A method and system for monitoring intelligent new energy power generation equipment

Abstract

The application discloses a kind of intelligent new energy power generation equipment monitoring method and system, belong to power generation monitoring technical field, method includes: power generation equipment data integration, power generation equipment nearest neighbor number search, power generation equipment neighborhood guide clustering and power generation equipment monitoring.This scheme is according to the step length and nearest neighbor number of dynamic adjustment degree of density, obtain temporary optimal nearest neighbor number, based on the nearest neighbor overlap degree and near neighbor efficiency difference between adjacent iterations, carry out near neighbor stability test, determine optimal nearest neighbor number, improve the accuracy of equipment state identification;According to local density screening neighborhood optimization point, combine correlation degree and fluctuation index two-dimensional sorting, determine attribution guide point, based on local density and near neighbor representative proportion setting threshold value, determine condensation center point, according to condensation center point and correlation degree constructs initial cluster, based on affinity and correlation degree calculation cluster correlation, obtain final clustering result, provide reliable basis for the state judgment of new energy power generation equipment.

Description

Technical Field

[0001] This invention belongs to the field of power generation monitoring technology, specifically referring to an intelligent monitoring method and system for new energy power generation equipment. Background Technology

[0002] The monitoring method for new energy power generation equipment is a technical solution that integrates big data analysis and artificial intelligence algorithms. By analyzing multi-dimensional data of new energy power generation equipment in real time, it identifies abnormal characteristics and potential failure risks in equipment operation, providing scientific decision-making basis for operation and maintenance personnel, power generation companies and energy regulatory departments, thereby ensuring the stable and safe operation of new energy power generation equipment.

[0003] However, existing monitoring methods for new energy power generation equipment suffer from several problems. Equipment operation data is affected by environmental factors, resulting in significant fluctuations in data density distribution. This makes it difficult to accurately capture the dynamic changes in data distribution, leading to low accuracy in equipment status identification and poor consistency in monitoring results. Furthermore, existing monitoring methods for new energy power generation equipment suffer from non-uniform distribution of equipment operation data due to environmental changes. Traditional clustering methods tend to misclassify similar data under different environments into the same category, and the selection of cluster centers lacks clear guidance, failing to dynamically adapt to the data distribution characteristics under different equipment states. This results in low accuracy in equipment status identification and slow monitoring response. Summary of the Invention

[0004] To address the aforementioned issues and overcome the shortcomings of existing technologies, this invention provides an intelligent monitoring method and system for new energy power generation equipment. Addressing the problems of existing monitoring methods where equipment operation data is significantly affected by environmental factors, resulting in fluctuating data density distribution and difficulty in accurately capturing dynamic changes in data distribution, leading to low accuracy in equipment status identification and poor consistency in monitoring results, this solution performs high-density effective data filtering based on initial local density. It dynamically adjusts the step size and nearest neighbor number according to the density to obtain a temporary optimal nearest neighbor number. Based on the nearest neighbor overlap and nearest neighbor effectiveness difference between adjacent iterations, a nearest neighbor stability test is performed to determine the optimal nearest neighbor number, providing reliable parameters for subsequent clustering. This improves the accuracy of equipment status identification and the consistency of monitoring results, avoiding the impact of monitoring jumps on operation and maintenance decisions. This addresses the shortcomings of existing new energy power generation equipment monitoring methods. The current method suffers from the problem that equipment operation data is unevenly distributed due to environmental changes. Traditional clustering tends to misclassify similar data under different environments into the same category, and the selection of cluster centers lacks clear guidance, failing to dynamically adapt to the data distribution characteristics under different equipment states. This results in low accuracy of equipment status identification and slow monitoring response. This solution selects optimal neighborhood points based on local density, and ranks them using both correlation and volatility indices to determine the guiding points. Based on local density and the proportion of nearest neighbor representation, local aggregation thresholds and nearest neighbor attribution thresholds are set to determine agglomeration centers. Initial clusters are constructed based on agglomeration centers and correlation. Cluster correlation is calculated based on affinity and correlation, and the clusters with the highest correlation are iteratively merged to obtain the final clustering result. This improves the accuracy of equipment status classification and monitoring timeliness, providing a reliable basis for judging the status of new energy power generation equipment and ensuring that maintenance personnel can respond quickly to equipment anomalies.

[0005] The technical solution adopted by this invention is as follows: This invention provides an intelligent monitoring method for new energy power generation equipment, which includes the following steps:

[0006] Step S1: Data integration of power generation equipment;

[0007] Step S2: Nearest neighbor search for power generation equipment;

[0008] Step S3: Neighborhood-guided clustering of power generation equipment;

[0009] Step S4: Monitoring of power generation equipment.

[0010] Furthermore, in step S1, the data integration of the power generation equipment involves collecting historical and real-time data of new energy power generation equipment. Both the historical and real-time data include photovoltaic module-related data, environmental data, and inverter data. The historical data is labeled with tags, including normal status, abnormal warning status, and fault status. The collected historical and real-time data are then cleaned and normalized.

[0011] Further, in step S2, the nearest neighbor search of the power generation equipment specifically includes the following steps:

[0012] Step S21: High-density effective data screening; For each historical new energy power generation equipment data, calculate the initial local density of the data based on the initial nearest neighbor number, and count the number n of data whose initial local density is not lower than its own local density among its nearest neighbors. Mark historical new energy power generation equipment data with a number n≥2 as effective data. Integrate all effective data and real-time new energy power generation equipment data to construct a power generation equipment monitoring dataset.

[0013] Step S22: Adaptive step size neighbor number optimization; Set an adaptive density threshold based on the initial local density distribution of all data in the power generation equipment monitoring dataset. Based on the current nearest neighbor number, determine the nearest neighbor set for each data in the power generation equipment monitoring dataset. Based on the nearest neighbor set, obtain the density of the power generation equipment monitoring dataset. Set an initial step size. Dynamically adjust the step size according to the density. Adjust the nearest neighbor number again according to the adjusted step size. Repeat the iteration to adjust the step size and the nearest neighbor number until the proportion of data without anti-nearest neighbors in the power generation equipment monitoring dataset is no more than 5%. Stop the iteration and use the nearest neighbor number at this time as the temporary preferred nearest neighbor number.

[0014] Step S23: Nearest Neighbor Stability Check; Starting with the temporarily optimized nearest neighbor number, continue to perform two more iterations to obtain the nearest neighbor sets corresponding to three consecutive iterations. Calculate the nearest neighbor overlap and the difference in nearest neighbor efficiency between two groups of adjacent iterations. If the nearest neighbor overlap of the first group is not less than that of the second group, and the difference in nearest neighbor efficiency of the first group is not greater than that of the second group, then it is determined that the nearest neighbor set has reached a stable state, and the nearest neighbor number corresponding to the last iteration in the first group of adjacent iterations is taken as the optimal nearest neighbor number; otherwise, take the nearest neighbor number of the latest iteration as the new starting point, and perform two more iterations until the nearest neighbor set reaches a stable state and then stop iterating.

[0015] Furthermore, in step S3, the neighborhood-guided clustering of the power generation equipment specifically includes the following steps:

[0016] Step S31: Neighborhood Guiding Point Selection; For each data point in the power generation equipment monitoring dataset, select data points with local density not lower than the local density of the data itself from its best nearest neighbor set as neighborhood guiding points. Calculate the correlation degree between the data and the neighborhood guiding points, as well as the fluctuation index of the neighborhood guiding points. Sort the neighborhood guiding points by correlation degree from largest to smallest and by fluctuation index from smallest to largest. Only retain neighborhood guiding points that are simultaneously in the top 20% of both correlation degree and fluctuation index. If only one neighborhood guiding point is retained, select it as the guiding point for the data. If multiple neighborhood guiding points are retained, select the neighborhood guiding point with the highest correlation degree as the guiding point for the data. If no neighborhood guiding points are retained, designate the data itself as the guiding point.

[0017] Step S32: Adaptive extraction of agglomeration centers; Select data from the power generation equipment monitoring dataset that belong to itself as the guiding point and whose local density is not less than the local agglomeration threshold as agglomeration candidate points. For each agglomeration candidate point, calculate the proportion of data in its best nearest neighbor set that belong to the guiding point of the agglomeration candidate point, and obtain the nearest neighbor representative proportion of each agglomeration candidate point. Only retain agglomeration candidate points whose proportion is not less than the nearest neighbor belonging threshold as agglomeration center points.

[0018] Step S33: Assign and build initial clusters; create an initial cluster for each cohesive center point and include the cohesive center point itself in the cluster. For all non-cohesive center points in the power generation equipment monitoring data, calculate their correlation with all cohesive center points, and assign each non-cohesive center point to the cluster corresponding to the cohesive center point with the highest correlation, thus completing the construction of all initial clusters.

[0019] Step S34: Cluster-related adaptive merging; Based on the affinity between data and clusters, and the correlation between agglomeration centers, the cluster correlation between each pair of clusters is obtained. The two clusters with the highest cluster correlation are merged into a new cluster, and the mean of the agglomeration centers of the two clusters is used as the agglomeration center of the new cluster. The cluster merging operation is repeated until the cluster correlation between all clusters is less than the merging threshold, and the final clustering result is obtained. The label with the most historical new energy power generation equipment data in each cluster is selected as the cluster label.

[0020] Furthermore, in step S4, the monitoring of the power generation equipment is based on the final clustering result, and the cluster label to which the real-time new energy power generation equipment data belongs is used as the monitoring result to obtain the status corresponding to the real-time new energy power generation equipment data.

[0021] The present invention provides an intelligent monitoring system for new energy power generation equipment, comprising a power generation equipment data integration module, a power generation equipment nearest neighbor number search module, a power generation equipment neighborhood guided clustering module, and a power generation equipment monitoring module;

[0022] The power generation equipment data integration module collects historical and real-time new energy power generation equipment data, performs data cleaning and normalization, and sends the data to the power generation equipment nearest neighbor search module.

[0023] The nearest neighbor search module for power generation equipment performs high-density effective data filtering based on the initial local density, dynamically adjusts the step size and the number of nearest neighbors according to the density, obtains a temporary preferred number of nearest neighbors, performs a nearest neighbor stability test based on the nearest neighbor overlap and the difference in nearest neighbor effectiveness between adjacent iterations, determines the optimal number of nearest neighbors, and sends the data to the power generation equipment neighborhood guided clustering module.

[0024] The power generation equipment neighborhood-guided clustering module filters neighborhood optimization points according to local density, sorts them by correlation degree and fluctuation index to determine the belonging guidance point, sets a threshold based on local density and the proportion of nearest neighbor representatives to determine the agglomeration center point, constructs an initial cluster based on the agglomeration center point and correlation degree, calculates the cluster correlation based on affinity and correlation degree, iteratively merges the cluster with the highest correlation to obtain the final clustering result, and sends the data to the power generation equipment monitoring module.

[0025] The power generation equipment monitoring module uses the cluster label to which the real-time new energy power generation equipment data belongs as the monitoring result based on the final clustering result.

[0026] The beneficial effects achieved by adopting the above solution are as follows:

[0027] (1) In view of the problems in the existing monitoring methods for new energy power generation equipment, the equipment operation data is affected by environmental factors, the data density distribution fluctuates significantly, and it is difficult to accurately capture the dynamic change structure of the data distribution, resulting in low equipment status identification accuracy and poor consistency of monitoring results. This solution uses high-density effective data screening based on the initial local density to effectively eliminate redundant data interference and improve the data quality foundation for equipment status identification. The step size and the number of nearest neighbors are dynamically adjusted according to the density to obtain a temporary preferred number of nearest neighbors, so that the number of nearest neighbors is dynamically adjusted with the equipment data density, adapting to the characteristics of environmentally sensitive data and accurately matching the real-time data distribution structure. Based on the nearest neighbor overlap and the difference in nearest neighbor effectiveness between adjacent iterations, the nearest neighbor stability test is performed to determine the optimal number of nearest neighbors, providing reliable parameters for subsequent clustering, improving the accuracy of equipment status identification and the consistency of monitoring results, and avoiding the impact of monitoring jumps on operation and maintenance decisions.

[0028] (2) In response to the problems of non-uniform distribution of equipment operation data due to environmental changes in existing monitoring methods for new energy power generation equipment, traditional clustering is prone to misclassifying similar data under different environments into the same category, and the selection of cluster centers lacks clear guidance and cannot dynamically adapt to the data distribution characteristics under different equipment states, resulting in low accuracy of equipment state identification and slow monitoring response. This solution selects neighborhood optimization points based on local density, and determines the belonging guidance point by combining correlation degree and fluctuation index in a two-dimensional sorting, providing a clear clustering direction for each data and reducing state misjudgment under non-uniform data distribution; sets local aggregation threshold and nearest neighbor belonging threshold based on local density and nearest neighbor representative ratio to determine the aggregation center point, ensuring that the center matches the data distribution characteristics of each equipment state, providing a reliable core for subsequent clustering, and avoiding state identification errors caused by center deviation; constructs initial clusters based on aggregation centers and correlation degree, calculates cluster correlation based on affinity and correlation degree, iteratively merges the clusters with the highest correlation, and obtains the final clustering result, improving the accuracy of equipment state classification and monitoring timeliness, providing a reliable basis for the state judgment of new energy power generation equipment, and ensuring that maintenance personnel can respond quickly to equipment anomalies. Attached Figure Description

[0029] Figure 1 A flowchart illustrating an intelligent monitoring method for new energy power generation equipment provided by the present invention;

[0030] Figure 2 A schematic diagram of an intelligent new energy power generation equipment monitoring system provided by the present invention;

[0031] Figure 3 This is a flowchart illustrating step S2;

[0032] Figure 4 This is a flowchart illustrating step S3.

[0033] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used together with the embodiments of the invention to explain the invention and do not constitute a limitation thereof. Detailed Implementation

[0034] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0035] In the description of this invention, it should be understood that the terms "upper", "lower", "front", "rear", "left", "right", "top", "bottom", "inner", "outer", etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this invention.

[0036] Example 1, see Figure 1 The present invention provides an intelligent monitoring method for new energy power generation equipment, the method comprising the following steps:

[0037] Step S1: Data integration of power generation equipment; collect historical and real-time data of new energy power generation equipment, and perform data cleaning and normalization;

[0038] Step S2: Nearest neighbor search for power generation equipment; based on the initial local density, high-density effective data is filtered, and the step size and the number of nearest neighbors are dynamically adjusted according to the density to obtain a temporary preferred number of nearest neighbors. Based on the nearest neighbor overlap and the difference in nearest neighbor effectiveness between adjacent iterations, the nearest neighbor stability test is performed to determine the optimal number of nearest neighbors.

[0039] Step S3: Neighborhood-guided clustering of power generation equipment; select the best neighborhood points according to local density, sort them by correlation degree and fluctuation index to determine the guiding points, set the threshold based on local density and the proportion of nearest neighbor representatives to determine the agglomeration center points, construct the initial clusters based on the agglomeration center points and correlation degree, calculate the cluster correlation based on affinity and correlation degree, iteratively merge the clusters with the highest correlation to obtain the final clustering results;

[0040] Step S4: Power generation equipment monitoring; Based on the final clustering results, the cluster label to which the real-time new energy power generation equipment data belongs is used as the monitoring result.

[0041] Example 2, see Figure 1This embodiment is based on the above embodiment. In step S1, the data integration of power generation equipment involves collecting historical and real-time data from new energy power generation equipment. New energy power generation equipment refers to solar power generation equipment. Both the historical and real-time data include photovoltaic module-related data, environmental data, and inverter data. The photovoltaic module-related data, environmental data, and inverter data are collected every 5 minutes. The photovoltaic module-related data includes voltage, current, power, and surface temperature. The environmental data includes light intensity, ambient temperature, wind speed, and wind direction. The inverter data includes DC input voltage, DC input current, AC output voltage, and AC output current. The system collects data on current, AC output frequency, and conversion efficiency. Historical data on renewable energy power generation equipment is labeled with tags indicating normal status, abnormal warning status, and fault status, used only for clustering result evaluation. The labeling criteria are: a conversion efficiency of 90% or higher indicates a normal status; a conversion efficiency less than 90% but not less than 75% indicates an abnormal warning status; and a conversion efficiency less than 75% or an AC output frequency of 0 indicates a fault status. The collected historical and real-time renewable energy power generation equipment data undergo data cleaning and normalization. Data cleaning employs the 3σ principle to remove erroneous values ​​from the data. Data normalization uses a max-min scaling method to unify numerical data to the range [0, 1].

[0042] Example 3, see Figure 1 and Figure 3 This embodiment is based on the above embodiment. In step S2, the nearest neighbor search of the power generation equipment specifically includes the following:

[0043] Step S21: High-density effective data screening; Historical renewable energy power generation equipment data contains low-density redundant data, which are mostly abnormal data from equipment start-up / shutdown transitions or extreme environments. Directly including these in the monitoring dataset would lead to biases in subsequent clustering and reduce the accuracy of equipment status identification. For each historical renewable energy power generation equipment data point, based on the initial nearest neighbor count... The initial local density of the data is calculated, and the number n of data points whose initial local density is not lower than its own local density among its nearest neighbors is counted. Historical new energy power generation equipment data with a number n≥2 are considered valid data. All valid data and real-time new energy power generation equipment data are integrated to construct a power generation equipment monitoring dataset. Redundant low-density data are removed to provide high-quality data support for subsequent nearest neighbor number optimization and clustering, thereby improving the reliability of equipment status monitoring. The formulas used are as follows:

[0044] ;

[0045] In the formula, x i and x jThese are the i-th and j-th historical data points for new energy power generation equipment, where i and j are data indices. It is x i The initial local density, k0 is the minimum nearest neighbor number, It rounds up, where N is the feature dimension of historical new energy power generation equipment data. It is x i The initial nearest neighbor set, It is x i and x j The Euclidean distance between them It is x i and The average Euclidean distance of all nearest neighbor data in the dataset. It is a function that takes the minimum value. It is the natural exponential function;

[0046] Step S22: Adaptive Step Size Neighbor Optimization; The operating data of new energy power generation equipment changes dynamically with environmental factors such as sunlight and temperature. A fixed nearest neighbor number cannot adapt to fluctuations in data density, easily leading to overly coarse or overly fine neighborhood division during clustering, affecting the equipment status differentiation effect. An adaptive density threshold is set based on the initial local density distribution of all data in the power generation equipment monitoring dataset. Based on the current nearest neighbor number, the nearest neighbor set for each data point in the power generation equipment monitoring dataset is determined. The density of the power generation equipment monitoring dataset is obtained based on the nearest neighbor set, and an initial step size is set. The step size is dynamically adjusted based on the density, and the nearest neighbor number is adjusted again based on the adjusted step size. This iterative adjustment of the step size and the nearest neighbor number is repeated until the proportion of data without anti-nearest neighbors in the power generation equipment monitoring dataset is no greater than 5%. The nearest neighbor number at this point is used as the temporary preferred nearest neighbor number. This ensures that the nearest neighbor number accurately reflects the data distribution structure, provides a reasonable basis for neighborhood partitioning for subsequent clustering, and improves the accuracy and adaptability of equipment status clustering. The formula used is as follows:

[0047] ;

[0048] ;

[0049] ;

[0050] In the formula, A represents the density of the power generation equipment monitoring dataset at the t-th iteration, where t is the iteration number index. eff It is a dataset of power generation equipment monitoring data. It is A eff The number of data in x a and x b These are the a-th and b-th data points in the power generation equipment monitoring dataset, where a and b are data indices. It is x b Initial local density, k t and k t+1 These are the nearest neighbor numbers at the t-th and t+1-th iterations, respectively. When x is in the t-th iteration a The nearest neighbor set, and These are the step sizes for the t-th and t+1-th iterations, respectively. and These are the mean and standard deviation of the initial local density of all data in the power generation equipment monitoring dataset, x. a The anti-nearest neighbor refers to the centralization of power generation equipment monitoring data with x as the nearest neighbor. a For nearest neighbor data, It is a function that takes the maximum value;

[0051] Step S23: Nearest Neighbor Stability Check; The temporary nearest neighbor number obtained by adaptive step-size optimization may fluctuate and cannot ensure stability in scenarios with dynamic changes in new energy equipment data. If directly used for clustering, it may cause frequent jumps in equipment status monitoring results, affecting operation and maintenance decisions; Starting from the temporary optimal nearest neighbor number, continue to perform two iterations of adjustment to obtain t. temp t temp+1 and t temp+2 For the nearest neighbor sets corresponding to three consecutive iterations, calculate the nearest neighbor overlap and the difference in nearest neighbor efficiency between two adjacent sets of iterations. The first set is (t temp , t temp+1 The second group is (t) temp+1 , t temp+2 If the nearest neighbor overlap of the first group is not less than that of the second group, and the difference in nearest neighbor effectiveness between the first group and the second group is not greater than that between the two groups, then the nearest neighbor set is considered to have reached a stable state, and the last iteration t in the adjacent iterations of the first group is taken. temp+1 The corresponding nearest neighbor number is the optimal nearest neighbor number; otherwise, the nearest neighbor number from the latest iteration is used as the new starting point, and two more iterations are performed for adjustment until the nearest neighbor set reaches a stable state, at which point the iteration stops; this provides reliable parameters for subsequent neighborhood-guided clustering, improves the consistency of device status monitoring, and ensures that maintenance personnel can formulate maintenance strategies based on stable results; the formula used is as follows:

[0052] ;

[0053] ;

[0054] In the formula, k temp It is the temporary preferred nearest neighbor number, t temp It is the number of iterations corresponding to the temporary selection of the nearest neighbor, t. temp+1 and t temp+2It is t temp The number of iterations in the next two iterations, and They are respectively the tth temp The second and the tth temp+1 The nearest neighbor overlap and nearest neighbor efficiency difference of the power generation equipment monitoring dataset in the next iteration. and They are respectively the tth temp The second and the tth temp+1 In the next iteration, x a The nearest neighbor set, yes and The number of data in the intersection yes and And the amount of data in the collection, yes and The number of data in the intersection yes and The number of data in the intersection and They are and The amount of data in the data.

[0055] By performing the above operations, this solution addresses the problems in existing new energy power generation equipment monitoring methods, such as the significant fluctuations in data density distribution due to environmental factors affecting equipment operation data, making it difficult to accurately capture the dynamic changes in data distribution, leading to low accuracy in equipment status identification and poor consistency in monitoring results. This solution uses initial local density to filter high-density effective data, effectively eliminating redundant data interference and improving the data quality foundation for equipment status identification. It dynamically adjusts the step size and the number of nearest neighbors based on the density to obtain a temporary optimal number of nearest neighbors, allowing the number of nearest neighbors to dynamically adjust with the equipment data density, adapting to environmentally sensitive data characteristics and accurately matching the real-time data distribution structure. Based on the nearest neighbor overlap and the difference in nearest neighbor effectiveness between adjacent iterations, a nearest neighbor stability test is performed to determine the optimal number of nearest neighbors, providing reliable parameters for subsequent clustering, improving the accuracy of equipment status identification and the consistency of monitoring results, and avoiding the impact of monitoring jumps on operation and maintenance decisions.

[0056] Example 4, see Figure 1 and Figure 4 This embodiment is based on the above embodiment. In step S3, the neighborhood-guided clustering of power generation equipment specifically includes the following steps:

[0057] Step S31: Optimization of Neighborhood Guiding Points; The distribution of new energy power generation equipment data is non-uniform due to environmental factors. Traditional clustering easily misclassifies similar data under different environments into the same category, leading to misjudgment of equipment status. Furthermore, it lacks clear guiding criteria, resulting in strong randomness in clustering results. For each data point in the power generation equipment monitoring dataset, data with a local density no lower than its own local density are selected from its best nearest neighbor set as neighborhood guiding points. The correlation between this data point and the neighborhood guiding points, as well as the fluctuation index of the neighborhood guiding points, are calculated. The neighborhood guiding points are then sorted by correlation from largest to smallest and by fluctuation index from smallest to largest. The algorithm retains neighboring points that rank in the top 20% for both correlation and volatility. If only one neighboring point is retained, it is selected as the guiding point for that data. If multiple neighboring points are retained, the one with the highest correlation is selected as the guiding point. If no neighboring points are retained, the data itself is designated as the guiding point. This rule-based selection of guiding points provides a clear clustering direction for each data point, reducing mis-clustering under non-uniform data distribution and improving the accuracy of equipment status classification. This is particularly suitable for scenarios where data distribution fluctuates due to changes in sunlight for solar energy equipment. The formula used is as follows:

[0058] ;

[0059] ;

[0060] In the formula, x c It is the c-th preferred neighborhood point, where c is the index of the preferred neighborhood point. and They are x a and x c The Euclidean distance and correlation between them, d max It is the maximum Euclidean distance between all data pairs in the power generation equipment monitoring dataset. It is x c Volatility index It is x c The optimal nearest neighbor set, yes The number of data in x f It is the f-th data in the best nearest neighbor set, where f is the data index. x is determined based on the optimal nearest neighbor number. f Local density, It is x c The average local density of all data in the optimal nearest neighbor set;

[0061] Step S32: Adaptive extraction of agglomeration centers; Traditional clustering agglomeration centers are mostly manually set or randomly selected, which cannot adapt to the data distribution characteristics of new energy equipment under different states. This can easily lead to the cluster centers deviating from the core data of the actual equipment state, affecting the accuracy of state differentiation. Data with its own guiding point and local density not less than the local clustering threshold are selected from the power generation equipment monitoring dataset as agglomeration candidate points. The local clustering threshold is the 25th percentile of the local density of all data labeled as normal state in the power generation equipment monitoring dataset. For each agglomeration candidate point, the proportion of data with its own guiding point in its best nearest neighbor set is calculated to obtain the nearest neighbor representative proportion of each agglomeration candidate point. Only agglomeration candidate points with a proportion not less than the nearest neighbor belonging threshold are retained as agglomeration center points. The nearest neighbor belonging threshold is the 30th percentile of the nearest neighbor representative proportion of all data labeled as normal state in the power generation equipment monitoring dataset. Adaptive extraction of agglomeration centers requires no manual intervention, can dynamically adapt to the data distribution under different equipment states, improve the accuracy of clustering centers, provide a reliable core for subsequent initial cluster construction, and ensure the rationality of equipment state classification. The formula used is as follows:

[0062] ;

[0063] In the formula, x g It is the g-th agglomeration candidate point, where g is the index of the agglomeration candidate point. It is x g The proportion of neighboring representatives, It is x g The optimal nearest neighbor set, yes The amount of data in Is Central Order equals x g Data x h Quantity, x h It is the h-th data in the best nearest neighbor set, where h is the data index. It is x h The guiding point of belonging;

[0064] Step S33: Assigning and Building Initial Clusters; The amount of data from new energy equipment is enormous. Directly clustering the entire dataset would be computationally complex and time-consuming, and the dispersed data distribution could lead to chaotic initial cluster divisions, affecting the efficiency of subsequent cluster merging and the speed of status monitoring. Instead, an initial cluster is created for each agglomeration center, incorporating the agglomeration center itself into the cluster. For all non-agglomeration centers in the power generation equipment monitoring dataset, the correlation between each non-agglomeration center and all agglomeration centers is calculated. Each non-agglomeration center is then assigned to the cluster corresponding to the agglomeration center with the highest correlation, completing the construction of all initial clusters. This quickly groups the data, ensuring high similarity within the initial clusters, reducing the workload of subsequent cluster merging, and ensuring that the initial clusters can initially reflect different equipment states, laying the foundation for the final clustering results.

[0065] Step S34: Cluster-related adaptive merging; Initial clusters may be over-segmented, dividing similar data under the same device status into multiple clusters, leading to misjudgment of device status. Furthermore, the lack of a unified cluster merging standard fails to adapt to the dynamic characteristics of new energy device data. Based on the affinity between data and clusters, and the correlation between agglomeration centers, the cluster correlation between every two clusters is obtained. The two clusters with the highest correlation are merged into a new cluster, and the mean of the agglomeration centers of the two clusters is used as the agglomeration center of the new cluster. This cluster merging operation is repeated until the cluster correlation between all clusters is less than the merging threshold. The final clustering results are obtained, and the label with the most historical new energy power generation equipment data in each cluster is selected as the cluster label; the similarity between clusters is comprehensively measured to avoid erroneous merging due to a single dimension of judgment, ensuring that the merging standard is adapted to the data distribution and improving the accuracy of equipment status monitoring; the formula used is as follows:

[0066] ;

[0067] ;

[0068] In the formula, x s and x u These are the s-th and u-th data items in the cluster, respectively, where s and u are data indices. It is x a and x s The degree of correlation between them, C p and C q These are the p-th and q-th clusters, respectively. and They are C p and C q The number of data items in the array, where p and q are clustered indexes. and They are x a and x u With C pThe affinity between them It is x s With C q The affinity between them, T H It is the merging threshold, μ H and σ H These are the mean and standard deviation of the cluster correlation among all clusters, x lp and x lq They are C p and C q The center of cohesion in It is x lp and x lq The degree of correlation between them It is C p and C q Cluster correlation between them.

[0069] By performing the above operations, this solution addresses the problems in existing monitoring methods for new energy power generation equipment. These problems include the non-uniform distribution of equipment operation data due to environmental changes, the tendency of traditional clustering to misclassify similar data under different environments into the same category, and the lack of clear guidance in selecting cluster centers, which fails to dynamically adapt to the data distribution characteristics under different equipment states, leading to low accuracy in equipment state identification and slow monitoring response. This solution selects optimal neighborhood points based on local density and ranks them using both correlation and volatility indices to determine the guiding points, providing a clear clustering direction for each data point and reducing misjudgments of state under non-uniform data distribution. Based on local density and the proportion of nearest neighbor representation, local aggregation thresholds and nearest neighbor attribution thresholds are set to determine agglomeration centers, ensuring that the centers match the distribution characteristics of each equipment state data point. This provides a reliable core for subsequent clustering and avoids state identification errors caused by center bias. Initial clusters are constructed based on agglomeration centers and correlation. Cluster correlation is calculated based on affinity and correlation, and the clusters with the highest correlation are iteratively merged to obtain the final clustering result. This improves the accuracy of equipment state classification and monitoring timeliness, providing a reliable basis for judging the state of new energy power generation equipment and ensuring that maintenance personnel can quickly respond to equipment anomalies.

[0070] Example 5, see Figure 1 This embodiment is based on the above embodiment. In step S4, the monitoring of power generation equipment is based on the final clustering result. The cluster label to which the real-time new energy power generation equipment data belongs is used as the monitoring result to obtain the status corresponding to the real-time new energy power generation equipment data. If the monitoring result is a normal status, the real-time power generation equipment data and the corresponding status are automatically recorded to the equipment monitoring database. If the monitoring result is an abnormal warning status, a warning prompt is generated and stored in the equipment operation and maintenance terminal, and the operation and maintenance personnel are prompted to pay close attention to the equipment operation trend. If the monitoring result is a fault status, the equipment fault warning mechanism is immediately triggered, and the equipment fault information is pushed to the equipment operation and maintenance terminal at the same time.

[0071] Example 6, see Figure 2 Based on the above embodiments, this embodiment provides an intelligent new energy power generation equipment monitoring system, including a power generation equipment data integration module, a power generation equipment nearest neighbor number search module, a power generation equipment neighborhood guided clustering module, and a power generation equipment monitoring module;

[0072] The power generation equipment data integration module collects historical and real-time new energy power generation equipment data, performs data cleaning and normalization, and sends the data to the power generation equipment nearest neighbor search module.

[0073] The nearest neighbor search module for power generation equipment performs high-density effective data filtering based on the initial local density, dynamically adjusts the step size and the number of nearest neighbors according to the density, obtains a temporary preferred number of nearest neighbors, performs a nearest neighbor stability test based on the nearest neighbor overlap and the difference in nearest neighbor effectiveness between adjacent iterations, determines the optimal number of nearest neighbors, and sends the data to the power generation equipment neighborhood guided clustering module.

[0074] The power generation equipment neighborhood-guided clustering module filters neighborhood optimization points according to local density, sorts them by correlation degree and fluctuation index to determine the belonging guidance point, sets a threshold based on local density and the proportion of nearest neighbor representatives to determine the agglomeration center point, constructs an initial cluster based on the agglomeration center point and correlation degree, calculates the cluster correlation based on affinity and correlation degree, iteratively merges the cluster with the highest correlation to obtain the final clustering result, and sends the data to the power generation equipment monitoring module.

[0075] The power generation equipment monitoring module uses the cluster label to which the real-time new energy power generation equipment data belongs as the monitoring result based on the final clustering result.

[0076] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0077] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention.

[0078] The present invention and its embodiments have been described above. This description is not restrictive, and the accompanying drawings are only one embodiment of the present invention; the actual structure is not limited thereto. In conclusion, if those skilled in the art are inspired by this description and design similar structures and embodiments without departing from the spirit of the invention, such designs should fall within the protection scope of the present invention.

Claims

1. A method for monitoring intelligent new energy power generation equipment, characterized in that: The method comprises the following steps: Step S1: power generation equipment data integration; collecting historical new energy power generation equipment data and real-time new energy power generation equipment data, and performing data cleaning and data normalization; Step S2: searching for the nearest neighbor number of the power generation equipment; performing high-density effective data screening based on the initial local density, dynamically adjusting the step size and the nearest neighbor number according to the density degree, obtaining a temporarily preferred nearest neighbor number, performing near neighbor stability inspection based on the nearest neighbor overlap degree and the near neighbor efficiency difference between adjacent iterations, and determining the optimal nearest neighbor number; Step S3: power generation equipment neighborhood guided clustering; screening neighborhood preferred points according to the local density, determining the attribution guide point by combining the correlation degree and the fluctuation index two-dimensional sorting, setting a threshold based on the local density and the near neighbor representative ratio to determine the condensation center point, constructing an initial cluster according to the condensation center point and the correlation degree, calculating the cluster correlation based on the affinity and the correlation degree, and iteratively merging the cluster with the largest correlation to obtain the final clustering result; Step S4: power generation equipment monitoring; taking the cluster label to which the real-time new energy power generation equipment data belongs as the monitoring result; In step S3, the power generation equipment neighborhood guided clustering specifically comprises the following steps: Step S31: neighborhood guide point optimization; for each data in the power generation equipment monitoring data set, in its optimal nearest neighbor set, the data with a local density not lower than the local density of the data itself is selected as a neighborhood preferred point, the correlation degree between the data and the neighborhood preferred point is calculated, and the fluctuation index of the neighborhood preferred point is calculated, the neighborhood preferred points are sorted in descending order of the correlation degree and in ascending order of the fluctuation index, only the neighborhood preferred points simultaneously ranking in the top 20% in the correlation degree and in the top 20% in the fluctuation index are reserved, if there is only one reserved neighborhood preferred point, the neighborhood preferred point is selected as the attribution guide point of the data; if there are multiple reserved neighborhood preferred points, the neighborhood preferred point with the highest correlation degree is selected as the attribution guide point of the data; if there is no reserved neighborhood preferred point, the data itself is defined as the attribution guide point; Step S32: condensation center adaptive extraction; the data with the attribution guide point being itself and the local density not being less than the local aggregation threshold value is selected from the power generation equipment monitoring data set as a condensation candidate point, for each condensation candidate point, the proportion of the data with the attribution guide point being the condensation candidate point in its optimal nearest neighbor set is calculated to obtain the near neighbor representative ratio of each condensation candidate point, only the condensation candidate points with the proportion not less than the near neighbor attribution threshold value are reserved as the condensation center points; Step S33: correlation assignment and initial cluster construction; an initial cluster is created for each condensation center point, and the condensation center point itself is included in the cluster, the correlation degrees between all non-condensation center points in the power generation equipment monitoring data set and all condensation center points are calculated, each non-condensation center point is assigned to the cluster corresponding to the condensation center point with the largest correlation degree, and the construction of all initial clusters is completed; Step S34: cluster correlation adaptive merging; based on the affinity between data and clusters and the correlation between condensation center points, the cluster correlation between each two clusters is obtained, the two clusters with the largest cluster correlation are merged into a new cluster, and the mean of the condensation center points of the two clusters is taken as the condensation center point of the new cluster, the cluster merging operation is repeatedly performed until the cluster correlation between all clusters is less than the merging threshold, and the final clustering result is obtained, and the label with the largest number of historical new energy power generation equipment data in each cluster is selected as the cluster label. 2.The intelligent new energy power generation equipment monitoring method of claim 1, wherein: In step S2, the power generation equipment nearest neighbor number search specifically includes the following steps: Step S21: high-density effective data screening; for each historical new energy power generation equipment data, the initial local density of the data is calculated based on the initial nearest neighbor number, and the number n of data with an initial local density not lower than the local density of the data itself is counted, the historical new energy power generation equipment data with a number n greater than or equal to 2 is marked as effective data, all effective data and real-time new energy power generation equipment data are integrated to construct a power generation equipment monitoring data set; Step S22: adaptive step length neighbor number optimization; an adaptive density threshold is set according to the initial local density distribution of all data in the power generation equipment monitoring data set, the nearest neighbor set of each data in the power generation equipment monitoring data set is determined based on the current nearest neighbor number, the density of the power generation equipment monitoring data set is obtained based on the nearest neighbor set, an initial step length is set, the step length is dynamically adjusted according to the density, and the nearest neighbor number is adjusted again according to the adjusted step length, the step length and the nearest neighbor number are repeatedly iterated and adjusted until the proportion of data without a reverse nearest neighbor in the power generation equipment monitoring data set is not greater than 5%, and the nearest neighbor number at this time is taken as a temporarily preferred nearest neighbor number; Step S23: nearest neighbor stability test. 3.The intelligent new energy power generation equipment monitoring method of claim 2, wherein: In step S23, the nearest neighbor stability test is performed with the temporarily preferred nearest neighbor number as the starting point, and the nearest neighbor set corresponding to three consecutive iterations is obtained by continuing to perform two iterations of adjustment, the nearest neighbor overlap degree and the nearest neighbor efficiency difference between the two adjacent iterations are calculated, if the nearest neighbor overlap degree of the first group is not less than that of the second group, and the nearest neighbor efficiency difference of the first group is not greater than that of the second group, it is determined that the nearest neighbor set has reached a stable state, and the nearest neighbor number corresponding to the last iteration in the first group of adjacent iterations is taken as the optimal nearest neighbor number; otherwise, the nearest neighbor number of the latest iteration is taken as a new starting point, and two iterations of adjustment are performed again until the nearest neighbor set reaches a stable state. 4.The intelligent new energy power generation device monitoring method of claim 3, wherein: In step S1, the power generation equipment data integration is to collect historical new energy power generation equipment data and real-time new energy power generation equipment data, the historical new energy power generation equipment data and the real-time new energy power generation equipment data both include photovoltaic module related data, environmental data and inverter data, the historical new energy power generation equipment data is labeled with a label, the label includes a normal state, an abnormal early warning state and a fault state, and the collected historical new energy power generation equipment data and real-time new energy power generation equipment data are subjected to data cleaning and data normalization. 5.The intelligent new energy power generation device monitoring method of claim 4, wherein: In step S4, the power generation equipment monitoring is based on the final clustering result, and the cluster label to which the real-time new energy power generation equipment data belongs is taken as the monitoring result to obtain the state corresponding to the real-time new energy power generation equipment data.

6. An intelligent new energy power generation equipment monitoring system for implementing the intelligent new energy power generation equipment monitoring method of any one of claims 1-5, characterized in that: The power generation equipment data integration module, the power generation equipment nearest neighbor number searching module, the power generation equipment neighborhood guided clustering module and the power generation equipment monitoring module are included. The power generation equipment data integration module collects historical new energy power generation equipment data and real-time new energy power generation equipment data, performs data cleaning and data normalization, and sends the data to the power generation equipment nearest neighbor number searching module. The power generation equipment nearest neighbor number searching module performs high-density effective data screening based on initial local density, dynamically adjusts the step size and the nearest neighbor number according to the density degree, obtains a temporary optimal nearest neighbor number, performs nearest neighbor stability inspection based on the nearest neighbor overlap degree and the nearest neighbor efficiency difference between adjacent iterations, determines the optimal nearest neighbor number, and sends the data to the power generation equipment neighborhood guided clustering module. The power generation equipment neighborhood guided clustering module screens neighborhood preferred points according to local density, determines attribution guide points in combination with correlation degree and fluctuation index two-dimensional sorting, sets a threshold based on local density and nearest neighbor representative proportion, determines condensation center points, constructs initial clusters according to the condensation center points and the correlation degree, calculates cluster correlation based on affinity and correlation degree, iteratively merges the clusters with the largest correlation, obtains the final clustering result, and sends the data to the power generation equipment monitoring module. The power generation equipment monitoring module is based on the final clustering result, and the cluster label to which the real-time new energy power generation equipment data belongs is taken as the monitoring result. The power generation equipment neighborhood guided clustering module specifically includes the following contents: Neighborhood guide point optimization; for each data in the power generation equipment monitoring data set, in its optimal nearest neighbor set, the data with local density not lower than the local density of the data itself is screened as a neighborhood preferred point, the correlation degree between the data and the neighborhood preferred point is calculated, and the fluctuation index of the neighborhood preferred point is calculated, the neighborhood preferred points are sorted in descending order of correlation degree and in ascending order of fluctuation index, only the neighborhood preferred points that are simultaneously in the top 20% in correlation degree and in the top 20% in fluctuation index are retained, if there is only one neighborhood preferred point retained, the neighborhood preferred point is selected as the attribution guide point of the data; if there are multiple neighborhood preferred points retained, the neighborhood preferred point with the highest correlation degree is selected as the attribution guide point of the data; if there is no neighborhood preferred point retained, the data itself is defined as the attribution guide point; Condensation center adaptive extraction; the data with the attribution guide point being itself and the local density not being less than the local aggregation threshold value is screened from the power generation equipment monitoring data set as a condensation candidate point, for each condensation candidate point, the proportion of the data with the attribution guide point being the condensation candidate point in its optimal nearest neighbor set is calculated to obtain the nearest neighbor representative proportion of each condensation candidate point, only the condensation candidate points with the proportion not less than the nearest neighbor attribution threshold value are retained as condensation center points; The initial clusters are assigned in association; an initial cluster is created for each condensation center point, and the condensation center point itself is included in the cluster, all non-condensation center points in the power generation equipment monitoring data set are calculated for their association degrees with all condensation center points, each non-condensation center point is assigned to the cluster corresponding to the condensation center point with the maximum association degree, and the construction of all initial clusters is completed; Cluster correlation adaptive merging; based on the affinity between data and clusters and the association degree between condensation center points, the cluster correlation between each two clusters is obtained, the two clusters with the maximum cluster correlation are merged into a new cluster, and the mean of the two cluster condensation center points is taken as the condensation center point of the new cluster, the cluster merging operation is repeatedly executed until the cluster correlation between all clusters is less than the merging threshold, the final clustering result is obtained, and the label with the most number of historical new energy power generation equipment data in each cluster is selected as the cluster label.

Images