvement, two types of improvements are made; Improvement 1 is to change the randomized number to a specific one. For a random number between [-1 1], it is specified as: , in, is the total force of the sum of the forces of each subsequence; Improvement 2 is to Improved to: , Among them, t is the current iteration number, and T is the maximum iteration number of the algorithm.

[0013] As a preferred solution of the energy storage frequency modulation instruction prediction method of the centripetal force method described in the present invention, the optimization of the decomposition layer number K of VMD includes optimizing the K value in VMD using an improved algorithm, outputting each subsequence corresponding to the optimal K, solving the objective function, outputting K prediction results and adding them to obtain the K value corresponding to the maximum value of the total resultant force.

[0014] Another object of the present invention is to provide a storage frequency modulation instruction prediction system based on the centripetal force method. The present invention aims to solve the technical problems in frequency modulation signal processing through the collaborative work of the VMD decomposition module, the centripetal force evaluation module and the frost and ice optimization improvement module, including optimizing the number of decomposition layers to improve the efficiency of time-frequency analysis, accurately identifying signal characteristics, and automatically adjusting parameters through centripetal force evaluation and frost and ice optimization algorithms to achieve the best signal decomposition effect, thereby improving the accuracy and automation level of signal processing.

[0015] As a preferred solution of the energy storage frequency modulation instruction prediction system of the centripetal force method described in the present invention, it is characterized by comprising a VMD decomposition module, a centripetal force evaluation module, and a frost and ice optimization and improvement module; The VMD decomposition module decomposes the original FM sequence into multiple subsequences by VMD decomposition, sets the number of decomposition layers to K, identifies convex points and concave points in the sequence, and obtains the resultant force of each subsequence to evaluate the features; The centripetal force evaluation module uses the centripetal force method to evaluate the centripetal force direction and magnitude of each subsequence decomposed by VMD, and determines the maximum centripetal force as the objective function; The frost ice optimization improvement module applies the improved frost ice optimization algorithm to optimize the decomposition layer number K of VMD to maximize the objective function and obtain the best decomposition effect.

[0016] A computer device includes a memory and a processor, wherein the memory stores a computer program, and is characterized in that when the processor executes the computer program, the steps of the energy storage frequency modulation instruction prediction method of the centripetal force method are implemented.

[0017] 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 energy storage frequency modulation instruction prediction method of the centripetal force method are implemented.

[0018] Beneficial effects of the present invention: The present invention decomposes the original frequency modulation sequence through VMD, converts complex signals into easy-to-process subsequences, facilitates detailed analysis, and improves prediction accuracy. The centripetal force method is used to evaluate the convex and concave points in the subsequence, accurately locates the signal characteristics, and provides data support for subsequent optimization. By solving the centripetal force maximization and integrating the centripetal forces of multiple key points, the signal strength is optimized globally to enhance the credibility of the overall prediction. The improved frost and ice optimization algorithm is applied to dynamically optimize the decomposition layer number K to achieve the best decomposition effect, ensuring that the system has more flexible information extraction capabilities. Finally, this method effectively improves the processing efficiency and accuracy of frequency modulation signals, and provides important support for the real-time prediction of frequency modulation instructions in energy management systems. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of 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 work.

[0020] Figure 1 An overall flow chart of a method for predicting energy storage frequency modulation instructions using a centripetal force method is provided for one embodiment of the present invention.

[0021] Figure 2 A prediction flow chart of a method for predicting energy storage frequency modulation instructions using a centripetal force method provided by an embodiment of the present invention.

[0022] Figure 3 A system solution flow chart of a centripetal force method energy storage frequency modulation instruction prediction system provided in one embodiment of the present invention. DETAILED DESCRIPTION

[0023] 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.

[0024] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0025] Example 1, reference Figure 1-Figure 2, which is the first embodiment of the present invention, and which provides a method for predicting energy storage frequency modulation instructions using a centripetal force method, comprising: S1: Decompose the original FM sequence into multiple subsequences by VMD, set the decomposition layer to K, identify the convex and concave points in the sequence, and obtain the resultant force of each subsequence to evaluate the features. VMD is a signal decomposition method, namely variational mode decomposition, which decomposes the original FM sequence into multiple subsequences.

[0026] Because the sequence is irregular, the present invention is inspired by the centripetal force in physics and proposes a set of processes to evaluate the strength of the regularity of a sequence. The stronger the centripetal force, the stronger the regularity, and vice versa.

[0027] Furthermore, the original FM sequence is decomposed into K subsequences with the decomposition layer number K. , the resultant force of each subsequence is , the total force is ,in, is the resultant force of the Kth subsequence; The centripetal force method evaluation includes setting any subsequence of any original sequence after VMD decomposition to , each group of three consecutive points is read through the entire sequence. When the middle point among the three consecutive points is larger than the two adjacent points, the current point is determined to be a convex point. When the middle point among the three consecutive points is the smallest, the current point is determined to be a concave point. When convex or concave points are identified in the sequence, it is allowed to find the centripetal points because such points have more variations.

[0028] S2: The centripetal force method is used to evaluate the direction and magnitude of the centripetal force of each subsequence of VMD decomposition, and the maximum centripetal force is determined as the objective function.

[0029] Furthermore, when If it is a convex point, the centripetal force direction of the current convex point determines the center position of the circle, and the sequence The mean of is A, and the point closest to A ,sequence The median of is B, and the point closest to B , then the center of the circle is , and Belongs to M, then the direction of the current convex point is: , when If it is a concave point, the direction of the current concave point is: , in, Indicates the direction of the centripetal force of a convex or concave point, i represents the variable index, that is, the i-th value; s represents the subscript of the value closest to A, and q represents the subscript of the value closest to B.

[0030] when is a sequence of convex or concave points The mean of is A, the sequence The median of is B, then the centripetal force is: , Among them, M is The length of the subsequence, is the magnitude of the centripetal force.

[0031] It should be noted that the centripetal force and the direction of the corresponding force corresponding to all convex or concave points in the entire sequence are solved, and the magnitude of the final resultant force F is: , Among them, c is the number of all concave and convex points set .

[0032] S3: Apply the improved frost optimization algorithm to optimize the number of decomposition levels K of VMD to maximize the objective function and obtain the best decomposition effect.

[0033] Furthermore, the K value of VMD is optimized by the improved frost optimization algorithm, and the soft frost search strategy before the improvement is performed: , in, is the updated position of the particle, ab represents the bth particle of the ath rime-agent; is the best b-th particle in the population, is a parameter representing a random number in the range (-1, 1), and r controls the direction of particle movement. Will change with the number of iterations: , is the environmental factor, which simulates the influence of the external environment by following the number of iterations; h is the adhesion, a random number in the range of (0, 1), which controls the distance between the centers of two fog particles; and are the upper and lower bounds of the escape space, respectively, and E is the adhesion coefficient, which increases with the number of iterations; For parameters A random number in the range of (0, 1) determines whether the particle position is updated; For the soft cream search strategy before improvement, two types of improvements are made; Improvement 1 is to change the randomized number to a specific one. For a random number between [-1 1], it is specified as: , in, is the total force of the sum of the forces of each subsequence; Improvement 2 is to Improved to: , Among them, t is the current iteration number, and T is the maximum iteration number of the algorithm.

[0034] The improved algorithm is used to optimize the K value in VMD, and the corresponding subsequences under the optimal K are output. The objective function is solved through the GRU gated loop, and the K prediction results are output and added to obtain the K value corresponding to the maximum total force.

[0035] In order to further verify the advantages of the present invention, the present invention uses the method of the present invention and the prediction method of VMD-GRU to predict the frequency modulation sequence, and the four evaluation indicators are: Indicator one is MAE, which is the mean absolute prediction error of the prediction results: , The second indicator is SSE, which reflects the discrete state of each observation value of each sample, also known as the within-group sum of squares or residual sum of squares: , The third indicator is RMSE, which is the root mean square prediction error: , Indicator 4 is MAPE, which is the mean of absolute errors: , Where N represents the sample size, and Represent the actual value and predicted value at time n respectively.

[0036] The results are shown in Table 1: Table 1 Frequency modulation evaluation index results , Embodiment 2, the second embodiment of the present invention, is different from the previous embodiment in that: If the functions are implemented in the form of software functional units and sold or used as independent products, they 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 method described in 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., which can store program codes.

[0037] 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.

[0038] More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic device), a portable computer disk case (magnetic device), 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.

[0039] 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 logic 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.

[0040] Example 3, reference Figure 3 , which is the third embodiment of the present invention, and which provides a centripetal force method energy storage frequency modulation instruction prediction system, including a VMD decomposition module, a centripetal force evaluation module, and a frost and ice optimization improvement module; The VMD decomposition module decomposes the original FM sequence into multiple subsequences by VMD decomposition, sets the number of decomposition layers to K, identifies convex points and concave points in the sequence, and obtains the resultant force of each subsequence to evaluate the features; The centripetal force evaluation module uses the centripetal force method to evaluate the centripetal force direction and magnitude of each subsequence decomposed by VMD, and determines the maximum centripetal force as the objective function; The frost ice optimization improvement module applies the improved frost ice optimization algorithm to optimize the decomposition layer number K of VMD to maximize the objective function and obtain the best decomposition effect.

[0041] 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 the present invention. 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 centripetal force method, characterized in that: include, Decompose the original FM sequence into multiple subsequences by VMD, set the decomposition layer to K, identify the convex points and concave points in the sequence, and obtain the resultant force of each subsequence to evaluate the features; The centripetal force method is used to evaluate the centripetal force direction and magnitude of each subsequence of VMD decomposition, and the maximum centripetal force is determined as the objective function; The improved frost optimization algorithm is applied to optimize the number of decomposition layers K of VMD to maximize the objective function and obtain the best decomposition effect; The improved frost optimization algorithm includes optimizing the K value of VMD through the improved frost optimization algorithm and performing the soft frost search strategy before the improvement: ， in, is the updated position of the particle, ab represents the bth particle of the ath rime-agent; is the best b-th particle in the population, is a parameter representing a random number in the range (-1, 1), and r controls the direction of particle movement. Will change with the number of iterations: ， is the environmental factor; h is the adhesion, which is a random number in the range of (0, 1); and are the upper and lower bounds of the escape space, respectively, and E is the adhesion coefficient, which increases with the number of iterations; For parameters A random number in the range of (0, 1) determines whether the particle position is updated.

2. The method for predicting energy storage frequency modulation instructions using the centripetal force method according to claim 1, characterized in that: The VMD decomposition of the original frequency modulation sequence includes decomposing the original frequency modulation sequence into K subsequences, assuming the number of decomposition layers is K. , the resultant force of each subsequence is , the total force is ,in, is the resultant force of the Kth subsequence; The centripetal force method evaluation includes setting any subsequence of any original sequence after VMD decomposition to , where i is the variable index, indicating the i-th value. Each group of three consecutive points goes through the entire sequence. When the middle point among three consecutive points is larger than the two adjacent points, the current point is determined to be a convex point. When the middle point among three consecutive points is the smallest, the current point is determined to be a concave point. When a convex or concave point is determined in the sequence, it is allowed to find the point of centripetal force.

3. The energy storage frequency modulation instruction prediction method of the centripetal force method as claimed in claim 2, characterized in that: The centripetal force direction and magnitude include: If it is a convex point, the centripetal force direction of the current convex point determines the center position of the circle, and the sequence The mean of is A, and the point closest to A ,sequence The median of is B, and the point closest to B , then the center of the circle is , and Belongs to M, where M is the length of any subsequence after VMD decomposition, then the direction of the current point is: ， when If it is a concave point, the direction of the current concave point is: ， in, Indicates the direction of the centripetal force of a convex or concave point, i represents the i-th value, s represents the subscript of the value closest to A, and q represents the subscript of the value closest to B.

4. The method for predicting energy storage frequency modulation instructions using the centripetal force method as claimed in claim 3, characterized in that: The centripetal force direction and magnitude also include when is a sequence of convex or concave points The mean of is A, the sequence The median of is B, then the centripetal force is: ， Among them, M is The length of the subsequence, is the magnitude of the centripetal force.

5. The method for predicting energy storage frequency modulation instructions by centripetal force method according to claim 4, characterized in that: Determining the maximum centripetal force as the objective function includes solving the centripetal force and the direction of the corresponding force corresponding to all convex points or concave points in the entire sequence, and the magnitude of the final resultant force is: ， Among them, c is the number of all concave and convex points set .

6. The method for predicting energy storage frequency modulation instructions using the centripetal force method as claimed in claim 5, characterized in that: The improved frost and ice optimization algorithm includes two types of improvements to the soft frost search strategy before the improvement; Improvement 1 is to change the randomized number to a specific one. For a random number between [-1 1], it is specified as: ， in, is the total force of the sum of the forces of each subsequence; Improvement 2 is to Improved to: ， Among them, t is the current iteration number, and T is the maximum iteration number of the algorithm.

7. The method for predicting energy storage frequency modulation instructions using the centripetal force method as claimed in claim 6, characterized in that: The optimization of the decomposition layer number K of VMD includes optimizing the K value in VMD using an improved algorithm, outputting each subsequence corresponding to the optimal K, solving the objective function, outputting K prediction results and adding them to obtain the K value corresponding to the maximum value of the total resultant force.

8. A system for predicting energy storage frequency modulation instructions using a centripetal force method as claimed in any one of claims 1 to 7, characterized in that: Including VMD decomposition module, centripetal force evaluation module, frost and ice optimization and improvement module; The VMD decomposition module decomposes the original FM sequence into multiple subsequences by VMD decomposition, sets the number of decomposition layers to K, identifies convex points and concave points in the sequence, and obtains the resultant force of each subsequence to evaluate the features; The centripetal force evaluation module uses the centripetal force method to evaluate the centripetal force direction and magnitude of each subsequence decomposed by VMD, and determines the maximum centripetal force as the objective function; The frost ice optimization improvement module applies the improved frost ice optimization algorithm to optimize the decomposition layer number K of VMD to maximize the objective function and obtain the best decomposition effect.

9. 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 a method for predicting energy storage frequency modulation instructions using a centripetal force method according to any one of claims 1 to 7 are implemented.

10. 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 centripetal force method according to any one of claims 1 to 7 are implemented.

Images