Prediction method and system for optimizing frequency modulation instruction of fused salt heat storage system

Abstract

The invention relates to the technical field of power grid frequency modulation, in particular to a prediction method and system for optimizing a frequency modulation instruction of a fused salt heat storage system, and the method comprises the steps: carrying out the clustering of all sample data points, and obtaining a plurality of class clusters; adjusting abnormal sample data points in all the sample data points through the discrete degree of all the data in the plurality of class clusters to obtain target data points; dimension reduction is carried out on the time series data of all dimensions corresponding to all the target data points, and reference features after dimension reduction are obtained; obtaining a plurality of reference data points through the frequency modulation instruction signal data at a plurality of moments and various reference feature data; through a plurality of reference data points, obtaining errors corresponding to different sample sizes of each decision tree; and obtaining the optimal sample size of each decision-making tree according to the trend change of the corresponding error under different sample sizes of each decision-making tree, and predicting instruction signal data. The method improves the prediction precision of the frequency modulation instruction, and reduces the calculation amount.

Description

Technical Field

[0001] The present invention relates to the technical field of power grid frequency modulation, and particularly relates to a method and system for optimizing the prediction of frequency modulation commands for a molten salt thermal energy storage system. Background Art

[0002] The molten salt thermal energy storage peak shaving and frequency modulation system combines molten salt energy storage technology with the peak shaving and frequency modulation functions of the power system, which can balance the supply and demand fluctuations of energy and improve the stability of the power grid. The molten salt thermal energy storage peak shaving and frequency modulation system is mainly used to store excess electric energy and release the stored energy during peak demand periods, thereby balancing the load fluctuations of the power grid. Molten salt thermal energy storage stores thermal energy by heating molten salt; when needed, the molten salt can release thermal energy to drive thermoelectric conversion equipment (such as steam turbines, combined heat and power, etc.) to generate electricity.

[0003] To ensure the stability and reliability of the power grid and solve problems such as power grid load and power generation fluctuations, unstable renewable energy power generation, and frequency fluctuations. By predicting the frequency modulation command signal, the generator output and energy storage system can be adjusted in advance to avoid frequency anomalies, system overloads, and large-scale power outages, thereby improving the stability of the power grid, reducing operating costs, and enhancing the power grid's ability to respond to emergencies.

[0004] When predicting the frequency modulation command signal using existing methods, a decision tree can be constructed through the random forest algorithm for prediction. However, in the conventional prediction process, the data volume of each decision tree and the depth of the decision tree are both set manually. During the prediction process, when the data volume of each decision tree is small, the prediction result may be poor, reducing the prediction accuracy; when the data volume of each decision tree is large, although the prediction accuracy is improved, the computational complexity also increases. Moreover, because the data volume of each decision tree is large, the model will memorize every detail in the historical data, and when new data appears, the generalization ability of the model will be poor. And a relatively large depth of the decision tree may lead to overfitting, while a relatively small depth may lead to underfitting. By reasonably controlling the depth of the decision tree, a balance can be found between avoiding overfitting and underfitting. Summary of the Invention

[0005] The present invention provides a method and system for optimizing the prediction of frequency modulation commands for a molten salt thermal energy storage system, which are used to solve the problems of unstable prediction accuracy, increased computational complexity, and overfitting or underfitting caused by improper setting of the data volume and depth of the decision tree in existing methods.

[0006] The object of the present invention can be achieved by the following technical solutions: In the first aspect of the present invention, a method for optimizing the prediction of frequency modulation commands for a molten salt thermal energy storage system is provided, including: Continuously obtain the frequency modulation command signal data at several moments and various characteristic data affecting the stability of the power system; obtain several sample data points through the frequency modulation command signal data and various characteristic data at several moments; Cluster all the sample data points to obtain several clusters; screen out the abnormal sample data points from all the sample data points according to the dispersion degree of all the data in several clusters to obtain the target sample points; perform interpolation through all the target sample points to obtain the target data points; perform dimensionality reduction on the time series data corresponding to each dimension of all the target data points to obtain the reference features after dimensionality reduction; obtain several reference data points through the frequency modulation command signal data and various reference feature data at several moments; Obtain the corresponding errors of each decision tree under different sample sizes through several reference data points; obtain the optimal sample size of each decision tree according to the trend change of the corresponding errors of each decision tree under different sample sizes; Predict the command signal data according to the optimal sample size of each decision tree through the random forest algorithm.

[0007] Further, the obtaining of several sample data points through the frequency modulation command signal data and various characteristic data at several moments includes: Form a sample data point with the data of all dimensions at each moment, and several sample data points can be obtained; wherein, all dimensions include all characteristic data and frequency modulation command signal data.

[0008] Further, the clustering of all the sample data points to obtain several clusters includes: Cluster all the sample data points using the K-means clustering algorithm to obtain several clusters; wherein, the number of clusters is obtained by the elbow method.

[0009] Further, the screening out of the abnormal sample data points from all the sample data points according to the dispersion degree of all the data in several clusters to obtain the target sample points; performing interpolation through all the target sample points to obtain the target data points includes: Calculate and obtain the abnormal outlier factor of each sample data point through the LOF algorithm; record the mean value of the abnormal outlier factors of all the sample data points in each cluster as the noise abnormal factor of each cluster; record all the sample data points corresponding to the cluster with the largest noise abnormal factor among all the clusters as the abnormal sample data points; screen out the abnormal sample data points from all the sample data points, and record the remaining sample data points as the target sample points; Perform interpolation on all the target sample points through the linear interpolation method to obtain the interpolated data points; form the target data points with the interpolated data points and all the target sample points.

[0010] Further, dimensionality reduction is performed on the time-series data of each dimension corresponding to all target data points to obtain the reference features after dimensionality reduction; several reference data points are obtained from the frequency modulation command signal data and various reference feature data at several moments, including: Perform dimensionality reduction on the time-series data of each dimension corresponding to all target data points through the PCA algorithm to obtain the features after dimensionality reduction, denoted as reference features; Form a reference data point by combining all reference feature data and frequency modulation command signal data at each moment, then several reference data points can be obtained.

[0011] Further, the error corresponding to different sample sizes of each decision tree is obtained through several reference data points, including: Sort several reference data points in chronological order to form a sequence, denoted as the data point sequence; take the first reference data points in the data point sequence to form a sequence, denoted as the front data point sequence; take the last reference data points in the data point sequence to form a sequence, denoted as the back data point sequence; form a sequence by combining all the frequency modulation command signal data corresponding to the reference data points in the back data point sequence, denoted as the back command signal data sequence; where, is the preset division percentage factor; Randomly select a preset parameter data points from the front data point sequence with replacement to form a data set, and then select times to form data sets; construct decision trees from data sets, predict the frequency modulation command signal data through the random forest constructed by decision trees, and form a sequence by arranging the predicted frequency modulation command signal data in chronological order, denoted as the first predicted command signal data sequence; calculate the error between the first predicted command signal data sequence and the back command signal data sequence through MAPE, denoted as the first error ; Take the first error as the error corresponding to the sample size of each decision tree being ; Randomly select a preset parameter data points from the front data point sequence with replacement to form a data set, and then select times to form data sets; construct decision trees from data sets, predict the frequency modulation command signal data through the random forest constructed by ​​​​Predict the frequency modulation command signal data using a random forest constructed by decision trees, and form a sequence of the predicted frequency modulation command signal data in chronological order, denoted as the second predicted command signal data sequence; calculate the error between the second predicted command signal data sequence and the post-command signal data sequence through MAPE, denoted as the second error ; Take the second error as the error corresponding to when the sample size of each decision tree is ; And so on, successively obtain the errors corresponding to different sample sizes of each decision tree; Among them, represents the preset number of decision trees, represents rounding up ; represents the number of all data points in the previous data point sequence.

[0012] Furthermore, obtaining the optimal sample size of each decision tree through the trend change of the errors corresponding to different sample sizes of each decision tree includes: Construct a reference coordinate system with the sample size of each decision tree as the horizontal axis and the error corresponding to the sample size of each decision tree as the vertical axis; map the different sample sizes and corresponding errors of each decision tree in the reference coordinate system to obtain several coordinate points; perform curve fitting on the several coordinate points by the least squares method to obtain an error fitting curve; Obtain the inflection point where the second derivative of the function corresponding to the error fitting curve is equal to 0, and take as the optimal sample size of each decision tree; among them, represents the horizontal value of the inflection point, represents rounding up ;

[0013] The second aspect of the present invention is to provide a system for optimizing the frequency modulation command prediction of a molten salt thermal energy storage system, including: Data acquisition module: used to continuously obtain frequency modulation command signal data and various characteristic data affecting the stability of the power system at several moments; the various characteristic data includes humidity, temperature, wind speed, and load; obtain several sample data points through the frequency modulation command signal data and various characteristic data at several moments; Data processing module: used to cluster all sample data points to obtain several clusters; screen out abnormal sample data points from all sample data points according to the dispersion degree of all data in the several clusters to obtain target sample points; perform interpolation through all target sample points to obtain target data points; perform dimensionality reduction on the time series data of each dimension corresponding to all target data points to obtain the reference features after dimensionality reduction; obtain several reference data points through the frequency modulation command signal data and various reference feature data at several moments; Error analysis module: used to obtain the corresponding errors of each decision tree under different sample sizes through a number of reference data points; obtain the optimal sample size of each decision tree through the trend changes of the corresponding errors of each decision tree under different sample sizes; Prediction module: used to predict the command signal data through the random forest algorithm according to the optimal sample size of each decision tree.

[0014] The third aspect of the present invention is to provide an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the method for optimizing the frequency modulation command prediction of the molten salt energy storage system is implemented.

[0015] The fourth aspect of the present invention is to provide a computer-readable storage medium. The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method for optimizing the frequency modulation command prediction of the molten salt energy storage system is implemented.

[0016] Compared with the prior art, the beneficial effects of the present invention are as follows: clustering all sample data points to obtain a number of clusters; screening out abnormal sample data points from all sample data points through the dispersion degree of all data in a number of clusters to obtain target sample points; interpolating through all target sample points to obtain target data points, reducing the influence of noise data; performing dimensionality reduction on the time series data corresponding to each dimension of all target data points to obtain the reference features after dimensionality reduction; obtaining a number of reference data points through the frequency modulation command signal data and various reference feature data at a number of moments, reducing the calculation amount through dimensionality reduction; obtaining the corresponding errors of each decision tree under different sample sizes through a number of reference data points; obtaining the optimal sample size of each decision tree through the trend changes of the corresponding errors of each decision tree under different sample sizes; predicting the command signal data through the random forest algorithm according to the optimal sample size of each decision tree, improving the accuracy of the frequency modulation command prediction and reducing the calculation amount. Brief Description of the Drawings

[0017] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained without creative efforts based on these drawings.

[0018] Figure 1 It is a schematic flow chart of the steps of a method for optimizing the frequency modulation command prediction of a molten salt energy storage system provided by the present invention; Figure 2This invention provides a schematic diagram of the module process for optimizing the frequency modulation command prediction system of a molten salt energy storage system. Detailed implementation manners

[0019] In order to enable those skilled in the art to better understand the solution of this invention, the technical solutions in the embodiments of this invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of this invention. Obviously, the described embodiments are only a part of the embodiments of this invention, rather than all of the embodiments. Based on the embodiments of this invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the scope of protection of this invention.

[0020] It should be noted that the terms "first", "second", etc. in the description and claims of this invention and the above-mentioned drawings are used to distinguish similar objects, and do not necessarily need to describe a specific order or sequence. It should be understood that such data can be interchanged under appropriate circumstances, so that the embodiments of this invention described here can be implemented in an order different from those illustrated or described here. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device that includes a series of steps or units does not necessarily need to be limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0021] In response to the problems existing in the background technology, it is of great practical significance to research and design a method and system for optimizing the frequency modulation command prediction of a molten salt energy storage system.

[0022] As Figure 1 shown, the first aspect of this invention is to provide a method for optimizing the frequency modulation command prediction of a molten salt energy storage system, including the following steps: Step S001: Continuously collect the frequency modulation command signal data at several moments and various characteristic data affecting the stability of the power system.

[0023] It should be noted that when there is a deviation in the power supply and demand balance (such as load changes, power generation fluctuations, etc.). The dispatching center or automation system of the power system generates a frequency modulation command according to the frequency fluctuation situation, load demand and power generation situation of the system, and maintains frequency stability through the frequency modulation command, which can respond to emergencies and reduce the possibility of failures.

[0024] Furthermore, it should be noted that during the normal power generation process of the power system, the power system is vulnerable to interference from the environment, weather, etc. When the power system is interfered, its stability is poor; when the power system is interfered, in order to maintain the stability of the system, the molten salt thermal energy storage peak shaving and frequency modulation system will perform frequency modulation and peak shaving to maintain the stability of the power system; therefore, the frequencies in the command signal data corresponding to different environmental conditions are different. Therefore, the frequencies in the command signal data can be determined through environmental and weather data, so it is necessary to collect all types of data related to the frequencies in the command signal data.

[0025] Specifically, at a preset time interval to obtain the frequency modulation command signal data and various characteristic data of all moments of the power system dispatching center for a preset duration hours before the current moment; among them, various characteristic data include humidity, temperature, wind speed, and load, etc.

[0026] Combining the data of all dimensions at each moment into a sample data point, several sample data points can be obtained; among them, all dimensions include all characteristic data and frequency modulation command signal data.

[0027] Among them, in this embodiment, the preset duration , where, in this embodiment, the preset duration is not specifically limited, and the implementer can determine it according to the specific situation. Among them, in this embodiment, the preset time interval seconds, where the preset time interval is not specifically limited, and the implementer can determine it according to the specific situation.

[0028] So far, several sample data points are obtained.

[0029] Step S002: Cluster all the sample data points to obtain several clusters; screen out abnormal sample data points from all the sample data points according to the dispersion degree of all the data in the several clusters to obtain target sample points; perform interpolation through all the target sample points to obtain target data points; perform dimensionality reduction on the time series data of each dimension corresponding to all the target data points to obtain the reference features after dimensionality reduction; obtain several reference data points through the frequency modulation command signal data and various reference feature data of several moments.

[0030] It should be noted that in order to determine the prediction accuracy of each constructed decision tree and the number of layers of the corresponding decision tree, the characteristic data affecting the frequency modulation command signal data is analyzed. However, the redundancy among some characteristic data is very high. Constructing a decision tree with a higher number of layers will only increase the computational amount and has little impact on the improvement of the performance of predicting the frequency modulation command signal data. Therefore, the number of layers of the decision tree can be optimized by reducing the data of some characteristic dimensions through dimensionality reduction. In order to determine the prediction accuracy of each constructed decision tree and the number of leaf nodes of the corresponding decision tree, the prediction analysis error is analyzed by iteratively analyzing the number of leaf nodes of each decision tree in turn to determine the optimal number of leaf nodes of the decision tree.

[0031] Furthermore, it should be noted that due to the interference of noise during the data acquisition process, there is a certain difference between the acquired data and the real data. Therefore, denoising is required. When there is a large distance between a sample data point and other sample data points among all the acquired sample data points, it indicates that the sample data point is interfered by noise. Therefore, denoising processing is performed based on the distribution of all sample data points.

[0032] Specifically, all sample data points are clustered using the K-means clustering algorithm (K-means clustering) to obtain several clusters; among them, the number of clusters is obtained by the elbow method; both the K-means clustering algorithm and the elbow method are well-known technologies and will not be specifically elaborated here.

[0033] The local outlier factor (LOF) algorithm is used to calculate the outlier factor of each sample data point; the mean value of the outlier factors of all sample data points in each cluster is denoted as the noise outlier factor of each cluster; all sample data points corresponding to the cluster with the largest noise outlier factor among all clusters are denoted as outlier sample data points; the outlier sample data points are screened out from all sample data points, and the remaining sample data points are denoted as target sample points; among them, the LOF algorithm is a well-known technology and will not be specifically elaborated here.

[0034] Interpolation is performed on all target sample points using the linear interpolation method to obtain the data points after interpolation; the data points after interpolation and all target sample points are combined to form target data points. Among them, the linear interpolation method is a well-known technology and will not be specifically elaborated here.

[0035] Dimensionality reduction is performed on the time-series data of all target data points corresponding to each dimension through the PCA (Principal Component Analysis) algorithm to obtain the features after dimensionality reduction, denoted as reference features. Among them, the PCA algorithm is a well-known technology and will not be specifically elaborated here.

[0036] Combining all the reference feature data and the frequency modulation command signal data at each moment to form a reference data point, several reference data points can be obtained.

[0037] So far, several reference data points have been obtained through the above method.

[0038] Step S003: Obtain the errors corresponding to different sample sizes of each decision tree through several reference data points; obtain the optimal sample size of each decision tree through the trend change of the errors corresponding to different sample sizes of each decision tree.

[0039] It should be noted that in order to reduce the number of data points of each decision tree and thus reduce the calculation amount, the optimal number of data points is obtained by iteratively analyzing the influence of the number of data points of each decision tree on the prediction error.

[0040] Specifically, sort several reference data points in chronological order to form a sequence, denoted as the data point sequence; form a sequence with the first reference data points in the data point sequence, denoted as the front data point sequence; form a sequence with the last reference data points in the data point sequence, denoted as the back data point sequence; form a sequence with all the frequency modulation command signal data corresponding to the reference data points in the back data point sequence, denoted as the back command signal data sequence. Among them, is a preset division percentage factor.

[0041] Among them, in this embodiment, the preset division percentage factor is In this embodiment, no specific limitation is imposed on the preset division percentage factor and the implementer can determine it according to the specific situation.

[0042] Randomly select data points from the front data point sequence with replacement to form a data set, and then select times to form data sets; construct decision trees from data sets, and through Predict the frequency modulation command signal data using a random forest constructed by decision trees, and form a sequence of the predicted frequency modulation command signal data in chronological order, denoted as the first predicted command signal data sequence; calculate the error between the first predicted command signal data sequence and the subsequent command signal data sequence through MAPE (Mean Absolute Percentage Error), denoted as the first error ; Take the first error as the error corresponding to the sample size of each decision tree being ; Randomly select a preset parameter data points from the previous data point sequence with replacement to form a data set, and then select times to form data sets; Construct decision trees from data sets, and predict the frequency modulation command signal data using a random forest constructed by decision trees. Form a sequence of the predicted frequency modulation command signal data in chronological order, denoted as the second predicted command signal data sequence; calculate the error between the second predicted command signal data sequence and the subsequent command signal data sequence through MAPE, denoted as the second error ; Take the second error as the error corresponding to the sample size of each decision tree being ; And so on, successively obtain the errors corresponding to different sample sizes of each decision tree; Among them, represents the preset number of decision trees; among them, in this embodiment, the preset number of decision trees , where in this embodiment, the preset number of decision trees is not specifically limited, and the implementer can determine it according to the specific situation. Among them, represents rounding up ; represents the number of all data points in the previous data point sequence.

[0043] It should be noted that when the sample size of each decision tree is small, there is a large deviation in its prediction result. Therefore, the prediction performance is often improved by increasing the sample size of each decision tree. However, during the process of increasing the sample size of each decision tree, its prediction performance gradually improves. When it reaches a certain level, increasing the number of decision trees may not improve the prediction effect or the improvement speed is slow. Therefore, analyze the change curve of the error during the process of increasing the sample size of each decision tree to obtain the most suitable sample size.

[0044] ​​​Specifically, a reference coordinate system is constructed with the sample size of each decision tree as the horizontal axis and the error corresponding to the sample size of each decision tree as the vertical axis; the different sample sizes and corresponding errors of each decision tree are mapped in the reference coordinate system to obtain a number of coordinate points; the least squares method is used to perform curve fitting on the number of coordinate points to obtain an error fitting curve. Among them, the least squares method is a well-known technique and will not be specifically described here.

[0045] Obtain the inflection point where the second derivative of the function corresponding to the error fitting curve is equal to 0, and use as the optimal sample size of each decision tree; among them, represents the horizontal value of the inflection point, and represents rounding up upward.

[0046] Thus, the optimal sample size of each decision tree is obtained through the above method.

[0047] Step S004: According to the optimal sample size of each decision tree, use the random forest algorithm to predict the command signal data.

[0048] According to the optimal sample size of each decision tree, use the random forest algorithm to predict the command signal data, so as to improve the prediction accuracy and reduce the calculation amount in the prediction process.

[0049] As Figure 2 shown, the second aspect of the present invention is to provide a system for optimizing the frequency modulation command prediction of a molten salt thermal energy storage system, including the following modules: Data acquisition module 101: used to continuously acquire the frequency modulation command signal data and various characteristic data affecting the stability of the power system at several moments; the various characteristic data include humidity, temperature, wind speed, and load; through the frequency modulation command signal data and various characteristic data at several moments, a number of sample data points are obtained; Data processing module 102: used to cluster all sample data points to obtain a number of clusters; screen out abnormal sample data points from all sample data points according to the dispersion degree of all data in the number of clusters to obtain target sample points; perform interpolation through all target sample points to obtain target data points; perform dimensionality reduction on the time series data corresponding to each dimension of all target data points to obtain the reference features after dimensionality reduction; through the frequency modulation command signal data and various reference feature data at several moments, a number of reference data points are obtained; Error analysis module 103: used to obtain the error corresponding to different sample sizes of each decision tree through a number of reference data points; obtain the optimal sample size of each decision tree through the trend change of the error corresponding to different sample sizes of each decision tree; Prediction module 104: It is used to predict the instruction signal data through the random forest algorithm according to the optimal sample size of each decision tree.

[0050] The third aspect of the present invention is to provide an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements a method for optimizing the prediction of frequency modulation instructions for a molten salt thermal energy storage system.

[0051] The fourth aspect of the present invention is to provide a computer-readable storage medium. The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, it implements a method for optimizing the prediction of frequency modulation instructions for a molten salt thermal energy storage system.

[0052] Those skilled in the art should understand that the embodiments of the present invention can be provided as a method, a system, or a computer program product. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, optical storage, etc.) containing computer-usable program code.

[0053] The present invention is described with reference to the flowcharts and / or block diagrams of methods, systems, and computer program products according to the embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as 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 the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices generate for implementing in the process Figure 1 one process or multiple processes and / or blocks Figure 1 a device for the functions specified in one block or multiple blocks.

[0054] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable memory generate a manufactured article including an instruction device, and the instruction device implements in the process Figure 1 one process or multiple processes and / or blocks Figure 1 a device for the functions specified in one block or multiple blocks.

[0055] These computer program instructions can also be loaded onto a computer or other programmable data processing apparatus, so that a series of operation steps are executed on the computer or other programmable apparatus to produce a computer-implemented process, thereby providing instructions for implementing the steps of the process Figure 1 one process or a plurality of processes and / or blocks Figure 1 steps for the functions specified in one block or a plurality of blocks.

[0056] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit them. Although the present invention has been described in detail with reference to the above embodiments, those of ordinary skill in the art should understand that: modifications or equivalent replacements can still be made to the specific embodiments of the present invention, and any modifications or equivalent replacements that do not depart from the spirit and scope of the present invention should be covered by the protection scope of the present invention.

Claims

1. A method for optimizing the prediction of frequency modulation commands in a molten salt thermal energy storage system, characterized in that, Including: Continuously obtain the frequency modulation command signal data at several moments and various characteristic data affecting the stability of the power system; Obtain a number of sample data points through the frequency modulation command signal data and various characteristic data at several moments; Cluster all sample data points to obtain a number of clusters; screen out abnormal sample data points from all sample data points according to the dispersion degree of all data in several clusters to obtain target sample points; perform interpolation through all target sample points to obtain target data points; perform dimensionality reduction on the time series data corresponding to each dimension of all target data points to obtain the reference features after dimensionality reduction; obtain a number of reference data points through the frequency modulation command signal data and various reference feature data at several moments; Obtain the errors corresponding to different sample sizes of each decision tree through a number of reference data points; obtain the optimal sample size of each decision tree through the trend change of the errors corresponding to different sample sizes of each decision tree; According to the optimal sample size of each decision tree, predict the command signal data through the random forest algorithm.

2. The method for predicting a frequency modulation command for optimizing a molten salt thermal energy storage system according to claim 1, wherein The obtaining of a number of sample data points through the frequency modulation command signal data and various characteristic data at several moments includes: Form a sample data point with the data of all dimensions at each moment, then a number of sample data points can be obtained; where all dimensions include all characteristic data and frequency modulation command signal data.

3. A method for predicting frequency modulation commands for optimizing a molten salt thermal energy storage system according to claim 1, characterized in that, The clustering of all sample data points to obtain a number of clusters includes: Cluster all sample data points using the K-means clustering algorithm to obtain a number of clusters; where the number of clusters is obtained by the elbow method.

4. A method for predicting frequency modulation commands for optimizing a molten salt thermal energy storage system according to claim 1, characterized in that, The screening out of abnormal sample data points from all sample data points according to the dispersion degree of all data in several clusters to obtain target sample points; performing interpolation through all target sample points to obtain target data points includes: Calculate and obtain the abnormal outlier factor of each sample data point through the LOF algorithm; record the mean value of the abnormal outlier factors of all sample data points in each cluster as the noise abnormal factor of each cluster; record all sample data points corresponding to the cluster with the largest noise abnormal factor among all clusters as abnormal sample data points; screen out abnormal sample data points from all sample data points, and record the remaining sample data points as target sample points; Perform interpolation on all target sample points through the linear interpolation method to obtain the data points after interpolation; form the data points after interpolation and all target sample points into target data points.

5. A method for predicting frequency modulation commands for optimizing a molten salt thermal energy storage system according to claim 1, wherein Perform dimensionality reduction on the time series data corresponding to each dimension of all target data points to obtain the reference features after dimensionality reduction; The obtaining of a number of reference data points through the frequency modulation command signal data and various reference feature data at several moments includes: Perform dimensionality reduction on the time series data corresponding to each dimension of all target data points through the PCA algorithm to obtain the features after dimensionality reduction, denoted as reference features; Form a reference data point with all reference feature data and frequency modulation command signal data at each moment, then a number of reference data points can be obtained.

6. The method for predicting the frequency modulation command for optimizing the molten salt thermal energy storage system according to claim 1, wherein Obtaining the errors corresponding to different sample sizes of each decision tree through a number of reference data points, including: Sort a number of reference data points in chronological order to form a sequence, denoted as the data point sequence; form a sequence from the first reference data points in the data point sequence, denoted as the front data point sequence; form a sequence from the last reference data points in the data point sequence, denoted as the back data point sequence; form a sequence from all the frequency modulation command signal data corresponding to the reference data points in the back data point sequence, denoted as the back command signal data sequence; where is a preset division percentage factor. Select preset parameters from the previous data point sequence with replacement data points to form a data set, and then select times to form data sets; construct data sets to build decision trees, and use the random forest constructed by decision trees to predict the frequency modulation command signal data. Arrange the predicted frequency modulation command signal data in chronological order to form a sequence, denoted as the first predicted command signal data sequence; calculate the error between the first predicted command signal data sequence and the subsequent command signal data sequence through MAPE, denoted as the first error ; take the first error as the error corresponding to the sample size of each decision tree being ; Select preset parameters from the previous data point sequence with replacement data points to form a data set, and then select times to form data sets; construct decision trees from data sets, and predict the frequency modulation command signal data through the random forest constructed by decision trees. The predicted frequency modulation command signal data is formed into a sequence in chronological order, denoted as the second predicted command signal data sequence; calculate the error between the second predicted command signal data sequence and the post-command signal data sequence through MAPE, denoted as the second error ; use the second error as the error corresponding to the sample size of each decision tree being ;​​​​ And so on, obtaining the errors corresponding to different sample sizes of each decision tree in sequence; Among them, represents the preset number of decision trees, represents rounding up ; represents the number of all data points in the previous data point sequence.

7. A method for predicting frequency modulation commands for optimizing a molten salt thermal energy storage system according to claim 1, characterized in that, Obtaining the optimal sample size of each decision tree through the trend change of the errors corresponding to different sample sizes of each decision tree, including: Taking the sample size of each decision tree as the horizontal axis and the error corresponding to the sample size of each decision tree as the vertical axis to construct a reference coordinate system; mapping the different sample sizes and corresponding errors of each decision tree in the reference coordinate system to obtain a number of coordinate points; performing curve fitting on the number of coordinate points by the least squares method to obtain an error fitting curve; Obtain the inflection point where the second derivative of the function corresponding to the error fitting curve is equal to 0, and take as the optimal sample size for each decision tree; where represents the horizontal value of the inflection point, represents rounding up.

8. A frequency modulation command prediction system for optimizing a molten salt thermal energy storage system, characterized in that, Including: Data acquisition module: used to continuously obtain the frequency modulation command signal data at a number of moments and various characteristic data affecting the stability of the power system; the various characteristic data include humidity, temperature, wind speed, and load; obtaining a number of sample data points through the frequency modulation command signal data and various characteristic data at a number of moments; Data processing module: used to cluster all sample data points to obtain a number of clusters; screening out abnormal sample data points from all sample data points according to the degree of dispersion of all data in the number of clusters to obtain target sample points; performing interpolation through all target sample points to obtain target data points; performing dimensionality reduction on the time series data corresponding to each dimension of all target data points to obtain the reference features after dimensionality reduction; obtaining a number of reference data points through the frequency modulation command signal data and various reference feature data at a number of moments; Error analysis module: used to obtain the errors corresponding to different sample sizes of each decision tree through a number of reference data points; obtaining the optimal sample size of each decision tree through the trend change of the errors corresponding to different sample sizes of each decision tree; Prediction module: used to predict the command signal data through the random forest algorithm according to the optimal sample size of each decision tree.

9. An electronic device, characterized in that, Including a memory, a processor, and a computer program stored in the memory and executable on the processor, and when the processor executes the computer program, it implements the method for optimizing the frequency modulation command prediction of the molten salt thermal energy storage system according to any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, and when the computer program is executed by the processor, it implements the method for optimizing the frequency modulation command prediction of the molten salt thermal energy storage system according to any one of claims 1-7.

Images