Power distribution network operation reliability evaluation method, system, equipment, medium and product

Abstract

The invention relates to the technical field of power distribution networks, and discloses a power distribution network operation reliability evaluation method, system and device, a medium and a product. And inputting the operation data and the environment change types corresponding to the operation data into a pre-trained operation state prediction model to obtain operation state prediction results of the power distribution network under various environment change types, and determining the fault probability of the power distribution network by using the operation state prediction results. And the operation reliability of the power distribution network is evaluated according to the fault probability of the power distribution network, so that the objectivity and the accuracy of an evaluation result are improved. Compared with a traditional reliability evaluation technology depending on human experience, the method can more accurately predict the operation states of the power distribution network under different environment change types, thereby more accurately evaluating the operation reliability of the power distribution network.

Description

Technical Field

[0001] The present invention relates to the technical field of distribution networks, and particularly to a method, system, device, medium and product for evaluating the operation reliability of a distribution network. Background Art

[0002] The reliable operation of a power system is a complex issue, which is affected by many factors such as equipment failures, natural disasters, human operation errors, etc. Therefore, the power system reliability assessment technology has become an important guarantee for the safe operation of the power system. The distribution network technology is the trend of future power grid development, which realizes the intelligent management and optimized operation of the power grid through digital technology. The model-driven system optimization technology is to construct a mathematical model to predict and evaluate the operation state of the system in order to achieve the optimal operation of the system.

[0003] The traditional power system reliability evaluation technology mainly collects the data of the power grid operation for a period of time recently, and then experts conduct reliability evaluation according to experience. However, this method relies too much on human experience, resulting in overly subjective evaluation results and low accuracy. Summary of the Invention

[0004] In view of this, the present invention provides a method, system, device, medium and product for evaluating the operation reliability of a distribution network, which solves the technical problem that the traditional power system reliability evaluation technology relies too much on human experience, resulting in overly subjective evaluation results and low accuracy.

[0005] The first aspect of the present invention provides a method for evaluating the operation reliability of a distribution network, including:

[0006] Collecting the operation data of the distribution network under various types of environmental changes;

[0007] Inputting the operation data and the corresponding types of environmental changes into a pre-trained operation state prediction model, and based on the operation state prediction model, outputting the operation state prediction results of the distribution network under various types of environmental changes;

[0008] Determining the probability of the distribution network having a fault according to the operation state prediction results, and determining the operation reliability evaluation result of the distribution network according to the probability of the distribution network having a fault.

[0009] Preferably, the training process of the operation state prediction model includes:

[0010] Obtaining the historical operation data and historical operation states of the distribution network under various types of environmental changes;

[0011] Constructing a training data set according to the historical operation data and the historical operation states;

[0012] Train a machine learning network model with the training data set to obtain an initial operating state prediction model;

[0013] Optimize the network parameters of the initial operating state prediction model according to the predicted operating state value and the true operating state value output by the initial operating state prediction model to obtain an operating state prediction model.

[0014] Preferably, the optimizing the network parameters of the initial operating state prediction model according to the predicted operating state value and the true operating state value output by the initial operating state prediction model to obtain an operating state prediction model includes:

[0015] Calculate the gradient of the loss function according to the predicted operating state value and the true operating state value, where the gradient represents the gap between the predicted operating state value and the true operating state value;

[0016] Use the gradient as the target value to train a decision tree to obtain a trained decision tree;

[0017] Use the trained decision tree to update the initial operating state prediction model, and use the updated initial operating state prediction model to perform iterative training on the training data set;

[0018] After the number of iterations reaches the termination number, the iteration terminates, and an operating state prediction model that has been trained is obtained.

[0019] Preferably, the objective function corresponding to the target value is:

[0020]

[0021] In the formula, is the objective function, is the labeling result of the i-th sample data in the t-th iteration and the predicted operating state The loss between them, n represents the total number of sample data, is the regularization term.

[0022] Preferably, the determining the probability of a failure occurring in the distribution network according to the predicted operating state result, and determining the operating reliability evaluation result of the distribution network according to the probability of a failure occurring in the distribution network includes:

[0023] According to the predicted operating state result, the probability of a failure occurring in each subsystem in the distribution network;

[0024] Determine the operating reliability evaluation result of the distribution network according to the weighted result of the probabilities of failures occurring in each subsystem.

[0025] Preferably, determining the operation reliability evaluation result of the distribution network according to the weighted result of the failure probabilities of the subsystems includes:

[0026] Determining the data standard deviation of each subsystem according to the historical operation data respectively corresponding to each subsystem;

[0027] Determining the conflict degree of each subsystem according to the correlation coefficient between any two of the subsystems;

[0028] Determining the weight corresponding to each subsystem according to the data standard deviation and the conflict degree of each subsystem;

[0029] Performing weighted calculation on the failure probabilities of the subsystems according to the weights respectively corresponding to the subsystems to obtain the operation reliability evaluation result of the distribution network.

[0030] In a second aspect, the present invention also provides a distribution network operation reliability evaluation system, including:

[0031] A data acquisition module for acquiring the operation data of the distribution network under various environmental change types;

[0032] A state prediction module for inputting the operation data and the environmental change type corresponding to the operation data into a pre-trained operation state prediction model, and outputting the operation state prediction result of the distribution network under various environmental change types based on the operation state prediction model;

[0033] A reliability evaluation module for determining the failure probability of the distribution network according to the operation state prediction result, and determining the operation reliability evaluation result of the distribution network according to the failure probability of the distribution network.

[0034] In a third aspect, the present invention also provides an electronic device, which includes a memory and a processor. A computer program is stored in the memory, and when the computer program is executed by the processor, the processor executes the steps of the distribution network operation reliability evaluation method as described in the first aspect.

[0035] In a fourth aspect, the present invention also provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed, the steps of the distribution network operation reliability evaluation method as described in the first aspect are implemented.

[0036] In a fifth aspect, the present invention also provides a computer program product, which includes a computer program stored on a non-transitory computer-readable storage medium. The computer program includes program instructions. When the program instructions are executed by a computer, the computer is caused to execute the steps of the method for evaluating the operation reliability of a distribution network as described in the first aspect.

[0037] As can be seen from the above technical solutions, the present invention collects the operation data of the distribution network under various types of environmental changes, and inputs the operation data and the corresponding types of environmental changes into a pre-trained operation state prediction model to obtain the operation state prediction results of the distribution network under various types of environmental changes. By using the operation state prediction results, the probability of the distribution network having a fault is determined, and the operation reliability of the distribution network is evaluated based on the probability of the distribution network having a fault, thereby improving the objectivity and accuracy of the evaluation results. Compared with the traditional reliability evaluation technology that relies on human experience, the embodiments of the present application can more accurately predict the operation state of the distribution network under different types of environmental changes, and thus more accurately evaluate the operation reliability of the distribution network. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0039] Figure 1 The application environment of a method for evaluating the operation reliability of a distribution network provided by an embodiment of the present invention;

[0040] Figure 2 The flowchart of a method for evaluating the operation reliability of a distribution network provided by an embodiment of the present invention;

[0041] Figure 3 The structural schematic diagram of a system for evaluating the operation reliability of a distribution network provided by an embodiment of the present invention;

[0042] Figure 4 The structural schematic diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0043] To enable those skilled in the art to better understand the solution of the present invention, the following will clearly and completely describe the technical solution in the embodiments of the present invention in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0044] The method for evaluating the operation reliability of a distribution network provided by the embodiments of the present application can be applied to an application environment as Figure 1 shown. Among them, the terminal 101 communicates with the server 102 through a network. The data storage system can store the data that the server 102 needs to process. The data storage system can be integrated on the server 102, or can be placed in the cloud or other network servers. The server 102 collects the operation data of the distribution network under various types of environmental changes; inputs the operation data and the corresponding environmental change types of the operation data into a pre-trained operation state prediction model, and based on the operation state prediction model, outputs the operation state prediction results of the distribution network under various types of environmental changes; determines the probability of the distribution network failing according to the operation state prediction results, and determines the operation reliability evaluation results of the distribution network according to the probability of the distribution network failing.

[0045] The terminal 101 can be, but is not limited to, various personal computers, laptop computers, smart phones, tablet computers, etc.

[0046] The server 102 can be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server providing cloud computing services.

[0047] As Figure 2 shown, the embodiments of the present application provide a method for evaluating the operation reliability of a distribution network. Taking the application of this method to the Figure 1 terminal 101 or server 102 in it as an example, the following steps S1 to step S3 are included. Among them:

[0048] Step S1: Collect the operation data of the distribution network under various types of environmental changes.

[0049] Among them, the types of environmental changes include normal weather and extreme weather. Among them, extreme weather includes heavy rain, lightning, strong wind, high temperature, low temperature, hail, haze, etc. By collecting the operation data of the distribution network under various types of environmental changes, the operation data of the distribution network under different environments can be comprehensively obtained. The operation data includes key parameters such as voltage, current, power factor, and frequency.

[0050] Exemplarily, monitoring devices (such as sensors) in the smart grid are used to monitor various aspects of the power system in real time, including data such as the temperature, vibration, and current of power equipment; these real-time monitored data are transmitted to the data center or cloud in real time through various communication means (such as high-speed networks, wireless communications, etc.) to ensure the real-time and accuracy of the data. Data preprocessing and quality control can also be performed on the data in the data center or cloud, including data cleaning, outlier processing, data verification, etc., to improve the reliability and accuracy of the data.

[0051] Then, record the acquisition time of real-time operation data: During the data acquisition process, record the acquisition time of each real-time operation data, which can usually be achieved through the timestamp method. Exemplarily, the timestamp should be accurate to the second or millisecond level to ensure that the exact time of data acquisition can be accurately recorded. In this step, considering the seasonal changes, periodicity of extreme weather, etc. between different times and weather, therefore, when collecting real-time operation data, the acquisition time of the data and real-time weather data should also be considered.

[0052] At the same time, a reliable meteorological data platform or API interface can also be selected, such as the Xihe Energy Big Data Platform, etc.; select the geographical location on the meteorological data platform, select the same or similar geographical location as the distribution network data acquisition point; confirm the data source, select a reliable meteorological data source to ensure the accuracy of weather data; input the time period of data acquisition, that is, the acquisition time of the real-time operation data to be queried, and select the corresponding start and end times; select the required meteorological data, such as temperature, humidity, wind speed, etc., and download or export the data in the required format.

[0053] Step S2: Input the operation data and the corresponding environmental change types of the operation data into a pre-trained operation state prediction model, and based on the operation state prediction model, output the operation state prediction results of the distribution network under various environmental change types.

[0054] Among them, the operation state prediction model is trained with historical sample data, where the historical sample data includes environmental change types, the operation data corresponding to the environmental change types, and the operation state of the distribution network, where the operation state includes the fault state of the distribution network and the fault states of each subsystem within the distribution network.

[0055] Among them, the operation state prediction results can include various possible weather scenarios, and the stable or fault states of the digital power grid operation under various weather scenarios, so as to estimate whether the distribution network can operate stably under different weather scenarios.

[0056] Step S3: Determine the probability of the distribution network failing according to the operation state prediction results, and determine the operation reliability evaluation results of the distribution network according to the probability of the distribution network failing.

[0057] It is understandable that by determining the number of occurrences of power grid failures in each weather scenario, the probability of power grid failures under each weather condition is statistically estimated, and the probabilities of power grid failures under each weather condition are weighted to obtain the operation reliability evaluation result of the distribution network.

[0058] It should be noted that in the embodiments of the present application, by collecting the operation data of the distribution network under various types of environmental changes and inputting the operation data and the corresponding types of environmental changes into a pre-trained operation state prediction model, the operation state prediction result of the distribution network under various types of environmental changes is obtained. Using the operation state prediction result, the probability of power grid failures in the distribution network is determined, and the operation reliability of the distribution network is evaluated according to the probability of power grid failures in the distribution network, thereby improving the objectivity and accuracy of the evaluation result. Compared with the traditional reliability evaluation technology that relies on human experience, the embodiments of the present application can more accurately predict the operation state of the distribution network under different types of environmental changes, and thus more accurately evaluate the operation reliability of the distribution network.

[0059] In some embodiments, the training process of the operation state prediction model includes:

[0060] Step S21: Obtain the historical operation data and historical operation state of the distribution network under various types of environmental changes;

[0061] Step S22: Construct a training data set according to the historical operation data and historical operation state;

[0062] Step S23: Train a machine learning network model through the training data set to obtain an initial operation state prediction model;

[0063] Among them, the machine learning network model can use the XGBoost (eXtreme Gradient Boosting) machine learning network. Since the XGBoost model has advantages such as high efficiency, accuracy, flexibility, interpretability, and robustness, in this step, iterative training is performed through the XGBoost model, and the prediction efficiency and accuracy of the final operation state prediction model are higher.

[0064] The XGBoost model provides various tree learning algorithms, such as Gradient-based TreeBoosting, Label-wise Prediction, and Leaf-wise Prediction. For example, the process of iteratively training the XGBoost model based on the predicted operating states and labeled results of the digital power grid in various weather scenarios to obtain a trained operating state prediction model includes:

[0065] Step S24: Optimize the network parameters of the initial operating state prediction model according to the predicted operating state value and the true operating state value output by the initial operating state prediction model to obtain the operating state prediction model.

[0066] In some embodiments, optimizing the network parameters of the initial operating state prediction model according to the predicted operating state value and the true operating state value output by the initial operating state prediction model to obtain the operating state prediction model includes:

[0067] Step S241: Calculate the gradient of the loss function according to the predicted operating state value and the true operating state value, where the gradient represents the gap between the predicted operating state value and the true operating state value;

[0068] Step S242: Use the gradient as the target value to train the decision tree to obtain the trained decision tree;

[0069] In this example, the process of the XGBoost model fitting the training data by constructing multiple decision trees and using the gradient boosting algorithm to optimize the model's prediction performance can be as follows: At the beginning, XGBoost initializes a weak prediction model (usually a constant or the mean) as the basis for prediction. For each sample in the training dataset, calculate the gradient (or negative gradient) of the loss function at the current predicted value. This gradient reflects the gap between the model's predicted value and the actual value, and how to reduce this gap by adjusting the model parameters. Use the calculated gradient as the target value (label) to train a new decision tree. This decision tree attempts to fit the residual of the current model (i.e., the difference between the predicted value and the actual value), so as to reduce the prediction error of the entire model on the training data. Add the newly trained decision tree to the model and update the model's predicted value. This is usually achieved by adding the predicted value of the new tree to the predicted value of the previous model. Repeat the above steps (calculate the gradient, construct a new decision tree, update the model). Each iteration will generate a new decision tree and add it to the model. This process continuously reduces the residual of the model on the training data, thereby gradually improving the model's prediction performance. The iteration process will continue until a certain stopping condition is met. Common stopping conditions include reaching the preset maximum number of iterations, the model's performance on the validation set reaching the preset threshold, or the model's performance improvement on the training set being no longer significant, etc. When the iteration process stops, XGBoost outputs a strong learner composed of multiple decision trees, that is, the final prediction model. This model can be used to predict new, unseen data. In this process, XGBoost adopts the idea of the gradient boosting algorithm, and gradually reduces the model's prediction error by continuously adding new decision trees, so as to obtain a prediction model with better performance. At the same time, XGBoost can also adopt some optimization techniques (such as column subsampling, row subsampling, regularization, etc.) to prevent overfitting and improve the model's generalization ability.

[0070] The objective function in the above XGBoost model includes a loss function and a regularization term. The regularization term is used to control the complexity of the XGBoost model; by designing an objective function including a loss function and a regularization term, it aims to minimize the prediction error while preventing overfitting.

[0071] The XGBoost model supports custom loss functions and evaluation metrics. Among them, the objective function corresponding to the target value is:

[0072]

[0073] In the formula, is the objective function, is the labeled result of the i-th sample data in the t-th iteration and the prediction running state The loss between, n represents the total number of sample data, is the regularization term.

[0074] Regularization term is:

[0075]

[0076] In the formula, is the coefficient of regularization on the weights, represents the number of leaf nodes of the decision tree in the t-th round of iteration, is the coefficient of regularization, is the output weight of the j-th leaf node, is the regularization strength parameter specific to each task, is the regularization term for each task separately, is a hyperparameter used to control the shared regularization strength, is the shared regularization. This objective function and the regularization term can promote knowledge transfer between tasks and improve the overall learning efficiency, and ensure the stability of the model, thereby improving the accuracy of the final prediction result.

[0077] Step S243: Update the initial operating state prediction model using the trained decision tree, and perform iterative training on the training data set using the updated initial operating state prediction model;

[0078] Step S244: After the number of iterations reaches the termination number, the iteration terminates, and the trained operating state prediction model is obtained.

[0079] It can be understood that in the traditional solution, only the recent operating state is considered, only a single operating scenario is considered, and the prediction of the power grid operating state under multiple scenarios is insufficient, while the power grid operating state under multiple scenarios is ignored, which may lead to one-sidedness and inaccuracy of the evaluation results. In the embodiments of the present invention, the impacts of multiple weather scenarios are considered, and these weather scenarios may have different impacts on the operation of the power grid. Therefore, when predicting and evaluating, the comprehensiveness and accuracy of the prediction and evaluation can be improved.

[0080] Considering that the business system, physical system, and information system may all affect the reliability of the digital power grid, therefore, in some embodiments, according to the operating state prediction result, the probability of the distribution network failing is determined, and according to the probability of the distribution network failing, the operating reliability evaluation result of the distribution network is determined, including:

[0081] Step S301: According to the operating state prediction result, according to the probability of each subsystem in the distribution network failing.

[0082] Among them, each subsystem in the distribution network includes a business system, a physical system, an information system, etc. Among them, the business system refers to the systems related to operations, dispatching, control, etc. of the distribution network; the physical system refers to the physical components such as equipment, lines, and substations in the distribution network; the information system refers to the information-based system for functions such as monitoring, control, and management. By analyzing the prediction results of the operating states of each subsystem, the probability of each subsystem failing can be determined, thereby providing more detailed data support for subsequent reliability evaluation.

[0083] Step S302: Determine the operating reliability evaluation result of the distribution network according to the weighted result of the probability of each subsystem failing.

[0084] In some embodiments, determining the operating reliability evaluation result of the distribution network according to the weighted result of the probability of each subsystem failing includes:

[0085] Step S3021: Determine the data standard deviation of each subsystem according to the historical operating data corresponding to each subsystem.

[0086] Among them, the historical operating data is statistically analyzed to calculate the mean and variance of each index during the operation of each subsystem. Then, based on the variance, the data standard deviation of each subsystem is calculated. The data standard deviation reflects the degree of dispersion of the operating data of the subsystem. The larger the standard deviation, the greater the fluctuation of the operating data of the subsystem and the greater the possible operating risks.

[0087] Exemplarily, the standard deviations of the physical system, the business system, and the information system can be implemented by using the calculation methods of sample standard deviation or population standard deviation, which are not limited here. After calculating the standard deviation, standard deviation standardization (such as Z-Score standardization) can also be performed. Standard deviation standardization can eliminate the influence of dimensions, make different features comparable, and thus improve accuracy.

[0088] Step S3022: Determine the conflict degree of each subsystem according to the correlation coefficient between any two subsystems in each subsystem. Among them, the correlation coefficient between any two subsystems is calculated, and this correlation coefficient reflects the relevance or mutual influence degree of the operating states between the two subsystems. If the correlation coefficient between two subsystems is relatively high, it means that their operating states are relatively consistent and there may be similar operating risks; if the correlation coefficient is relatively low, it means that their operating states are quite different and there may be different operating risks. Based on the correlation coefficient, the conflict degree of each subsystem can be calculated. The higher the conflict degree, the greater the difference in the operating states of the subsystem and other subsystems, and the greater the possible operating risks.

[0089] Exemplarily, determining the conflict degree of the physical system, business system, and information system according to the correlation coefficient between any two metrics among them includes:

[0090]

[0091] In the formula, is the conflict degree of the i-th metric, is the global dynamic adjustment factor, which can be a value greater than 1 to increase the importance of the conflict degree, or a value less than 1 to reduce the impact brought by this global dynamic adjustment factor. is the local adjustment factor between the i-th metric and the j-th metric, and the local adjustment factor represents the relative importance between different metric pairs. is the correlation coefficient between the i-th metric and the j-th metric. Both i and j are positive integers not greater than m, where m is the total number of metrics, and the metrics include the physical system, business system, and information system. The global dynamic adjustment factor and the local adjustment factor can be determined based on historical data analysis or the characteristics of the data itself. This formula can reflect the average absolute value of the correlation between a metric and other metrics, so as to quantify the conflict or information richness of the metric.

[0092] Step S3023: Determine the weights corresponding to each subsystem according to the data standard deviation and conflict degree of each subsystem.

[0093] Among them, by constructing a weight determination model, this model comprehensively considers two factors: data standard deviation and conflict degree. The data standard deviation reflects the dispersion degree of the operation data of the subsystem. The larger the standard deviation, the greater the fluctuation of the operation data of this subsystem, and there may be greater operation risks. Therefore, higher attention should be given in weight determination. The conflict degree reflects the difference degree of the operation states between the subsystem and other subsystems. The higher the conflict degree, the greater the possible unique operation risks of this subsystem, and it should also be considered in weight determination.

[0094] Among them, the model for determining the weights corresponding to each subsystem is expressed as:

[0095]

[0096] In the formula, is the weight of the i-th metric, and are both adjustment factors used to balance the influence of the standard deviation and conflict degree on the weight, , are the standard deviations of the i-th metric and the j-th metric respectively.

[0097] Step S3024: Calculate the weighted probabilities of failures of each subsystem according to the weights corresponding to each subsystem, and obtain the operation reliability evaluation result of the distribution network.

[0098] It should be noted that in the embodiments of the present application, by distinguishing the operation states of each subsystem in the distribution network, calculating the probabilities of their failures respectively, and then comprehensively evaluating in combination with the weights of each subsystem, the evaluation result is more accurate and comprehensive. This evaluation method not only considers the operation states of each subsystem itself, but also considers the mutual influence between subsystems, so as to more truly reflect the overall operation reliability of the distribution network. Through this evaluation system, operation and maintenance personnel can timely discover potential operation risks, take corresponding measures for prevention and repair, thereby improving the stability and safety of the distribution network, and enhancing the operation efficiency and reliability of the entire power grid.

[0099] Based on the same inventive concept, the embodiments of the present application also provide a distribution network operation reliability evaluation system for implementing the above-mentioned distribution network operation reliability evaluation method.

[0100] The implementation solutions provided by this system to solve problems are similar to those recorded in the above method. Therefore, the specific limitations in one or more embodiments of the following distribution network operation reliability evaluation system can refer to the limitations on the distribution network operation reliability evaluation method in the above text, and will not be elaborated here.

[0101] As Figure 3 shown, the embodiments of the present application provide a distribution network operation reliability evaluation system, which includes:

[0102] A data acquisition module 100 for acquiring operation data of the distribution network under various environmental change types;

[0103] A state prediction module 200 for inputting the operation data and the corresponding environmental change types of the operation data into a pre-trained operation state prediction model, and outputting an operation state prediction result of the distribution network under various environmental change types based on the operation state prediction model;

[0104] A reliability evaluation module 300 for determining the probability of failure of the distribution network according to the operation state prediction result, and determining the operation reliability evaluation result of the distribution network according to the probability of failure of the distribution network.

[0105] In some embodiments, this system further includes a model training module for:

[0106] A historical data acquisition module for acquiring historical operation data and historical operation states of the distribution network under various environmental change types;

[0107] A training data module for constructing a training data set according to historical operation data and historical operation status;

[0108] A training module for training a machine learning network model through the training data set to obtain an initial operating status prediction model;

[0109] A model optimization module for optimizing the network parameters of the initial operating status prediction model according to the predicted operating status value and the true operating status value output by the initial operating status prediction model to obtain an operating status prediction model.

[0110] In some embodiments, the model optimization module is used for:

[0111] Calculating the gradient of the loss function according to the predicted operating status value and the true operating status value, where the gradient represents the gap between the predicted operating status value and the true operating status value;

[0112] Using the gradient as the target value to train a decision tree to obtain a trained decision tree;

[0113] Updating the initial operating status prediction model using the trained decision tree, and iteratively training the training data set using the updated initial operating status prediction model;

[0114] After the number of iterations reaches the termination number, the iteration terminates, and a trained operating status prediction model is obtained.

[0115] In some embodiments, the objective function corresponding to the target value is:

[0116]

[0117] In the formula, is the objective function, is the labeling result of the i-th sample data in the t-th iteration and the predicted operating status The loss between them, n represents the total number of sample data, is the regularization term.

[0118] In some embodiments, the reliability evaluation module 300 is used for:

[0119] A probability determination module for determining the probability of failure of each subsystem in the distribution network according to the operating status prediction result;

[0120] An evaluation weighting module for determining the operating reliability evaluation result of the distribution network according to the weighted result of the probability of failure of each subsystem.

[0121] In some embodiments, the evaluation weighting module is used for:

[0122] Determine the data standard deviation of each subsystem according to the historical operation data corresponding to each subsystem;

[0123] Determine the conflict degree of each subsystem according to the correlation coefficient between any two subsystems in each subsystem;

[0124] Determine the weight corresponding to each subsystem according to the data standard deviation and conflict degree of each subsystem;

[0125] Perform weighted calculation on the probability of each subsystem failing according to the weight corresponding to each subsystem, and obtain the operation reliability evaluation result of the distribution network.

[0126] As Figure 4 shown, an embodiment of the present application provides an electronic device. The electronic device 10 includes a memory 20 and a processor 30. A computer program is stored in the memory 20. When the computer program is executed by the processor 30, the processor 30 is caused to execute the steps of the distribution network operation reliability evaluation method in the above embodiment.

[0127] An embodiment of the present application provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed, the steps of the distribution network operation reliability evaluation method in the above embodiment are implemented.

[0128] An embodiment of the present application provides a computer program product. The computer program product includes a computer program stored on a non-transitory computer-readable storage medium. The computer program includes program instructions. Wherein, when the program instructions are executed by a computer, the computer is caused to execute the steps of the distribution network operation reliability evaluation method in the above embodiment.

[0129] Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working processes of the above-described system, electronic device, computer storage medium, and computer program product can refer to the corresponding processes in the foregoing method embodiments, and will not be described herein again.

[0130] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above drawings are used to distinguish similar objects, and do not have to be used to describe a specific order or sequence. It should be understood that such used data may be interchanged under appropriate circumstances so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or device that includes a series of steps or units does not necessarily have to be limited to those clearly listed steps or units, but may include other steps or units that are not clearly listed or are inherent to these processes, methods, products, or devices.

[0131] It should be understood that although the steps in the flowcharts involved in the above-described embodiments are shown in sequence according to the indication of the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless there is a clear description in this article, there is no strict order restriction for the execution of these steps, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-described embodiments may include multiple steps or multiple stages. These steps or stages are not necessarily executed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be executed alternately or in turn with at least a part of the steps or stages in other steps or other steps.

[0132] In several embodiments provided by the present invention, it should be understood that the disclosed systems, electronic devices, computer storage media, computer program products, and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the units is only a logical function division, and there may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed couplings or direct couplings or communication connections to each other can be through some interfaces, indirect couplings or communication connections of devices or units, and can be in electrical, mechanical or other forms.

[0133] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place, or may be distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0134] In addition, the functional units in each embodiment of the present invention can be integrated into one processing unit, or each unit can exist physically alone, or two or more units can be integrated into one unit. The above-mentioned integrated units can be implemented in the form of hardware or in the form of software functional units.

[0135] When the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for executing all or part of the steps of the methods described in various embodiments of the present invention through a computer device (which can be a personal computer, a server, or a network device, etc.). The aforementioned storage medium includes: various media that can store program codes, such as USB flash drives, mobile hard disks, read-only memories (English full name: Read-Only Memory, English abbreviation: ROM), random access memories (English full name: Random Access Memory, English abbreviation: RAM), magnetic disks, or optical discs.

[0136] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements on some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of various embodiments of the present invention.

Claims

1. A method for evaluating the operation reliability of a distribution network, characterized in that, Including: Collecting the operation data of the distribution network under various environmental change types; Inputting the operation data and the corresponding environmental change types of the operation data into a pre-trained operation status prediction model, and based on the operation status prediction model, outputting the operation status prediction results of the distribution network under various environmental change types; Determining the probability of the distribution network having a fault according to the operation status prediction results, and determining the operation reliability evaluation results of the distribution network according to the probability of the distribution network having a fault.

2. The method for evaluating the operation reliability of a distribution network according to claim 1, wherein The training process of the operation status prediction model includes: Obtaining the historical operation data and historical operation status of the distribution network under various environmental change types; Constructing a training data set according to the historical operation data and the historical operation status; Training a machine learning network model through the training data set to obtain an initial operation status prediction model; Optimizing the network parameters of the initial operation status prediction model according to the operation status prediction values and the true operation status values output by the initial operation status prediction model to obtain an operation status prediction model.

3. The method for evaluating the operation reliability of a distribution network according to claim 2, characterized in that The optimizing the network parameters of the initial operation status prediction model according to the operation status prediction values and the true operation status values output by the initial operation status prediction model to obtain an operation status prediction model includes: Calculating the gradient of the loss function according to the operation status prediction values and the true operation status values, where the gradient represents the gap between the operation status prediction values and the true operation status values; Using the gradient as the target value to train a decision tree to obtain a trained decision tree; Updating the initial operation status prediction model with the trained decision tree, and iteratively training the training data set with the updated initial operation status prediction model; After the number of iterations reaches the termination number, the iteration terminates, and an operation status prediction model that has been trained is obtained.

4. The method for evaluating the operation reliability of a distribution network according to claim 3, wherein The objective function corresponding to the target value is: In the formula, is the objective function, is the labeling result of the i-th sample data in the t-th iteration and the predicted operating state The loss between them, n represents the total number of sample data, is the regularization term.

5. The method for evaluating the operation reliability of a distribution network according to claim 1, wherein The determining the probability of the distribution network having a fault according to the operation status prediction results, and determining the operation reliability evaluation results of the distribution network according to the probability of the distribution network having a fault includes: According to the operation status prediction results, the probability of each subsystem in the distribution network having a fault; Determining the operation reliability evaluation results of the distribution network according to the weighted result of the probabilities of each subsystem having a fault.

6. The method for evaluating the operation reliability of a distribution network according to claim 5, characterized in that, The determining the operation reliability evaluation results of the distribution network according to the weighted result of the probabilities of each subsystem having a fault includes: Determining the data standard deviation of each subsystem according to the historical operation data corresponding to each subsystem; Determining the conflict degree of each subsystem according to the correlation coefficient between any two subsystems in each subsystem; Determining the weight corresponding to each subsystem according to the data standard deviation and conflict degree of each subsystem; Performing weighted calculation on the probabilities of each subsystem having a fault according to the weights corresponding to each subsystem to obtain the operation reliability evaluation results of the distribution network.

7. A distribution network operation reliability evaluation system, characterized in that, Including: A data acquisition module for acquiring the operation data of the distribution network under various types of environmental changes; A status prediction module for inputting the operation data and the corresponding types of environmental changes of the operation data into a pre-trained operation status prediction model, and outputting the operation status prediction results of the distribution network under various types of environmental changes based on the operation status prediction model; A reliability evaluation module for determining the probability of the distribution network failing according to the operation status prediction results, and determining the operation reliability evaluation results of the distribution network according to the probability of the distribution network failing.

8. An electronic device, characterized in that, The electronic device includes a memory and a processor. When the computer program stored in the memory is executed by the processor, the processor executes the steps of the distribution network operation reliability evaluation method according to any one of claims 1-6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed, the steps of the distribution network operation reliability evaluation method according to any one of claims 1-6 are implemented.

10. A computer program product, characterized in that, The computer program product includes a computer program stored on a non-transitory computer-readable storage medium. The computer program includes program instructions. When the program instructions are executed by a computer, the computer executes the steps of the distribution network operation reliability evaluation method according to any one of claims 1-6.

Images