Power grid technical improvement project evaluation method and system based on data clustering

Abstract

The invention discloses a power grid technical improvement project evaluation method and system based on data clustering, and relates to the technical field of power grid technical improvement, and the method comprises the following steps: collecting multi-dimensional data of a power grid technical improvement project, and carrying out the preprocessing of the data; clustering the preprocessed data based on a clustering algorithm, and identifying project data with the same characteristics; obtaining technical improvement evaluation influence data of the distribution box in the to-be-tested area in the same characteristic item data; constructing a technical improvement project evaluation model based on a machine learning algorithm according to the technical improvement evaluation influence data; and obtaining technical improvement project evaluation coefficients according to the technical improvement project evaluation model, generating a descending sort table according to the technical improvement project evaluation coefficients, and preferentially processing the project with the maximum technical improvement project evaluation coefficient. Through data clustering and machine learning technologies, rapid evaluation and sorting of power grid technical improvement projects are realized, decision-making efficiency is greatly improved, and due to the fact that multiple dimensions and factors are considered in the evaluation process, the evaluation result is more accurate and reliable.

Description

Technical Field

[0001] The present invention relates to the technical field of technical transformation evaluation, and more specifically, to a method and system for evaluating a power grid technical transformation project based on data clustering. Background Art

[0002] With the rapid changes in the energy industry and the proposal of sustainable development goals, the renewal and transformation of power grid equipment has become one of the core links to ensure the safe and efficient operation of the power system. At the same time, with the continuous implementation and deepening of industry reform policies such as cost auditing and electricity price determination, and stricter and tighter supervision, the company's business development faces a new and complex situation. The contradictions between price reduction and reform, stable investment and growth are gradually highlighted, and the efficiency and benefits of supporting the improvement of power grid production technical transformation investment are faced with multiple influences and complex challenges. By combining the sorting out of technical transformation investment management issues and demand analysis, scientifically constructing a refined evaluation system, and proposing investment optimization allocation technology, it can help enterprises allocate resources more accurately, optimize investment decisions, and thus maximize investment returns.

[0003] Data cluster analysis is an exploratory analysis method that aims to classify similar data points into the same category, while data points of different categories have greater differences. In the evaluation of power grid technical transformation projects, cluster analysis can be applied to the processing and analysis of a large amount of power grid operation data, thereby identifying data points with similar characteristics and providing a more accurate basis for project evaluation.

[0004] Grid technology transformation is a series of technical upgrades to improve grid operation efficiency, safety and stability. With the continuous expansion of power system scale and technological advancement, grid technology transformation projects are becoming more and more complex, requiring more scientific and objective evaluation methods to guide project implementation.

[0005] The development of big data processing technology has provided strong support for the evaluation of power grid technical transformation projects. Big data has the characteristics of large data volume and fast processing speed, and can be applied to the real-time collection, storage and analysis of power grid operation data. The comprehensive power grid evaluation system based on big data can realize the evaluation of power grid operation level and power supply capacity, and provide a scientific basis for the decision-making and implementation of power grid technical transformation projects.

[0006] The above disclosed technical solutions have at least the following technical problems: when faced with a large number of samples and multi-dimensional data, the clustering algorithm can discover potential patterns and common features in the data, but it is difficult to quantitatively score and rank each sample. This may lead to the inability to accurately judge the actual effects and priorities of different projects in the evaluation of power grid technical transformation projects. In view of the above problems, the present invention proposes a solution. Summary of the invention

[0007] In order to overcome the above-mentioned defects of the prior art, an embodiment of the present invention provides a power grid technical transformation project evaluation method and system based on data clustering, which collects multi-dimensional data of power grid technical transformation projects and performs preprocessing, uses a clustering algorithm to identify project data with the same characteristics, and then combines a machine learning algorithm to build a technical transformation project evaluation model to solve the problems raised in the above-mentioned background technology.

[0008] To achieve the above object, the present invention provides the following technical solutions: The method and system for evaluating power grid technical transformation projects based on data clustering include the following steps: collecting multi-dimensional data of power grid technical transformation projects and preprocessing the data; clustering the preprocessed data based on a clustering algorithm to identify project data with the same characteristics; obtaining technical transformation evaluation impact data of distribution boxes in a test area in the project data with the same characteristics; building a technical transformation project evaluation model based on a machine learning algorithm according to the technical transformation evaluation impact data; obtaining a technical transformation project evaluation coefficient according to the technical transformation project evaluation model, generating a descending sorting table according to the technical transformation project evaluation coefficient, and giving priority to processing projects with the largest technical transformation project evaluation coefficient.

[0009] In a preferred embodiment, the multi-dimensional data of the power grid technical transformation project is collected and the data is pre-processed, specifically: according to the evaluation objectives of the power grid technical transformation project, multi-dimensional data is collected and integrated from the internal database and external database of the power grid company using a data crawler tool to obtain a unified data set; the data set is standardized to convert the data into a unified standard deviation and mean; based on the standardized data set, the covariance matrix of each element in the data set is calculated; the covariance matrix is ​​eigenvalue decomposed to obtain eigenvalues ​​and corresponding eigenvectors; based on the size of the eigenvalue, the eigenvectors corresponding to the first k largest eigenvalues ​​are selected; the original data is converted into a new space composed of the eigenvectors corresponding to the first k largest eigenvalues ​​to obtain the data after dimensionality reduction.

[0010] In a preferred embodiment, the clustering algorithm is based on which the preprocessed data is clustered to identify project data with the same characteristics, specifically: the preprocessed data is used as the input of the clustering algorithm; the sum of squared errors under different k values ​​is calculated to obtain the optimal number of clusters k; K data points are randomly selected as initial centroids, the distance between each data point and each centroid is calculated, and the distance is assigned to the cluster where the nearest centroid is located; the centroid of each cluster is updated to the mean of all data points in the cluster; the above calculation steps are repeated until the position of the centroid reaches a preset number of iterations; the compactness of data points within the cluster and the degree of separation between clusters are calculated to evaluate the clustering effect; the data points in each cluster are observed, and the clustering results and feature distribution are intuitively displayed based on a scatter plot; based on the clustering results, the common features of each cluster are extracted; and project data with the same features are classified into the same cluster.

[0011] In a preferred embodiment, the technical transformation assessment impact data of the distribution box include equipment status coefficient, load fluctuation coefficient and safety performance coefficient; the specific method for obtaining the equipment status coefficient is as follows: operate the distribution box to be tested under preset conditions, record its various operating parameters at different time points, and the various operating parameters include temperature, pressure, current, and voltage. Conduct multiple experiments to obtain multiple groups of operating data; pre-process the collected multiple groups of operating data to remove outliers and noise; calculate the average value and standard deviation of various operating parameters when the distribution box is running; obtain the status index of the distribution box based on the average value and standard deviation of various operating parameters combined with the design parameters and historical maintenance records of the distribution box; calculate the equipment status coefficient based on the status index of the distribution box combined with the response time based on a preset formula.

[0012] In a preferred embodiment, the specific steps for obtaining the load fluctuation coefficient are as follows: real-time monitoring and recording of the load value and its changes of each load data point within a certain time window; based on the recorded data, calculating the load difference between each load data point and the previous data point, and constructing a load difference sequence; recording the change range of the load difference sequence within a period of time, and calculating the volatility of the load difference; cumulatively analyzing the load data, and calculating the cumulative effect of load fluctuations over a period of time; defining a threshold for load fluctuations to trigger an unstable state based on historical load data and system operation experience; when the cumulative effect of load fluctuations exceeds the threshold, marking the current time period as a potential unstable period; performing time series segmentation on the historical load data to obtain several time series data segments, and judging whether they are potential unstable periods respectively as historical load fluctuation events; mapping the time series data with the historical load fluctuation events based on a machine learning algorithm to generate a probability model for load fluctuations to trigger an unstable state; combining the volatility of the load difference and the probability model for load fluctuations to trigger an unstable state, the load fluctuation coefficient is calculated.

[0013] In a preferred embodiment, the specific steps for obtaining the safety performance coefficient are as follows: each key component of the distribution box is regarded as a node, and a unique label is assigned to each node; the safety performance parameters of each node are collected, including the rated current, rated voltage, actual current, actual voltage, temperature, number of shutdowns and safety protection level of the node; based on a preset safety performance coefficient calculation formula, the safety performance coefficient of the node is calculated in combination with the safety capacity and actual load of the node.

[0014] The method and system for evaluating power grid technical transformation projects based on data clustering include a data acquisition module, a data clustering module, an impact data acquisition module, a technical transformation project evaluation model construction module and a processing module, and there are connections between the modules; the data acquisition module is used to collect multi-dimensional data of power grid technical transformation projects and pre-process the data; the data clustering module is used to cluster the pre-processed data based on a clustering algorithm to identify project data with the same characteristics; the impact data acquisition module is used to obtain the technical transformation evaluation impact data of the distribution box in the test area in the project data with the same characteristics; the technical transformation project evaluation model construction module is used to construct a technical transformation project evaluation model based on the machine learning algorithm according to the technical transformation evaluation impact data; the processing module is used to obtain the technical transformation project evaluation coefficient according to the technical transformation project evaluation model, generate a descending sort table according to the technical transformation project evaluation coefficient, and give priority to the power grid project with the largest technical transformation project evaluation coefficient.

[0015] The technical effects and advantages of the power grid technical transformation project evaluation method and system based on data clustering of the present invention are as follows: 1. The present invention collects multi-dimensional data of power grid technical transformation projects and performs preprocessing, uses clustering algorithms to identify project data with the same characteristics, and then combines machine learning algorithms to build a technical transformation project evaluation model, which can achieve efficient and accurate evaluation of technical transformation projects. This method not only takes into account the multiple dimensions of the project, but also deeply explores the inherent connections and laws between project data through data clustering and machine learning techniques, thereby improving the accuracy and reliability of the evaluation. According to the descending sort table generated by the technical transformation project evaluation coefficient, the priority of each project can be intuitively displayed, providing strong data support for decision makers.

[0016] 2. The present invention fully considers the impact data of the technical transformation evaluation of the distribution box, including equipment status coefficient, load fluctuation coefficient, safety performance coefficient and other aspects. The acquisition of these coefficients not only relies on real-time monitoring data, but also combines historical data and system operation experience to ensure the comprehensiveness and accuracy of the evaluation. Through the comprehensive consideration of these coefficients, the impact of the technical transformation project on the operation of the power grid can be more comprehensively evaluated, providing a scientific basis for the implementation of the project. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 The present invention is a schematic diagram of a power grid technical transformation project evaluation method and system structure based on data clustering.

[0018] Figure 2 The present invention is a schematic diagram of the structure of a power grid technical transformation project evaluation system based on data clustering. DETAILED DESCRIPTION

[0019] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0020] Embodiment 1, Figure 1 The present invention provides a power grid technical transformation project evaluation method based on data clustering, which includes the following steps: S1, collects multi-dimensional data of power grid technical transformation projects and pre-processes the data; The multi-dimensional data include but are not limited to project cost, implementation period, expected benefits, technical complexity, etc.; The multi-dimensional data of the power grid technical transformation project is collected and the data is pre-processed, specifically: According to the evaluation objectives of the power grid technical transformation project, multi-dimensional data is collected and integrated from the internal and external databases of the power grid company using data crawling tools to obtain a unified data set; Standardize the data set and convert the data into a uniform standard deviation and mean; Based on the standardized data set, calculate the covariance matrix of each element in the data set; Perform eigenvalue decomposition on the covariance matrix to obtain the eigenvalues ​​and corresponding eigenvectors.

[0021] According to the size of the eigenvalue, select the eigenvectors corresponding to the first k largest eigenvalues; The original data is converted into a new space consisting of the eigenvectors corresponding to the first k largest eigenvalues ​​to obtain the reduced-dimensional data.

[0022] S2, clustering the preprocessed data based on the clustering algorithm to identify project data with the same characteristics; Based on the clustering algorithm, the pre-processed data is clustered to identify project data with the same characteristics, specifically: Use the preprocessed data as input to the clustering algorithm; Calculate the sum of squared errors under different k values ​​to get the optimal number of clusters k; Randomly select K data points as the initial centroids, calculate the distance between each data point and each centroid, and assign it to the cluster where the nearest centroid is located; Update the centroid of each cluster to the mean of all data points in the cluster; Repeat the above calculation steps until the position of the centroid reaches the preset number of iterations; Calculate the closeness of data points within a cluster (such as average distance, standard deviation, etc.) and the separation between clusters (such as distance matrix, distance between cluster centers, etc.) to evaluate the clustering effect; Observe the data points in each cluster and intuitively display the clustering results and feature distribution based on scatter plots; According to the clustering results, extract the common features of each cluster; Item data with the same characteristics are grouped into the same cluster.

[0023] The evaluation method of power grid technical transformation projects based on clustering algorithms, especially the use of pre-processed data for clustering to identify project data with the same characteristics, has brought many benefits to the evaluation of power grid technical transformation projects: Automated processing: Through clustering algorithms, a large amount of power grid technical transformation project data can be automatically processed and analyzed, avoiding the tedious and time-consuming traditional manual evaluation.

[0024] Accurate classification: Clustering algorithms can accurately classify data points based on their characteristics, and classify project data with the same or similar characteristics into the same cluster, thereby improving the accuracy of the assessment.

[0025] Resource optimization: By clustering the technical transformation projects, we can clearly identify different categories of projects, so as to make targeted resource allocation according to the characteristics and needs of the projects and avoid waste of resources.

[0026] Decision support: Clustering results can provide strong support for decision-making on power grid technical transformation projects. For example, based on the clustering results, it can be determined which projects need to be prioritized for technical transformation, which projects can be postponed or merged, etc.

[0027] Technological innovation: Cluster analysis can reveal the commonalities and differences between different projects, thus providing inspiration and direction for technological innovation. For example, cluster analysis can be used to discover the applicability of certain technologies or methods in different projects, thereby promoting technological innovation and improvement.

[0028] Continuous improvement: Through continuous tracking and analysis of clustering results, problems and deficiencies in power grid technical transformation projects can be discovered in a timely manner, so that corresponding measures can be taken to improve and optimize them.

[0029] Operation efficiency: By clustering the technical transformation projects, we can more accurately evaluate the implementation effect of the projects and their impact on the grid operation efficiency, thereby formulating a more scientific operation and maintenance strategy.

[0030] Safety improvement: Cluster analysis can reveal potential safety hazards and weak links in power grid technical transformation projects, so that corresponding measures can be taken to strengthen and improve them and improve the safety of the power grid.

[0031] Visualization: Through visualization tools such as scatter plots, clustering results and feature distribution can be intuitively displayed, making the evaluation results more intuitive and easy to understand.

[0032] Intuitive understanding: Visual display helps relevant personnel to understand the clustering results and project characteristics more intuitively, so as to more accurately grasp the overall situation and development trend of power grid technical transformation projects.

[0033] S3, obtaining the technical transformation assessment impact data of the distribution boxes in the test area in the project data with the same characteristics; The technical transformation assessment impact data of the distribution box, including equipment status coefficient, load fluctuation coefficient and safety performance coefficient; The equipment status coefficient is a quantitative indicator for evaluating the performance and operating status of equipment. It comprehensively evaluates the quality and reliability of equipment during use by analyzing and comparing equipment operating data. The equipment status coefficient can reflect key indicators such as equipment integrity rate, availability coefficient, failure rate and repair rate, helping managers to understand the operating status of equipment in a timely manner, so as to take corresponding maintenance and management measures to ensure the long-term stable operation of equipment. This indicator is of great significance in equipment management, maintenance and production operations.

[0034] The specific method for obtaining the device status coefficient is as follows: Operate the distribution box to be tested under preset conditions, record its operating parameters at different time points, such as temperature, pressure, current, voltage, etc., conduct multiple experiments, and obtain multiple sets of operating data; Preprocess the collected multiple sets of operation data to remove outliers and noise; Calculate the average value and standard deviation of various operating parameters of the distribution box during operation; The status index of the distribution box is obtained based on the average value and standard deviation of various operating parameters combined with the design parameters and historical maintenance records of the distribution box; The equipment status coefficient is calculated based on a preset formula according to the status index of the distribution box and the response time.

[0035] The calculation formula of the state index of the distribution box is as follows:

[0036] The calculation formula of the equipment status coefficient is as follows:

[0037] In the formula, is the state coefficient, is the measured value of the i-th operating parameter, is the average value of the ith operating parameter, is the standard deviation of the ith operating parameter, is the weight of the i-th operating parameter, n is the total number of operating parameters, is the equipment status coefficient, is the device response time, is the maximum response time of the device, and k is the response time adjustment coefficient.

[0038] The load fluctuation coefficient is an indicator used to measure the degree of load change in a power system or a specific user. It usually reflects the ratio of the degree of dispersion of the load to its degree of concentration. The larger the load fluctuation coefficient, the greater the volatility of the load, and the stability of the power system may be affected, which will have an adverse impact on the power supply and consumption. On the contrary, the smaller the load fluctuation coefficient, the more stable the load and the more stable and reliable the operation of the power system. This indicator has important application value in power system planning, operation scheduling, load forecasting, and equipment status assessment. It helps power workers better understand the load characteristics of the power system, so as to take corresponding measures to optimize power supply and ensure the safe and stable operation of the power grid.

[0039] The specific steps for obtaining the load fluctuation coefficient are as follows: Real-time monitoring and recording of the load value and its changes of each load data point within a certain time window; According to the recorded data, the load difference between each load data point and its previous data point is calculated to construct a load difference sequence; Record the range of changes in the load difference sequence over a period of time and calculate the volatility of the load difference; Conduct cumulative analysis on load data to calculate the cumulative effect of load fluctuations over a period of time; Based on historical load data and system operation experience, define the threshold for load fluctuations to trigger unstable states; When the cumulative effect of load fluctuation exceeds the threshold, the current time period is marked as a potential unstable period; Perform time series segmentation on historical load data to obtain several time series data segments, and determine whether they are potential unstable periods, as historical load fluctuation events; Mapping time series data with historical load fluctuation events based on machine learning algorithms to generate a probability model of load fluctuation triggering unstable states; The load fluctuation coefficient is calculated by combining the fluctuation rate of the load difference and the probability model of load fluctuation triggering an unstable state.

[0040] The specific calculation formula of the load fluctuation coefficient is as follows:

[0041] In the formula, is the load fluctuation coefficient, n is the number of data points in the time window, is the standard deviation of the load data, is the average value of load data, x is the sum of load differences over a period of time, g is the critical value of load fluctuation triggering an unstable state, and P is the probability of load fluctuation triggering an unstable state.

[0042] The safety performance factor is a comprehensive evaluation index used to measure the safe operation capability and reliability of the distribution box in the power system. It takes into account the structural design, material selection, reliability of electrical components, load capacity, adaptability to environmental conditions, and fault prevention measures of the distribution box. The higher the safety performance factor, the higher the safety and stability of the distribution cabinet when facing various power loads and environmental changes, and effectively reduces the occurrence of faults and safety accidents. This factor is of great significance to ensure the stable operation of the power system and protect the safety of personnel and property. It is a key factor that must be considered in the design and selection of the distribution box.

[0043] The specific steps for obtaining the safety performance coefficient are as follows: Treat each key component of the distribution box as a node and assign a unique label to each node; Collect the safety performance parameters of each node, including the node's rated current, rated voltage, actual current, actual voltage, temperature, number of outages, and the node's safety protection level; Based on the preset safety performance coefficient calculation formula, combined with the node's safety capacity and actual load, the node's safety performance coefficient is calculated.

[0044] The calculation formula of the safety performance factor is as follows:

[0045] In the formula, is the safety performance coefficient, n is the number of nodes, is the rated current of the ith node, is the actual current of the ith node, is the rated voltage of the ith node, is the actual voltage of the ith node, is the temperature at the i-th node, is the ambient temperature, is the maximum temperature allowed for the node, is the safety protection level factor of the ith node, C is the safety capacity of the node, P is the actual load of the node, and L is the number of outages of the distribution box.

[0046] S4, constructing a technological transformation project evaluation model based on the technological transformation evaluation impact data and machine learning algorithm; The technical transformation project evaluation model is specifically as follows:

[0047] In the formula, is the technical transformation project evaluation coefficient, is the equipment status coefficient, is the load fluctuation coefficient, is the safety performance factor, is the preset equipment status coefficient weight factor, is the preset load fluctuation coefficient weight factor, It is the preset safety performance coefficient weight factor.

[0048] It should be noted that the larger the project's technical transformation project evaluation coefficient is, the more necessary it is to carry out technical transformation; conversely, the smaller the project's technical transformation project evaluation coefficient is, the less necessary it is to carry out technical transformation.

[0049] S5, obtaining a technical transformation project evaluation coefficient according to a technical transformation project evaluation model, generating a descending sort table according to the technical transformation project evaluation coefficient, and giving priority to processing the power grid technical transformation project with the largest technical transformation project evaluation coefficient.

[0050] The evaluation coefficient of the technical transformation project is obtained according to the technical transformation project evaluation model, and a descending sorting table is generated according to the evaluation coefficient of the technical transformation project, specifically: Organize the relevant information of all technical transformation projects, including project names, characteristic vectors, and evaluation coefficients of model outputs; the characteristic vectors include equipment status coefficients, load fluctuation coefficients, and safety performance coefficients; Organize the relevant information of all technical transformation projects into a table, including the following columns: project name, equipment status coefficient, load fluctuation coefficient, safety performance coefficient, and technical transformation evaluation coefficient; Sort the technological transformation projects in the table in descending order according to their evaluation coefficients, so that the project with the highest evaluation coefficient will be ranked first, and the project with the lowest evaluation coefficient will be ranked last; The sorted tables are exported as Excel and CSV files, and priority is given to the power grid technical transformation projects with the largest technical transformation project evaluation coefficients.

[0051] The power grid technical transformation project evaluation method based on data clustering has many benefits, especially in the process of generating technical transformation project evaluation coefficients and generating descending sorting tables based on them. The following is a detailed explanation of the benefits of this evaluation method: Data-driven: The evaluation method based on data clustering relies on a large amount of historical data and real-time data, which can fully reflect the actual situation and effect of the power grid technical transformation project. Through cluster analysis, similar projects can be grouped together, so as to more accurately evaluate the performance and benefits of each project.

[0052] Objective and fair: This method avoids subjective interference from human factors, and the evaluation results are more objective and fair. The clustering algorithm automatically classifies data based on similarities and differences, and is not affected by personal preferences or experience.

[0053] Priority sorting: By generating the evaluation coefficients of technical transformation projects and sorting them in descending order, it is possible to clearly see which projects have higher priority and greater potential value. This helps decision makers optimize resource allocation and invest limited funds and technical resources in the areas where they are most needed.

[0054] Improve return on investment: The evaluation method based on data clustering can identify those technological transformation projects with significant benefits and high cost-effectiveness. Prioritizing the implementation of these projects can ensure the maximum return on investment.

[0055] Fast screening: Clustering algorithms can quickly classify and screen a large number of technological transformation projects, reducing the time and cost of manual screening. This enables decision makers to make decisions faster and seize market opportunities.

[0056] Decision support: Through cluster analysis and the generation of evaluation coefficients, comprehensive project information and evaluation results can be provided to decision makers. This information can serve as an important reference for decision-making, helping decision makers make more intelligent and scientific decisions.

[0057] Discover potential problems: The evaluation method based on data clustering can discover potential problems and deficiencies in power grid technical transformation projects. By analyzing and studying the clustering results, it can be revealed which technologies or equipment have defects or need to be improved, thereby promoting technological innovation and upgrading.

[0058] Promote industry development: By continuously optimizing and improving the evaluation methods of technical transformation projects, the overall technological progress and development of the power grid industry can be promoted. This will help improve the reliability, safety and economy of the power grid and contribute to the sustainable development of society.

[0059] In summary, the power grid technical transformation project evaluation method based on data clustering has significant benefits in improving evaluation accuracy, optimizing resource allocation, improving decision-making efficiency and promoting technological innovation. This evaluation method can provide strong support and guarantee for the decision-making of power grid technical transformation projects and promote the sustainable and healthy development of the power grid industry.

[0060] Embodiment 2, Figure 2 The present invention provides a method and system for evaluating power grid technical transformation projects based on data clustering, including a data acquisition module, a data clustering module, an impact data acquisition module, a technical transformation project evaluation model building module and a processing module, and there are connections between the modules; Data acquisition module: used to collect multi-dimensional data of power grid technical transformation projects and pre-process the data; Data clustering module: used to cluster the pre-processed data based on the clustering algorithm and identify project data with the same characteristics; Impact data acquisition module: obtains the impact data of technical transformation assessment of distribution boxes in the tested area in the project data with the same characteristics; Technical transformation project evaluation model construction module: construct a technical transformation project evaluation model based on the impact data of technical transformation evaluation and machine learning algorithm; Processing module: obtain the technical transformation project evaluation coefficient according to the technical transformation project evaluation model, generate a descending sort table according to the technical transformation project evaluation coefficient, and give priority to the power grid technical transformation project with the largest technical transformation project evaluation coefficient.

[0061] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0062] The above embodiments may be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented by software, the above embodiments may be implemented in whole or in part in the form of a computer program product.

[0063] Those of ordinary skill in the art will appreciate that the modules and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0064] In addition, each functional module in each embodiment of the present application may be integrated into one processing module, or each module may exist physically separately, or two or more modules may be integrated into one module.

[0065] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

[0066] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A power grid technical transformation project evaluation method based on data clustering, characterized in that: include: Collect multi-dimensional data of power grid technical transformation projects and pre-process the data; Based on the clustering algorithm, the preprocessed data is clustered to identify project data with the same characteristics; Obtain the technical transformation assessment impact data of the distribution boxes in the test area in the project data with the same characteristics; Construct a technology transformation project evaluation model based on the technology transformation evaluation impact data and machine learning algorithm; The technical transformation project evaluation coefficient is obtained according to the technical transformation project evaluation model, a descending sorting table is generated according to the technical transformation project evaluation coefficient, and the power grid technical transformation project with the largest technical transformation project evaluation coefficient is processed first.

2. The power grid technical transformation project evaluation method based on data clustering according to claim 1 is characterized in that: The multi-dimensional data of the power grid technical transformation project is collected and the data is pre-processed, specifically: According to the evaluation objectives of the power grid technical transformation project, multi-dimensional data is collected and integrated from the internal and external databases of the power grid company using data crawling tools to obtain a unified data set; Standardize the data set and convert the data into a uniform standard deviation and mean; Based on the standardized data set, calculate the covariance matrix of each element in the data set; Perform eigenvalue decomposition on the covariance matrix to obtain the eigenvalues ​​and corresponding eigenvectors; According to the size of the eigenvalue, select the eigenvectors corresponding to the first k largest eigenvalues; The original data is converted into a new space consisting of the eigenvectors corresponding to the first k largest eigenvalues ​​to obtain the reduced-dimensional data.

3. The power grid technical transformation project evaluation method based on data clustering according to claim 2 is characterized in that: Based on the clustering algorithm, the pre-processed data is clustered to identify project data with the same characteristics, specifically: Use the preprocessed data as input to the clustering algorithm; Calculate the sum of squared errors under different k values ​​to get the optimal number of clusters k; Randomly select K data points as the initial centroids, calculate the distance between each data point and each centroid, and assign it to the cluster where the nearest centroid is located; Update the centroid of each cluster to the mean of all data points in the cluster; Repeat the above calculation steps until the position of the centroid reaches the preset number of iterations; Calculate the compactness of data points within a cluster and the separation between clusters to evaluate the clustering effect; Observe the data points in each cluster and intuitively display the clustering results and feature distribution based on scatter plots; According to the clustering results, extract the common features of each cluster; Item data with the same characteristics are grouped into the same cluster.

4. The power grid technical transformation project evaluation method based on data clustering according to claim 3 is characterized in that: The technical transformation assessment impact data of the distribution box, including equipment status coefficient, load fluctuation coefficient and safety performance coefficient; The specific method for obtaining the device status coefficient is as follows: Operate the distribution box to be tested under preset conditions, record its operating parameters at different time points, including temperature, pressure, current, and voltage, conduct multiple experiments, and obtain multiple sets of operating data; Preprocess the collected multiple sets of operation data to remove outliers and noise; Calculate the average value and standard deviation of various operating parameters of the distribution box during operation; The status index of the distribution box is obtained based on the average value and standard deviation of various operating parameters combined with the design parameters and historical maintenance records of the distribution box; The equipment status coefficient is calculated based on a preset formula according to the status index of the distribution box and the response time.

5. The power grid technical transformation project evaluation method based on data clustering according to claim 4 is characterized in that: The specific steps for obtaining the load fluctuation coefficient are as follows: Real-time monitoring and recording of the load value and its changes of each load data point within a certain time window; According to the recorded data, the load difference between each load data point and its previous data point is calculated to construct a load difference sequence; Record the range of changes in the load difference sequence over a period of time and calculate the volatility of the load difference; Conduct cumulative analysis on load data to calculate the cumulative effect of load fluctuations over a period of time; Based on historical load data and system operation experience, define the threshold for load fluctuations to trigger unstable states; When the cumulative effect of load fluctuation exceeds the threshold, the current time period is marked as a potential unstable period; Perform time series segmentation on historical load data to obtain several time series data segments, and determine whether they are potential unstable periods, as historical load fluctuation events; Mapping time series data with historical load fluctuation events based on machine learning algorithms to generate a probability model of load fluctuation triggering unstable states; The load fluctuation coefficient is calculated by combining the fluctuation rate of the load difference and the probability model of load fluctuation triggering an unstable state.

6. The power grid technical transformation project evaluation method based on data clustering according to claim 5 is characterized in that: The specific steps for obtaining the safety performance coefficient are as follows: Treat each key component of the distribution box as a node and assign a unique label to each node; Collect the safety performance parameters of each node, including the node's rated current, rated voltage, actual current, actual voltage, temperature, number of outages, and the node's safety protection level; Based on the preset safety performance coefficient calculation formula, combined with the node's safety capacity and actual load, the node's safety performance coefficient is calculated.

7. The method for evaluating power grid technical transformation projects based on data clustering according to claim 6 is characterized in that: The calculation formula of the state index of the distribution box is as follows: The calculation formula of the equipment status coefficient is as follows: In the formula, is the state coefficient, is the measured value of the i-th operating parameter, is the average value of the ith operating parameter, is the standard deviation of the ith operating parameter, is the weight of the i-th operating parameter, n is the total number of operating parameters, is the equipment status coefficient, is the device response time, is the maximum response time of the device, and k is the response time adjustment coefficient.

8. The power grid technical transformation project evaluation method based on data clustering according to claim 7 is characterized in that: The calculation formula of the safety performance factor is as follows: In the formula, is the safety performance coefficient, n is the number of nodes, is the rated current of the ith node, is the actual current of the ith node, is the rated voltage of the ith node, is the actual voltage of the ith node, is the temperature at the i-th node, is the ambient temperature, is the maximum temperature allowed for the node, is the safety protection level factor of the ith node, C is the safety capacity of the node, P is the actual load of the node, and L is the number of outages of the distribution box.

9. The power grid technical transformation project evaluation method based on data clustering according to claim 8 is characterized in that: The technical transformation project evaluation model is specifically as follows: In the formula, is the technical transformation project evaluation coefficient, is the equipment status coefficient, is the load fluctuation coefficient, is the safety performance factor, is the preset equipment status coefficient weight factor, is the preset load fluctuation coefficient weight factor, It is the preset safety performance coefficient weight factor.

10. A system using the power grid technical transformation project evaluation method based on data clustering as claimed in any one of claims 1 to 9, characterized in that: It includes data collection module, data clustering module, impact data acquisition module, technical transformation project evaluation model building module and processing module, and there are connections between the modules; Data acquisition module: used to collect multi-dimensional data of power grid technical transformation projects and pre-process the data; Data clustering module: used to cluster the pre-processed data based on the clustering algorithm and identify project data with the same characteristics; Impact data acquisition module: obtains the impact data of technical transformation assessment of distribution boxes in the tested area in the project data with the same characteristics; Technical transformation project evaluation model construction module: construct a technical transformation project evaluation model based on the impact data of technical transformation evaluation and machine learning algorithm; Processing module: obtain the technical transformation project evaluation coefficient according to the technical transformation project evaluation model, generate a descending sort table according to the technical transformation project evaluation coefficient, and give priority to the power grid technical transformation project with the largest technical transformation project evaluation coefficient.

Images