New energy load clustering method, device, equipment and medium

Abstract

The invention relates to a new energy load clustering method, device, equipment and medium, and the method comprises the steps: obtaining the sample point data of a new energy load of a to-be-clustered region, and determining the clustering center of each cluster and the membership matrix of each cluster; calculating a target function value of the current clustering center of the current cluster, and judging whether the change between the target function value of the current clustering center of the current cluster and the target function value of the last clustering center of the current cluster exceeds a threshold value or not; if the change exceeds the threshold value, updating the position of the current clustering center of the current cluster, and updating the membership matrix based on the updated clustering center until the change is smaller than or equal to the threshold value; and if the change is smaller than or equal to a threshold value, clustering the sample point data of the new energy load of the to-be-clustered region by taking the obtained current clustering center of the current cluster as a final clustering center. According to the method, the clustering precision can be remarkably improved, and the sensitivity to initial conditions is reduced, so that the method better adapts to the complexity of new energy loads.

Description

Technical Field

[0001] The present invention relates to the technical field of new energy load analysis, and in particular to a new energy load clustering method, device, equipment and medium. Background Art

[0002] Distributed power sources and energy storage devices, including photovoltaic, wind, and tidal power, are gaining increasing attention and application due to their advantages, such as low pollution emissions and high energy security. However, with the large-scale integration of new energy devices into the power grid, the randomness of new energy generation and load has created a dual uncertainty, adversely affecting the safe and stable operation of the power grid. This is especially true for urban production and living electricity loads, where loads such as electric vehicles often create a negative "peak-on-peak" impact. However, when faced with new energy loads, traditional grid dispatch methods often suffer from low dispatch efficiency or are unable to effectively control them due to large load fluctuations and strong unpredictability. Summary of the Invention

[0003] The technical problem to be solved by the present invention is to provide a new energy load clustering method, device, equipment and medium, which can significantly improve clustering accuracy, reduce sensitivity to initial conditions, and thus better adapt to the complexity of new energy loads.

[0004] The technical solution adopted by the present invention to solve the technical problem is to provide a new energy load clustering method, including the following steps:

[0005] Obtain the sample point data of the new energy load in the area to be clustered, and determine the cluster center of each cluster and its membership matrix;

[0006] For the cluster center of each cluster, the objective function value of the current cluster center of the current cluster is calculated according to the sample point data, the current cluster center of the current cluster and the membership matrix, and it is determined whether the change between the objective function value of the current cluster center of the current cluster and the objective function value of the previous cluster center of the current cluster exceeds the threshold;

[0007] If the change between the objective function value of the current cluster center of the current cluster and the objective function value of the previous cluster center of the current cluster exceeds a threshold, the current cluster center of the current cluster is updated, and its membership matrix is ​​updated based on the updated cluster center of the current cluster, until the change between the objective function value of the current cluster center of the current cluster and the objective function value of the previous cluster center of the current cluster is less than or equal to the threshold;

[0008] If the change between the objective function value of the current cluster center of the current cluster and the objective function value of the previous cluster center of the current cluster is less than or equal to the threshold, the obtained current cluster center is used as the final cluster center to cluster the sample point data of the new energy load in the clustering area.

[0009] The method of determining the cluster centers of each cluster and their membership matrices includes: randomly generating the cluster centers of multiple clusters through an initialization function, and initializing the membership matrices of the cluster centers according to the cluster centers, wherein the initialization function used when randomly generating the cluster centers of multiple clusters through the initialization function is: n =lb+rand×(ub-lb), where x n is the position of the initial cluster center, lb and ub are the lower and upper limits of the data range respectively, and rand is a random number between 0 and 1.

[0010] The objective function value of the cluster center is obtained by Calculated, where J(U,V) is the objective function value of the j-th cluster center, u ij is the membership value of the i-th row and j-th column in the membership matrix, d ij represents the distance from the i-th sample point to the j-th cluster center, k represents the number of clusters, and m represents the fuzzy index. This implementation uses the current cluster centers and the membership matrix to calculate the objective function value for clustering. This objective function value reflects the adaptability of the current clustering result. A lower objective function value indicates a smaller distance between the data point and its cluster center, indicating a better clustering effect. Therefore, by calculating the objective function value, the quality of the current cluster center can be evaluated.

[0011] When updating the current cluster center of the current cluster, use Update the current cluster center of the current cluster, where v j is the updated cluster center, u ij is the membership value of the i-th row and j-th column in the membership matrix, x n+1 is the updated location of the cluster center, x n is the location of the current cluster center of the current cluster, f(x n ) represents the location x of the current cluster center of the current cluster n The fitness value is Δx, and the perturbation amount is Δx. When the position of the current cluster center is updated, the search step size of the candidate solution of the cluster center is adjusted by the adaptive operator. The adaptive operator δ is expressed as: IT is the current iteration number, and Max_IT is the maximum iteration number.

[0012] When updating the membership matrix based on the updated cluster center, the Update the membership value of the i-th row and j-th column in the membership matrix, where u ij is the membership value of the i-th row and j-th column in the updated membership matrix, d ij Indicates the distance from the i-th sample point data to the j-th updated cluster center, dic It represents the distance from the i-th sample point data to the c-th updated cluster center, and m represents the fuzzy index.

[0013] Before clustering the sample point data of the new energy load in the clustering area using the current cluster center of the current cluster obtained as the final cluster center, the method further includes:

[0014] Use clustering evaluation criteria to evaluate whether the number of current cluster centers is appropriate;

[0015] If the number of current cluster centers obtained by the clustering evaluation criteria is inappropriate, the number of clusters is changed and the cluster centers of each cluster are searched again until the number of current cluster centers is appropriate.

[0016] The technical solution adopted by the present invention to solve the technical problem is to provide a new energy load clustering device, comprising:

[0017] The acquisition and determination module is used to obtain the sample point data of the new energy load in the area to be clustered, and to determine the cluster center of each cluster and its membership matrix;

[0018] The calculation and judgment module is used to calculate the objective function value of the current cluster center of each cluster based on the sample point data, the current cluster center of the current cluster and the membership matrix, and judge whether the change between the objective function value of the current cluster center of the current cluster and the objective function value of the previous cluster center of the current cluster exceeds a threshold;

[0019] An updating module is configured to update the position of the current cluster center of the current cluster when the change between the objective function value of the current cluster center of the current cluster and the objective function value of the previous cluster center of the current cluster exceeds a threshold, and update its membership matrix based on the updated cluster center of the current cluster until the change between the objective function value of the current cluster center of the current cluster and the objective function value of the previous cluster center of the current cluster is less than or equal to the threshold;

[0020] The clustering module is used to cluster the sample point data of the new energy load in the clustering area using the current cluster center of the current cluster as the final cluster center when the change between the objective function value of the current cluster center of the current cluster and the objective function value of the previous cluster center of the current cluster is less than or equal to a threshold.

[0021] The acquisition and determination module randomly generates the cluster centers of multiple clusters through an initialization function, and initializes its membership matrix according to each cluster center. When the cluster centers of multiple clusters are randomly generated through the initialization function, the initialization function used is: n =lb+rand×(ub-lb), where x nis the location of the cluster center, lb and ub are the lower and upper limits of the data range respectively, and rand is a random number between 0 and 1.

[0022] The calculation and judgment module is Calculate the objective function value of the current cluster center, where J(U,V) is the objective function value of the jth cluster center, u ij is the membership value of the i-th row and j-th column in the membership matrix, d ij It represents the distance from the i-th sample point data to the j-th cluster center, k represents the number of clusters, and m represents the fuzzy index.

[0023] The update module uses Update the current cluster center of the current cluster, where v j is the updated cluster center, u ij is the membership value of the i-th row and j-th column in the membership matrix, x n+1 is the updated location of the cluster center, x n is the location of the current cluster center of the current cluster, f(x n ) represents the location x of the current cluster center of the current cluster n The fitness value is Δx, and the perturbation amount is Δx. When the position of the current cluster center of the current cluster is updated, the search step size of the candidate cluster center solution is adjusted by the adaptive operator. The adaptive operator δ is expressed as: IT is the current iteration number, and Max_IT is the maximum iteration number.

[0024] The update module uses Update the membership value of the i-th row and j-th column in the membership matrix, where u ij is the membership value of the i-th row and j-th column in the updated membership matrix, d ij Indicates the distance from the i-th sample point data to the j-th updated cluster center, d ic It represents the distance from the i-th sample point data to the c-th updated cluster center, and m represents the fuzzy index.

[0025] The clustering module further includes:

[0026] An evaluation unit, configured to evaluate whether the number of current cluster centers is appropriate using a cluster evaluation standard;

[0027] The changing unit is used to change the number of clusters when the number of current cluster centers obtained by cluster evaluation criteria is not appropriate, and to re-search the cluster centers of each cluster until the number of current cluster centers obtained is appropriate.

[0028] The technical solution adopted by the present invention to solve its technical problem is: to provide an electronic device, including a memory, a processor and a computer program stored in the memory and capable of running on the processor, and the processor implements the steps of the above-mentioned new energy load clustering method when executing the computer program.

[0029] The technical solution adopted by the present invention to solve its technical problem is: providing a computer-readable storage medium on which a computer program is stored, and when the computer program is executed by a processor, the steps of the above-mentioned new energy load clustering method are implemented.

[0030] Beneficial effects

[0031] Due to the adoption of the above-mentioned technical solution, the present invention has the following advantages and positive effects compared with the prior art: the present invention randomly generates initial cluster centers through an initialization function, and combines the optimization mechanism of the initialization function to ensure the diversity and representativeness of the initial cluster centers. This improvement effectively reduces the sensitivity of the clustering results to the initial values, improves the clustering accuracy, and makes the clustering results more consistent with the actual data characteristics, thereby providing a more reliable basis for subsequent analysis. The present invention accurately evaluates the adaptability of the current clustering results by calculating the objective function value. This method not only evaluates the quality of the clustering, but also continuously optimizes the cluster centers based on the feedback of fitness by dynamically updating the individual positions. By optimizing the cluster centers and accurately updating the membership, the clustering results are more stable when facing complex load data in the real world, effectively avoiding clustering distortion caused by noise, and improving the credibility of the results. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 is a flow chart of a new energy load clustering method according to a first embodiment of the present invention;

[0033] Figure 2 It is a clustering index diagram of CH value and DBI value when the FCM clustering algorithm in the prior art is adopted;

[0034] Figure 3 This is a clustering result diagram when the FCM clustering algorithm in the prior art is used;

[0035] Figure 4 1 is a clustering index diagram of CH value and DBI value when the first embodiment of the present invention is adopted;

[0036] Figure 5 This is a diagram of clustering results when the first embodiment of the present invention is adopted. DETAILED DESCRIPTION

[0037] Below in conjunction with specific embodiment, further set forth the present invention.Should be understood that these embodiments are only used to illustrate the present invention and are not used in limiting the scope of the present invention.In addition, should be understood that after reading the content taught by the present invention, those skilled in the art can make various changes or modifications to the present invention, and these equivalent forms fall equally within the scope limited by the appended claims of the application.

[0038] The first embodiment of the present invention relates to a new energy load clustering method, which aims to improve the clustering accuracy and efficiency of new energy loads so as to provide more reliable data support for power grid dispatching. Cluster analysis, as an effective data processing tool, has gradually been widely used in the management of new energy loads. In recent years, technologies such as fuzzy C-means clustering (FCM) have been introduced to analyze and optimize new energy loads and improve the intelligence level of power grid dispatching. FCM is a clustering method widely used in data analysis, especially for processing uncertainty and fuzzy data. However, when processing high-dimensional data or data containing noise, traditional FCM is prone to fall into local optimal solutions, resulting in inaccurate clustering results. In addition, its sensitivity to the initial center also limits its application in complex load environments.

[0039] like Figure 1 As shown, this embodiment improves the traditional FCM and specifically includes the following steps:

[0040] Step 1: Obtain the sample point data of the new energy load in the area to be clustered, and determine the cluster center of each cluster and its membership matrix.

[0041] In this step, the cluster centers of multiple clusters are randomly generated by the initialization function, and the membership matrix of each cluster center is initialized according to the initialization function: n =lb+rand×(ub-lb), where x n is the location of the initial cluster center, lb and ub are the lower and upper limits of the data range, respectively, and rand is a random number between 0 and 1. This implementation ensures the diversity and representativeness of the initial cluster centers through the initialization function mechanism, effectively reducing the sensitivity of the clustering results to the initial values. At the same time, by setting the lower and upper limits of the data range, it ensures that the generated cluster centers are within the data range, avoiding invalid cluster center locations.

[0042] Step 2: For the cluster center of each cluster, calculate the objective function value of the current cluster center of the current cluster based on the sample point data, the current cluster center of the current cluster and the membership matrix, and judge whether the change between the objective function value of the current cluster center of the current cluster and the objective function value of the previous cluster center of the current cluster exceeds the threshold.

[0043] In this step, we use Calculate the objective function value of the cluster center, where J(U,V) is the objective function value of the jth cluster center, u ij is the membership value of the i-th row and j-th column in the membership matrix, d ij It represents the distance from the i-th sample point data to the j-th cluster center, k represents the number of clusters, and m represents the fuzzy index, where m>1.

[0044] This implementation uses the current cluster centers and the membership matrix to calculate the clustering objective function value. This objective function value reflects the adaptability of the current clustering result. A lower objective function value means that the distance between the data point and its cluster center is smaller, indicating a better clustering effect. Therefore, by calculating the objective function value, the quality of the current cluster center can be evaluated.

[0045] Step 3: If the change between the objective function value of the current cluster center of the current cluster and the objective function value of the previous cluster center of the current cluster exceeds the threshold, the position of the current cluster center of the current cluster is updated, and its membership matrix is ​​updated based on the updated cluster center of the current cluster, until the change between the objective function value of the current cluster center of the current cluster and the objective function value of the previous cluster center of the current cluster is less than or equal to the threshold.

[0046] In this step, when updating the current cluster center position of the current cluster, the Update the current cluster center of the current cluster, where v j is the updated cluster center, u ij is the membership value of the i-th row and j-th column in the membership matrix, x n+1 is the updated location of the cluster center, x n is the location of the current cluster center of the current cluster, f(x n ) represents the location x of the current cluster center of the current cluster n The fitness value is Δx, and the perturbation amount is Δx. When the position of the current cluster center is updated, the search step size of the candidate solution of the cluster center is adjusted by the adaptive operator. The adaptive operator δ is expressed as: IT is the current iteration number, and Max_IT is the maximum iteration number. This implementation uses the adaptive operator δ to adjust the search step size of the cluster center candidate solution, determining the movement direction and amplitude of the cluster center in each iteration. After multiple iterations, the final cluster center position is the result of the search process dynamically adjusted by the adaptive operator δ.

[0047] In this step, when updating the membership matrix, we use Update the membership value of the i-th row and j-th column in the membership matrix, where u ij is the membership value of the i-th row and j-th column in the updated membership matrix, d ijIndicates the distance from the i-th sample point data to the j-th updated cluster center, d ic It represents the distance from the i-th sample point data to the c-th updated cluster center, and m represents the fuzzy index.

[0048] When updating cluster centers and membership degrees, this implementation provides feedback based on the fitness of each cluster center, pushing the cluster center toward a more optimal position. This adjustment mechanism effectively improves the algorithm's convergence speed and clustering accuracy. By optimizing cluster centers and accurately updating membership degrees, the clustering results are more stable when dealing with complex real-world load data, effectively avoiding cluster distortion caused by noise and improving the credibility of the results. Furthermore, during the update process, an adaptive operator is employed to maintain a balance between the exploration and development phases of obtaining the optimal position. This adaptive operator automatically adjusts based on the number of iterations. This operator increases diversity through random actions during the optimization process, avoiding local optimality.

[0049] Step 4: If the change between the objective function value of the current cluster center of the current cluster and the objective function value of the previous cluster center of the current cluster is less than or equal to the threshold, the current cluster center of the current cluster is used as the final cluster center to cluster the sample point data of the new energy load in the clustering area.

[0050] This embodiment calculates the current objective function value in each iteration and compares it with the previous objective function value. By observing the change in the objective function value, it can be determined whether the clustering result is stable. When the change in the objective function value is less than the set threshold, it indicates that the clustering result may have converged, and the current clustering result can be output at this time. By continuously repeating steps 2 and 3, this embodiment can accumulate feedback information in each iteration and optimize the calculation of cluster centers and membership matrices. This feedback mechanism maintains flexibility throughout the optimization process and ultimately ensures that the clustering result is optimal.

[0051] The current cluster center obtained by this implementation method can better reflect the characteristics of load data, thereby improving the intelligence level of power grid dispatching and improving the operating efficiency and safety of the power grid. This lays a solid foundation for subsequent resource allocation and dispatching decisions and has important practical application value.

[0052] It is worth mentioning that, before clustering the sample point data of the new energy load in the clustering area using the current cluster center of the current cluster obtained in step 4 as the final cluster center, the following steps may also be included:

[0053] A clustering evaluation standard is used to evaluate whether the number of current cluster centers is appropriate. The clustering evaluation standard of this embodiment may adopt the Davidson-Boulding Index (DBI) value and the Calinski-Harabaz (CH) value.

[0054] If the number of current cluster centers obtained by the clustering evaluation standard is appropriate, the current cluster center of the current cluster is used as the final cluster center to cluster the sample point data of the new energy load in the clustering area. If the number of current cluster centers obtained by the clustering evaluation standard is inappropriate, the number of clusters is changed and the cluster centers of each cluster are searched again until the number of current cluster centers is appropriate.

[0055] like Figure 2 and Figure 3 As shown in the figure, the DBI value and CH value are used to evaluate the number of cluster centers obtained by the FCM clustering method and the method of this embodiment respectively, so as to judge whether the number of cluster centers obtained is appropriate. According to the CH value and DBI value of the FCM clustering effectiveness index, the best effect is achieved when the number of clusters is 2, but at this time, the load curves of 3kw and 7kw power cannot be effectively distinguished. Figure 4 and Figure 5 As shown, according to the effectiveness indicators CH value and DBI value of the clustering method proposed in this embodiment, the best effect is achieved when the number of clusters is 5, and the load curves of 3 kW and 7 kW can be effectively distinguished at this time.

[0056] It is not difficult to find that the present invention randomly generates initial cluster centers through an initialization function. This initialization function can randomly generate an initial cluster center for each cluster based on the set population size and search space range. This random value ensures the diversity and representativeness of the initial cluster centers. This improvement effectively reduces the sensitivity of the clustering results to the initial value, improves the clustering accuracy, and makes the clustering results more consistent with the actual data characteristics, thereby providing a more reliable foundation for subsequent analysis. By optimizing the cluster centers and accurately updating the membership, the present invention makes the clustering results more stable when facing complex load data in the real world, effectively avoiding clustering distortion caused by noise, and improving the credibility of the results.

[0057] A second embodiment of the present invention relates to a new energy load clustering device, comprising:

[0058] The acquisition and determination module is used to obtain the sample point data of the new energy load in the area to be clustered, and to determine the cluster center of each cluster and its membership matrix;

[0059] The calculation and judgment module is used to calculate the objective function value of the current cluster center of each cluster based on the sample point data, the current cluster center of the current cluster and the membership matrix, and judge whether the change between the objective function value of the current cluster center of the current cluster and the objective function value of the previous cluster center of the current cluster exceeds a threshold;

[0060] an updating module, configured to update the position of the current cluster center of the current cluster when a change between the objective function value of the current cluster center of the current cluster and the objective function value of the previous cluster center of the current cluster exceeds a threshold, and update its membership matrix based on the updated cluster center until the change between the objective function value of the current cluster center of the current cluster and the objective function value of the previous cluster center of the current cluster is less than or equal to the threshold;

[0061] The clustering module is used to cluster the sample point data of the new energy load in the clustering area using the current cluster center of the current cluster as the final cluster center when the change between the objective function value of the current cluster center of the current cluster and the objective function value of the previous cluster center of the current cluster does not exceed a threshold.

[0062] The acquisition and determination module randomly generates the cluster centers of multiple clusters through an initialization function, and initializes its membership matrix according to each cluster center. When the cluster centers of multiple clusters are randomly generated through the initialization function, the initialization function used is: n =lb+rand×(ub-lb), where x n is the location of the cluster center, lb and ub are the lower and upper limits of the data range respectively, and rand is a random number between 0 and 1.

[0063] The calculation and judgment module is Calculate the objective function value of the current cluster center, where J(U,V) is the objective function value of the jth cluster center, u ij is the membership value of the i-th row and j-th column in the membership matrix, d ij It represents the distance from the i-th sample point data to the j-th cluster center, k represents the number of clusters, and m represents the fuzzy index.

[0064] The update module uses Update the current cluster center position of the current cluster, where v j is the updated cluster center, u ij is the membership value of the i-th row and j-th column in the membership matrix, x n+1 is the updated location of the cluster center, x n is the location of the current cluster center of the current cluster, f(x n) represents the location x of the current cluster center of the current cluster n The fitness value is Δx, and the perturbation amount is Δx. When the position of the current cluster center of the current cluster is updated, the search step size of the candidate cluster center solution is adjusted by the adaptive operator. The adaptive operator δ is expressed as: IT is the current iteration number, and Max_IT is the maximum iteration number.

[0065] The update module uses Update the membership value of the i-th row and j-th column in the membership matrix, where u ij is the membership value of the i-th row and j-th column in the updated membership matrix, d ij Indicates the distance from the i-th sample point data to the j-th updated cluster center, d ic It represents the distance from the i-th sample point data to the c-th updated cluster center, and m represents the fuzzy index.

[0066] The clustering module further includes:

[0067] An evaluation unit, configured to evaluate whether the number of current cluster centers is appropriate using a cluster evaluation standard;

[0068] The changing unit is used to change the number of clusters when the number of current cluster centers obtained by cluster evaluation criteria is not appropriate, and to re-search the cluster centers of each cluster until the number of current cluster centers obtained is appropriate.

[0069] A third embodiment of the present invention relates to an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of the new energy load clustering method of the first embodiment are implemented.

[0070] A fourth embodiment of the present invention relates to a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the steps of the new energy load clustering method of the first embodiment are implemented.

[0071] Those skilled in the art will appreciate that the embodiments of the present application can be provided as methods, systems, or computer program products. Therefore, the present application can adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage and optical storage, etc.) that contain computer-usable program code.

[0072] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the steps in the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0073] These computer program instructions may 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 produce an article of manufacture including an instruction method, which is implemented in the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0074] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0075] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any modifications or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be based on the scope of protection of the claims.

Claims

1. A new energy load clustering method, characterized in that: The following steps are involved: Obtain the sample point data of the new energy load in the area to be clustered, and determine the cluster center of each cluster and its membership matrix; for the cluster center of each cluster, calculate the objective function value of the current cluster center of the current cluster based on the sample point data, the current cluster center of the current cluster and the membership matrix, and determine whether the change between the objective function value of the current cluster center of the current cluster and the objective function value of the previous cluster center of the current cluster exceeds a threshold; If the change between the objective function value of the current cluster center of the current cluster and the objective function value of the previous cluster center of the current cluster exceeds a threshold, the current cluster center of the current cluster is updated, and its membership matrix is ​​updated based on the updated cluster center of the current cluster, until the change between the objective function value of the current cluster center of the current cluster and the objective function value of the previous cluster center of the current cluster is less than or equal to the threshold; If the change between the objective function value of the current cluster center of the current cluster and the objective function value of the previous cluster center of the current cluster is less than or equal to the threshold, the obtained current cluster center is used as the final cluster center to cluster the sample point data of the new energy load in the clustering area.

2. The new energy load clustering method according to claim 1, characterized in that: The method of determining the cluster centers of each cluster and their membership matrices includes: randomly generating the cluster centers of multiple clusters through an initialization function, and initializing the membership matrices of the cluster centers according to the cluster centers, wherein the initialization function used when randomly generating the cluster centers of multiple clusters through the initialization function is: n =lb+rand×(ub-lb), where x n is the location of the cluster center, lb and ub are the lower and upper limits of the data range respectively, and rand is a random number between 0 and 1.

3. The new energy load clustering method according to claim 1, characterized in that: The objective function value of the cluster center is obtained by Calculated, where J(U,V) is the objective function value of the j-th cluster center, u ij is the membership value of the i-th row and j-th column in the membership matrix, d ij It represents the distance from the i-th sample point data to the j-th cluster center, k represents the number of clusters, and m represents the fuzzy index.

4. The new energy load clustering method according to claim 1, characterized in that: When updating the current cluster center of the current cluster, use Update the current cluster center of the current cluster, where v j is the updated cluster center, u ij is the membership value of the i-th row and j-th column in the membership matrix, x n+1 is the updated location of the cluster center, x n is the location of the current cluster center of the current cluster, f(x n ) represents the location x of the current cluster center of the current cluster n The fitness value is Δx, and the perturbation amount is Δx. When the position of the current cluster center is updated, the search step size of the candidate solution of the cluster center is adjusted by the adaptive operator. The adaptive operator δ is expressed as: IT is the current iteration number, and Max_IT is the maximum iteration number.

5. The new energy load clustering method according to claim 1, characterized in that: When updating the membership matrix based on the updated cluster center, the Update the membership value of the i-th row and j-th column in the membership matrix, where u ij is the membership value of the i-th row and j-th column in the updated membership matrix, d ij Indicates the distance from the i-th sample point data to the j-th updated cluster center, d ic It represents the distance from the i-th sample point data to the updated c-th cluster center, and m represents the fuzzy index.

6. The new energy load clustering method according to claim 1, characterized in that: Before clustering the sample point data of the new energy load in the clustering area using the current cluster center of the current cluster obtained as the final cluster center, the method further includes: Use clustering evaluation criteria to evaluate whether the number of current cluster centers is appropriate; If the number of current cluster centers obtained by the clustering evaluation criteria is inappropriate, the number of clusters is changed and the cluster centers of each cluster are searched again until the number of current cluster centers is appropriate.

7. A new energy load clustering device, characterized in that: include: The acquisition and determination module is used to obtain the sample point data of the new energy load in the area to be clustered, and to determine the cluster center of each cluster and its membership matrix; The calculation and judgment module is used to calculate the objective function value of the current cluster center of each cluster based on the sample point data, the current cluster center of the current cluster and the membership matrix, and judge whether the change between the objective function value of the current cluster center of the current cluster and the objective function value of the previous cluster center of the current cluster exceeds a threshold; an updating module, configured to update the current cluster center of the current cluster when a change between the objective function value of the current cluster center of the current cluster and the objective function value of the previous cluster center of the current cluster exceeds a threshold, and update its membership matrix based on the updated cluster center of the current cluster until a change between the objective function value of the current cluster center of the current cluster and the objective function value of the previous cluster center of the current cluster is less than or equal to the threshold; The clustering module is used to cluster the sample point data of the new energy load in the clustering area using the current cluster center of the current cluster as the final cluster center when the change between the objective function value of the current cluster center of the current cluster and the objective function value of the previous cluster center of the current cluster is less than or equal to a threshold.

8. The new energy load clustering device according to claim 7, characterized in that: The acquisition and determination module randomly generates the cluster centers of multiple clusters through an initialization function, and initializes its membership matrix according to each cluster center. When the cluster centers of multiple clusters are randomly generated through the initialization function, the initialization function used is: n =lb+rand×(ub-lb), where x n is the location of the cluster center, lb and ub are the lower and upper limits of the data range respectively, and rand is a random number between 0 and 1.

9. The new energy load clustering device according to claim 7, characterized in that: The calculation and judgment module is Calculate the objective function value of the cluster center, where J(U,V) is the objective function value of the jth cluster center, u ij is the membership value of the i-th row and j-th column in the membership matrix, d ij It represents the distance from the i-th sample point data to the j-th cluster center, k represents the number of clusters, and m represents the fuzzy index.

10. The new energy load clustering device according to claim 7, characterized in that: The update module uses Update the current cluster center of the current cluster, where v j is the updated cluster center, u ij is the membership value of the i-th row and j-th column in the membership matrix, x n+1 is the updated location of the cluster center, x n is the location of the current cluster center of the current cluster, f(x n ) represents the location x of the current cluster center of the current cluster n The fitness value is Δx, and the perturbation amount is Δx. When the position of the current cluster center of the current cluster is updated, the search step size of the candidate cluster center solution is adjusted by the adaptive operator. The adaptive operator δ is expressed as: IT is the current iteration number, and Max_IT is the maximum iteration number.

11. The new energy load clustering device according to claim 7, characterized in that: The update module uses Update the membership value of the i-th row and j-th column in the membership matrix, where u ij is the membership value of the i-th row and j-th column in the updated membership matrix, d ij Indicates the distance from the i-th sample point data to the j-th updated cluster center, d ic It represents the distance from the i-th sample point data to the c-th updated cluster center, and m represents the fuzzy index.

12. The new energy load clustering device according to claim 7, characterized in that: The clustering module further includes: An evaluation unit, configured to evaluate whether the number of current cluster centers is appropriate using a cluster evaluation standard; The changing unit is used to change the number of clusters when the number of current cluster centers obtained by cluster evaluation criteria is not appropriate, and to re-search the cluster centers of each cluster until the number of current cluster centers obtained is appropriate.

13. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the steps of the new energy load clustering method according to any one of claims 1 to 6 are implemented.

14. 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 new energy load clustering method according to any one of claims 1 to 6 are implemented.

Images