Intelligent photovoltaic power quality control system

Abstract

The invention discloses an intelligent photovoltaic power quality control system which comprises a photovoltaic data acquisition module, a feature space trend chart construction module, an isolation correction module, a preliminary clustering module, a boundary disturbance distribution module, a clustering optimization module and a photovoltaic power quality control module. The invention belongs to the field of electric energy quality control, and particularly relates to an intelligent photovoltaic electric energy quality control system. According to the scheme, hierarchical inflection points from tiny fluctuation to violent interference are captured in a self-adaptive mode, and events of different intensities are recognized more flexibly; trend deviation caused by abnormal disturbance is eliminated; a density consistency constraint is introduced to calculate a merging index; the photovoltaic power quality control accuracy is improved; the blind area probability is reduced through initialization, and the characteristics of photovoltaic data are better adapted; a dynamic balance mechanism is introduced to balance global search and local convergence; long-term retention of inferior solutions is avoided based on a local search enhancement mechanism; and finally, the photovoltaic power quality control effect is improved.

Description

Technical Field

[0001] The present invention relates to the field of power quality control, and specifically refers to an intelligent photovoltaic power quality control system. Background Art

[0002] A photovoltaic power quality control system is a comprehensive system used to monitor, analyze, and improve the power quality generated by a photovoltaic power generation system. However, general photovoltaic power quality control systems are prone to missing weak interference or misreporting strong interference, being severely affected by abnormal points, and unable to adapt to the density differences of photovoltaic systems under multiple working conditions, resulting in low accuracy of photovoltaic power quality control; general photovoltaic power quality control systems are unable to adapt to complex operating environments and are difficult to balance global exploration and local development simultaneously, leading to poor photovoltaic power quality control effects. Summary of the Invention

[0003] In view of the above situation, to overcome the defects of the prior art, the present invention provides an intelligent photovoltaic power quality control system. Regarding the problem that general photovoltaic power quality control systems are prone to missing weak interference or misreporting strong interference, being severely affected by abnormal points, and unable to adapt to the density differences of photovoltaic systems under multiple working conditions, resulting in low accuracy of photovoltaic power quality control, this solution adaptively captures the hierarchical inflection points from minute fluctuations to intense interference, and more flexibly identifies events of different intensities; captures transient harmonic clusters through LOF and captures mutation points through isolation forest to eliminate trend offsets caused by abnormal disturbances; introduces density consistency constraints to calculate the merging index to prevent mismerging between clusters of different densities and adapt to various working conditions of photovoltaic operation; thereby improving the accuracy of photovoltaic power quality control. Regarding the problem that general photovoltaic power quality control systems are unable to adapt to complex operating environments and are difficult to balance global exploration and local development simultaneously, leading to poor photovoltaic power quality control effects, this solution reduces the blind area probability through initialization, enhances initial diversity, and better adapts to the characteristics of wide fluctuation ranges and large distribution differences in photovoltaic data; introduces a dynamic balance mechanism to balance global search and local convergence and adapt to the non-stationarity of photovoltaic data changing over time; avoids the long-term retention of inferior solutions based on the local search enhancement mechanism; ultimately improves the photovoltaic power quality control effect.

[0004] The technical solution adopted by the present invention is as follows: An intelligent photovoltaic power quality control system provided by the present invention includes a photovoltaic data acquisition module, a feature space trend graph construction module, an isolation correction module, a preliminary clustering module, a boundary perturbation allocation module, a clustering optimization module, and a photovoltaic power quality control module;

[0005] The photovoltaic data acquisition module acquires the power quality data of the photovoltaic system to form a sample set;

[0006] The feature space trend graph construction module constructs a fully connected graph among samples, and constructs a core set by calculating the elbow distance and adjacency relationship;

[0007] The isolated point correction module detects isolated points and improves the core set through anomaly scores;

[0008] The preliminary clustering module forms preliminary clusters based on the core points and their neighborhood densities, and introduces a density consistency factor for cluster merging;

[0009] The boundary perturbation allocation module calculates the allocation confidence for boundary points based on local similarity, and finally performs boundary point allocation;

[0010] The clustering optimization module automatically adjusts clustering parameters based on the particle swarm optimization algorithm, and introduces a dynamic balance mechanism and a local search enhancement strategy;

[0011] The photovoltaic power quality control module realizes photovoltaic power quality control for real-time power quality monitoring data based on the clustering results of historical power quality monitoring data.

[0012] Further, the photovoltaic data acquisition module collects historical power quality monitoring data according to a fixed sampling time sequence during the operation of the photovoltaic system, preprocesses the collected data, and forms a sample set , where and are the first and nth samples respectively; each sample is an extracted feature vector; each sample is labeled with an operating state, and the operating state is used as a data label, only for use as a cluster label selection; the operating states include normal operation, mild anomaly, moderate anomaly, and severe anomaly.

[0013] Further, the feature space trend graph construction module constructs a fully connected graph among samples, with the edge weight being the Euclidean distance, and initializes all samples as unvisited; for each sample u, calculate the elbow distance between the sample and its k nearest neighbors , connect the neighbors with the Euclidean distance less than ; take sample u as the leaf node and sample v as the parent node, and jump to the nearest neighbor of itself each time, and take the stop point as the root node; repeat until all points are visited, and collect all root node sets as the core set R; where is the Euclidean distance from u to v; the k nearest neighbors are the set of k samples with the smallest Euclidean distances; the elbow distance is the distance corresponding to the point with the largest curvature on the curve obtained by plotting the Euclidean distances from sample u to all other samples in its k nearest neighbors and sorting them by size; v is a sample in the k nearest neighbors of u.

[0014] Furthermore, the isolated correction module uses the anomaly score to fuse the Local Outlier Factor (LOF) to detect outliers and calculates the anomaly score for each sample. , and the formula used is: ; ; Calculate the global anomaly score , and the formula used is: ; where is the path length of the isolated path; E[·] is the expectation; is the i-th sample the expected path length required to be isolated; is the sample normalization constant; is the harmonic number; is the anomaly weight; is the local outlier factor of the sample; is the maximum value of the local outlier factors among all samples; is the j-th sample; N is the size of the sample subset of each tree when the samples participate in constructing the isolation forest; Samples with high global anomaly scores are marked as the sparse region SR according to the threshold, and the rest are the dense region DR; For each trend tree edge (x a , x b ), if one end x a belongs to SR and the other end x b belongs to DR, then disconnect this trend tree edge and regard both ends as new root nodes and add them to R; x a and x b are the two ends of the trend tree edge.

[0015] Furthermore, the preliminary clustering module takes the i-th sample as the core point, and for the core point , set the k-nearest neighbor set of the core point as NN(x i ); For each core point , connect edges with the points belonging to NN(x i ) to form a subgraph ; Each subgraph corresponds to a cluster; Introduce the density consistency factor , and calculate the merging index. The formula used is: ; ; ; ; where is the merging index; i and j are both sample indices, and i ≠ j; is the subgraph obtained with the core point ; and are the local densities of samples x and y respectively; is the subgraph and the subgraph The nearest Euclidean distance between; is the density normalization parameter; is the set of k-nearest neighbors of the sample; Iteratively merge the largest pair until the number of remaining clusters is K to obtain the final core clustering; is the subgraph of the sample; is the subgraph of the sample; e is the natural constant.

[0016] Furthermore, the boundary perturbation allocation module treats samples that are not assigned to any cluster as boundary points and calculates its allocation confidence to the c-th cluster using the formula: ; where is the boundary point and is the total Euclidean distance between the points of; Assign the boundary point to the cluster with the maximum allocation confidence; For the clustering result of historical power quality monitoring data after boundary perturbation allocation, use the silhouette coefficient to evaluate, set the silhouette coefficient threshold, and if the average silhouette coefficient of all samples is lower than the silhouette coefficient threshold, perform clustering optimization; Otherwise, the clustering of historical power quality monitoring data is completed.

[0017] Furthermore, the clustering optimization module specifically includes the following:

[0018] Initialize the individual position; Construct a parameter optimization space based on the target number of clusters, density normalization parameter, k, and silhouette coefficient threshold; The quasi-initialization of the q-th dimension of each particle is expressed as: ; Generate the individual position to obtain the final initialized particle using the formula: ; Use the average silhouette coefficient of all samples clustered based on the individual position as the individual fitness value; The lower the individual fitness value, the worse the solution corresponding to the individual; where is the quasi-initial position of the q-th dimension of the m-th particle; and are the upper and lower limits of the q-th dimension respectively; is a random number obeying the uniform distribution ; is the final initial position of the q-th dimension of the m-th particle; rnd(·,·) is a function that uniformly and randomly takes values within a given interval;

[0019] Particle position update; Introduce a position dynamic balance mechanism and define a time function using the formula: ; ; Update the velocity and position using the formula: ; ; where is the dynamic cognitive learning factor; g is the current iteration number; is the maximum number of iterations; is the inertia weight; is the search mode switching threshold; and are the q-th dimension velocity of the m-th particle at the g+1-th iteration and the g-th iteration respectively; , and They are all random numbers between 0 and 1, independent of each other; and are the positions of the qth dimension of the mth particle at the g+1th iteration and the gth iteration respectively; It is a social learning factor; is the position of the optimal particle in the qth dimension; F is a random number that obeys the uniform distribution U(0,1), and Independent of each other; is the historical optimal position of the mth particle in the qth dimension;

[0020] Local search enhancement: every k iterations, the 10% individuals with the lowest fitness values ​​are perturbed and adjusted. The formula used is: ;in, is the position after local search enhancement; The degrees of freedom are t-distributed random variable;

[0021] Optimization judgment; when there is an individual fitness value higher than the silhouette coefficient threshold, the clustering result of the historical power quality monitoring data is obtained based on the individual position; if the maximum number of iterations is reached, return to the initialized individual position; otherwise, return to the particle position update.

[0022] Furthermore, the photovoltaic power quality control module selects the label with the largest number in the cluster as the cluster label based on the clustering result of historical power quality monitoring data; collects real-time power quality monitoring data and performs clustering allocation based on the minimum Euclidean distance principle; uses the cluster label corresponding to the real-time power quality monitoring data as the power quality monitoring result; if the monitoring result is moderately abnormal, increases the sampling frequency of the monitoring data; if the monitoring result is seriously abnormal, performs early warning processing; thereby realizing power quality control.

[0023] The beneficial effects achieved by the present invention using the above scheme are as follows:

[0024] (1) Aiming at the problems existing in general photovoltaic power quality control systems, such as being prone to missing weak interferences or misreporting strong interferences, being severely affected by abnormal points, and being unable to adapt to the density differences of photovoltaic systems under multiple working conditions, which further lead to low accuracy of photovoltaic power quality control, this solution adaptively captures the hierarchical inflection points from tiny fluctuations to severe interferences, and more flexibly identifies events of different intensities; captures transient harmonic clusters through LOF and captures mutation points through isolation forest to eliminate the trend offset caused by abnormal disturbances; introduces density consistency constraints to calculate the merging index to prevent mismerging between clusters of different densities and adapt to various working conditions of photovoltaic operation; thereby improving the accuracy of photovoltaic power quality control.

[0025] (2) Aiming at the problems existing in general photovoltaic power quality control systems, such as being unable to adapt to complex operating environments, being difficult to balance global exploration and local development simultaneously, and resulting in poor photovoltaic power quality control effects, this solution reduces the blind area probability through initialization, enhances initial diversity, and better adapts to the characteristics of wide fluctuation ranges and large distribution differences in photovoltaic data; introduces a dynamic balance mechanism to balance global search and local convergence and adapt to the non-stationarity of photovoltaic data changing over time; avoids the long-term retention of inferior solutions based on the local search enhancement mechanism; ultimately improving the photovoltaic power quality control effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 FIG. is a schematic flow chart of an intelligent photovoltaic power quality control system provided by the present invention;

[0027] Figure 2 FIG. is a schematic flow chart of the clustering optimization module.

[0028] The drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation to the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0029] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments; based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0030] In the description of the present invention, it should be understood that the terms "upper", "lower", "front", "rear", "left", "right", "top", "bottom", "inner", "outer", etc. indicate the orientation or position relationship based on the orientation or position relationship shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation to the present invention.

[0031] Example 1. Refer to Figure 1 , an intelligent photovoltaic power quality control system provided by the present invention includes a photovoltaic data acquisition module, a feature space trend graph construction module, an isolation correction module, a preliminary clustering module, a boundary perturbation allocation module, a clustering optimization module, and a photovoltaic power quality control module;

[0032] The photovoltaic data acquisition module acquires the power quality data of the photovoltaic system to form a sample set; and sends the data to the feature space trend graph construction module;

[0033] The feature space trend graph construction module constructs a fully connected graph between samples, and constructs a core set by calculating the elbow point distance and adjacency relationship; and sends the data to the isolation correction module;

[0034] The isolation correction module detects the isolated points and improves the core set through the anomaly score; and sends the data to the preliminary clustering module;

[0035] The preliminary clustering module forms preliminary clusters based on the core points and their neighborhood densities, and introduces a density consistency factor for cluster merging; and sends the data to the boundary perturbation allocation module;

[0036] The boundary perturbation allocation module calculates the allocation confidence for the boundary points that are not assigned to any cluster based on the local similarity, and finally performs the boundary point allocation; and sends the data to the clustering optimization module;

[0037] The clustering optimization module automatically adjusts the clustering parameters based on the particle swarm optimization algorithm, and introduces a dynamic balance mechanism and a local search enhancement strategy; and sends the data to the photovoltaic power quality control module;

[0038] The photovoltaic power quality control module realizes the photovoltaic power quality control for the real-time power quality monitoring data based on the clustering result of the historical power quality monitoring data.

[0039] Example 2. Refer to Figure 1 , based on the above example, the photovoltaic data acquisition module acquires the historical power quality monitoring data at fixed sampling time sequences during the operation of the photovoltaic system, preprocesses the acquired data to form a sample set , where and are the 1st and nth samples respectively; each sample is an extracted feature vector; the operation state of each sample is labeled, and the operation state is used as the data label, only for use as the cluster label selection; the feature vector includes voltage feature, frequency feature, harmonic feature, power feature, and current feature; the operation state includes normal operation, mild anomaly, moderate anomaly, and severe anomaly.

[0040] Example 3. Refer to Figure 1 . Based on the above example, the feature space trend graph construction module constructs a fully connected graph between samples, with the edge weight being the Euclidean distance, and initializes all samples as unvisited; for each sample u, calculate the elbow distance between the sample and its k nearest neighbors , and connect the neighbors with Euclidean distance less than ; take sample u as the leaf node and sample v as the parent node, and each time jump to the neighbor closest to itself, and take the stop point as the root node; repeat until all points are visited, and collect all root node sets as the core set R; where is the Euclidean distance from u to v; the k nearest neighbors are the set of k samples with the smallest Euclidean distances; the elbow distance is the distance corresponding to the point with the largest curvature on the curve obtained by plotting the Euclidean distances from sample u to all other samples in its k nearest neighbors and sorting them by size; v is a sample in the k nearest neighbors of u; it is used to capture multi-level density peaks, corresponding to the hierarchical inflection points from minor fluctuations to strong events.

[0041] Example 4. Refer to Figure 1 . Based on the above example, the outlier correction module uses the anomaly score to fuse the Local Outlier Factor (LOF) to detect outliers, and calculates the anomaly score of each sample , and the formula used is: ; ; calculate the global anomaly score , and the formula used is: ; where is the length of the path to be isolated; E[·] is the expectation; is the expected path length required for sample to be isolated; is the sample normalization constant; is the harmonic number; is the anomaly weight; is the local outlier factor of the sample; is the maximum value of the local outlier factors among all samples; is the j-th sample; N is the size of the sample subset of each tree when the samples participate in constructing the isolation forest; according to the threshold, mark the samples with high global anomaly scores as the sparse region SR, and the rest as the dense region DR; for each edge (x a , x b ) of the trend tree, if one end x a belongs to SR and the other end x b belongs to DR, then disconnect this edge of the trend tree, and regard both ends as new root nodes and add them to R; x a and x bThey are the two ends of the edges connecting the trend trees; capture instantaneous harmonic clusters through LOF and capture isolated mutations through anomaly scores; clean the offsets in the trend structure caused by abnormal perturbations to improve the recognition accuracy of the core set for abnormal events.

[0042] Example Five, refer to Figure 1 , based on the above example, the preliminary clustering module takes the i-th sample as the core point, and for the core point , let the k-nearest neighbor set of the core point be NN(x i ); for each core point , connect edges with the points belonging to NN(x i ) to form a subgraph ; each subgraph corresponds to a cluster; introduce a density consistency factor to resist merges with overly large density differences, and calculate the merge index, with the formula used being: ; ; ; ; where is the merge index; both i and j are sample indices, i ≠ j; is the subgraph obtained with the core point ; and are the local densities of samples x and y respectively; is the subgraph and the subgraph 's nearest Euclidean distance; is the density normalization parameter; is the k-nearest neighbor set of the sample; iteratively merge the maximum pairs until the number of remaining clusters is K to obtain the final core clustering; is the sample of the subgraph ; is the sample of the subgraph ; Subdivide the normal fluctuation patterns into K typical operating states to support quality control under different working conditions; e is the natural constant.

[0043] Example Six, refer to Figure 1 , based on the above example, the edge boundary perturbation allocation module takes samples that have not been assigned to any cluster as boundary points , corresponding to the uncertain points of slight flicker and harmonics, and calculates its allocation confidence for the c-th cluster, with the formula used being: ; where is the boundary point and The total Euclidean distance between points; assign boundary points to the cluster with the maximum distribution confidence; ensure that boundary points are smoothly assigned according to local similarity to avoid hard threshold misclassification; for the clustering results of historical power quality monitoring data after boundary disturbance allocation, use silhouette coefficient evaluation and set silhouette coefficient threshold. If the average silhouette coefficient of all samples is lower than the silhouette coefficient threshold, clustering optimization is performed; otherwise, the clustering of historical power quality monitoring data is completed.

[0044] By performing the above operations, the general photovoltaic power quality control system is prone to miss weak interference or falsely report strong interference, is seriously disturbed by abnormal points, and cannot adapt to the density differences of the photovoltaic system under multiple working conditions, which leads to low accuracy of photovoltaic power quality control. This scheme adaptively captures the layered inflection points from small fluctuations to severe interference, and more flexibly identifies events of different intensities; captures transient harmonic clusters through LOF, captures mutation points through isolated forests, and eliminates trend deviations caused by abnormal disturbances; introduces density consistency constraints to calculate the merging index to prevent erroneous merging between clusters of different densities, and adapts to various working conditions of photovoltaic operation; thereby improving the accuracy of photovoltaic power quality control.

[0045] Embodiment 7, see Figure 1 and Figure 2 This embodiment is based on the above embodiment, and the clustering optimization module specifically includes the following contents:

[0046] Initialize individual positions; construct parameter optimization space based on target cluster number, density normalization parameter, k and silhouette coefficient threshold; quasi-initialization of the qth dimension of each particle is expressed as: ; Generate individual positions and obtain the final initialized particles. The formula used is: ; The average silhouette coefficient of all samples clustered based on individual positions is taken as the individual fitness value; the lower the individual fitness value, the worse the solution corresponding to the individual; where, is the quasi-initial position of the mth particle in the qth dimension; and are the upper and lower limits of the qth dimension respectively; It is uniformly distributed A random number; is the final initial position of the mth particle in the qth dimension; rnd(·,·) is a function that takes uniform random values ​​in a given interval;

[0047] Particle position update; introduce position dynamic balance mechanism and define time function. The formula used is: ; ; The formula used to update speed and position is: ; ;in, is the dynamic cognitive learning factor; g is the current iteration number; is the maximum number of iterations; is the inertia weight; is the search mode switching threshold; and are the q-th dimension velocity of the m-th particle at the g+1-th iteration and the g-th iteration respectively; , and They are all random numbers between 0 and 1, independent of each other; and are the positions of the qth dimension of the mth particle at the g+1th iteration and the gth iteration respectively; It is a social learning factor; is the position of the optimal particle in the qth dimension; F is a random number that obeys the uniform distribution U(0,1), and Independent of each other; is the historical optimal position of the mth particle in the qth dimension;

[0048] Local search enhancement: every k iterations, the 10% individuals with the lowest fitness values ​​are perturbed and adjusted. The formula used is: ;in, is the position after local search enhancement; The degrees of freedom are t-distributed random variable;

[0049] Optimization judgment; when there is an individual fitness value higher than the silhouette coefficient threshold, the clustering result of the historical power quality monitoring data is obtained based on the individual position; if the maximum number of iterations is reached, return to the initialized individual position; otherwise, return to the particle position update.

[0050] By performing the above operations, in view of the fact that general photovoltaic power quality control systems are unable to adapt to complex operating environments and find it difficult to take into account both global exploration and local development at the same time, resulting in poor photovoltaic power quality control effects, this scheme reduces the probability of blind spots through initialization, enhances initial diversity, and is more adaptable to the characteristics of wide fluctuation range and large distribution differences in photovoltaic data; introduces a dynamic balance mechanism to balance global search and local convergence, and adapts to the non-stationarity of photovoltaic data over time; avoids long-term retention of inferior solutions based on a local search enhancement mechanism; and ultimately improves the photovoltaic power quality control effect.

[0051] Embodiment 8, see Figure 1, based on the above embodiment, the photovoltaic power quality control module selects the label with the largest number in the cluster as the cluster label based on the clustering results of historical power quality monitoring data; collects real-time power quality monitoring data and performs clustering assignment based on the principle of the minimum Euclidean distance; uses the cluster label corresponding to the real-time power quality monitoring data as the power quality monitoring result; if the monitoring result is moderately abnormal, increases the sampling frequency of the monitoring data; if the monitoring result is severely abnormal, conducts a warning process; thus realizing power quality control; the power quality monitoring data after clustering assignment is regarded as historical power quality monitoring data, and periodically returns to update the clustering results of historical power quality monitoring data.

[0052] It should be noted that in this article, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device comprising a series of elements not only includes those elements, but also includes other elements not expressly listed, or further includes elements inherent to such process, method, article or device.

[0053] Although the embodiments of the present invention have been shown and described, those of ordinary skill in the art can understand that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principle and spirit of the present invention.

[0054] The above describes the present invention and its implementation manners. Such description is not restrictive. What is shown in the drawings is only one of the implementation manners of the present invention, and the actual structure is not limited thereto. Generally speaking, if those of ordinary skill in the art are inspired by it and design similar structural manners and embodiments without creative efforts without departing from the purpose of the present invention, they shall fall within the protection scope of the present invention.

Claims

1. An intelligent photovoltaic power quality control system, characterized by: The system includes a photovoltaic data acquisition module, a feature space trend graph construction module, an isolated correction module, a preliminary clustering module, a boundary disturbance allocation module, a clustering optimization module and a photovoltaic power quality control module; The photovoltaic data acquisition module collects power quality data of the photovoltaic system to form a sample set; The feature space trend graph construction module constructs a fully connected graph between samples and constructs a core set by calculating elbow point distances and adjacency relationships; The isolated correction module detects isolated points and improves the core set through anomaly scores; The preliminary clustering module forms preliminary clusters based on the core points and their neighborhood densities, and introduces a density consistency factor to merge clusters; The boundary disturbance allocation module allocates confidence to the boundary points based on local similarity calculation, and finally allocates the boundary points; The clustering optimization module automatically adjusts clustering parameters based on the particle swarm optimization algorithm, and introduces a dynamic balance mechanism and a local search enhancement strategy; The photovoltaic power quality control module implements photovoltaic power quality control on real-time power quality monitoring data based on the clustering results of historical power quality monitoring data.

2. The intelligent photovoltaic power quality control system according to claim 1, characterized in that: The feature space trend graph construction module is to construct a fully connected graph between samples, with edge weights as Euclidean distances, and initialize all samples to be unvisited; for each sample u, the elbow distance between the sample and its k nearest neighbors is calculated. , connect the Euclidean distance Less than neighbors; take sample u as the leaf node and sample v as the parent node, jump to the nearest neighbor each time, and take the stopping point as the root node; repeat until all points are visited, and collect all root node sets as the core set R; where, is the Euclidean distance from u to v; k nearest neighbors is the set of k samples with the smallest Euclidean distance; elbow distance It is the distance corresponding to the point with the largest curvature on the curve after plotting the Euclidean distances of sample u to all other k-nearest neighbors and sorting them by size; v is the sample among u's k-nearest neighbors.

3. The intelligent photovoltaic power quality control system according to claim 2 is characterized by: The isolated correction module uses the anomaly score to fuse the local outlier factor LOF to detect isolated points and calculate the anomaly score of each sample. , the formula used is: ; ; Calculating the global anomaly score , the formula used is: ;in, is the isolated path length; E[·] is the expectation; It is a sample the expected path length required to be isolated; is the sample normalization constant; is a harmonic number; is the anomaly weight; is the local outlier factor of the sample; is the maximum value of the local outlier factor among all samples; is the jth sample; N is the size of each tree sample subset when the sample participates in building an isolation forest; according to the threshold, the samples with high global anomaly scores are marked as sparse areas SR, and the rest are dense areas DR; for each trend tree edge (x a ,x b ), if one end x a Belongs to SR, the other end x b If it belongs to DR, disconnect the trend tree edge and treat both ends as new root nodes and add R; x a and x b They are the two ends of the trend tree edge.

4. The intelligent photovoltaic power quality control system according to claim 3 is characterized by: The preliminary clustering module is to cluster the i-th sample As the core point, the core point , let the core point k nearest neighbor set be NN(x i ); For each core point , and belongs to NN(x i ) to form a subgraph ; Each subgraph corresponds to a cluster; introduce density consistency factor , calculate the merger index, the formula used is: ; ; ; ;in, is the merge index; i and j are the indices of the samples, i≠j; The core point The obtained subgraph; and are the local densities of samples x and y respectively; It is a subgraph and subgraph The closest Euclidean distance between them; is the density normalization parameter; is the k-nearest neighbor set of the sample; iterative merging of the maximum Yes, until the number of remaining clusters is K, and the final core clustering is obtained; It is a subgraph Samples of It is a subgraph A sample of ; e is a natural constant.

5. The intelligent photovoltaic power quality control system according to claim 4 is characterized by: The boundary perturbation allocation module is to use samples that are not classified into any cluster as boundary points , calculate its allocation confidence for the cth cluster , the formula used is: ;in, It is a boundary point and The total Euclidean distance between points; the boundary points are assigned to the cluster with the maximum allocation confidence; for the clustering results of the historical power quality monitoring data after the boundary disturbance is assigned, the silhouette coefficient is evaluated and the silhouette coefficient threshold is set. If the average silhouette coefficient of all samples is lower than the silhouette coefficient threshold, clustering optimization is performed; otherwise, the clustering of the historical power quality monitoring data is completed.

6. The intelligent photovoltaic power quality control system according to claim 5, characterized in that: The clustering optimization module specifically includes the following contents: Initialize individual positions; construct parameter optimization space based on target cluster number, density normalization parameter, k and silhouette coefficient threshold; quasi-initialization of the qth dimension of each particle is expressed as: ; Generate individual positions and obtain the final initialized particles. The formula used is: ; The average silhouette coefficient of all samples clustered based on individual positions is used as the individual fitness value; The lower the individual fitness value, the worse the solution corresponding to the individual; is the quasi-initial position of the mth particle in the qth dimension; and are the upper and lower limits of the qth dimension respectively; It follows a uniform distribution A random number; is the final initial position of the mth particle in the qth dimension; rnd(·,·) is a function that takes uniform random values ​​in a given interval; Particle position update; introduce the position dynamic balance mechanism and define the time function. The formula used is: ; ; The formula used to update speed and position is: ; ;in, is the dynamic cognitive learning factor; g is the current iteration number; is the maximum number of iterations; is the inertia weight; is the search mode switching threshold; and are the q-th dimension velocity of the m-th particle at the g+1-th iteration and the g-th iteration respectively; , and They are all random numbers between 0 and 1, independent of each other; and are the positions of the mth particle in the qth dimension at the g+1th iteration and the gth iteration respectively; It is a social learning factor; is the position of the optimal particle in the qth dimension; F is a random number that obeys the uniform distribution U(0,1), and Independent of each other; is the historical optimal position of the mth particle in the qth dimension; Local search enhancement: every k iterations, the 10% individuals with the lowest fitness values ​​are perturbed and adjusted. The formula used is: ;in, is the position after local search enhancement; The degrees of freedom are t-distributed random variable; Optimization judgment; when there is an individual fitness value higher than the silhouette coefficient threshold, the clustering result of the historical power quality monitoring data is obtained based on the individual position; if the maximum number of iterations is reached, return to the initialized individual position; otherwise, return to the particle position update.

7. The intelligent photovoltaic power quality control system according to claim 6, characterized in that: The photovoltaic data acquisition module collects historical power quality monitoring data according to a fixed sampling time sequence during the operation of the photovoltaic system, pre-processes the collected data, and forms a sample set. ,in, and are the 1st and nth samples respectively; each sample is an extracted feature vector; the operating status of each sample is labeled, and the operating status is used as a data label and is only selected and used as a cluster label; the operating status includes normal operation, slight abnormality, moderate abnormality and severe abnormality.

8. The intelligent photovoltaic power quality control system according to claim 7, characterized in that: The photovoltaic power quality control module selects the label with the largest number in the cluster as the cluster label based on the clustering result of historical power quality monitoring data; collects real-time power quality monitoring data and performs clustering allocation based on the minimum Euclidean distance principle; uses the cluster label corresponding to the real-time power quality monitoring data as the power quality monitoring result; if the monitoring result is moderately abnormal, increases the sampling frequency of the monitoring data; if the monitoring result is seriously abnormal, performs early warning processing; thereby realizing power quality control.

Images