A distribution network status assessment method and system

Abstract

The present invention discloses a distribution network state assessment method and system, which relates to the technical field of distribution network state assessment. The method and system obtain multiple measurement data of the distribution network to be assessed, filter each measurement data based on a preset window length, obtain corresponding mutation measurement data and target measurement data, perform anomaly detection on each mutation measurement data based on a preset false data detection model, obtain corresponding target false measurement data, perform adaptive filtering on each target false measurement data, obtain corresponding target state values, and then perform state assessment on the distribution network to be assessed based on each target state value and target measurement data. The method overcomes the technical problem that the existing technology mainly assesses the distribution network state through a robust state estimation method, but usually regards all fluctuation data as false data, resulting in misjudgment of real fluctuation data and reduced accuracy of distribution network state assessment.

Description

Technical Field

[0001] The present invention relates to the technical field of distribution network status assessment, and in particular to a distribution network status assessment method and system. Background Art

[0002] As the proportion of renewable energy (such as wind and solar energy) in the power system continues to increase, the operating characteristics of the power system have become more complex and uncertain. The intermittent and fluctuating nature of renewable energy generation leads to increased volatility in the measured data of the power system. Furthermore, false data injection attacks (FDIA) can disguise themselves as fluctuating data and infiltrate the power system's information network, affecting the accuracy of state estimation. Traditional state assessment methods typically use the Kalman filter algorithm and its improved algorithms to evaluate the state of the distribution network. However, these methods suffer from inaccurate identification and large estimation errors when faced with the volatility of renewable energy power systems and false data injection attacks.

[0003] At present, existing technologies mainly evaluate the distribution network status through robust state estimation methods, but usually all fluctuation data are regarded as false data, resulting in misjudgment of real fluctuation data and reducing the accuracy of distribution network status assessment. Summary of the Invention

[0004] The present invention provides a distribution network state assessment method and system, which solves the technical problem that the existing technology mainly evaluates the distribution network state through a robust state estimation method, but usually regards all fluctuation data as false data, resulting in misjudgment of real fluctuation data and reducing the accuracy of distribution network state assessment.

[0005] A first aspect of the present invention provides a distribution network status assessment method, comprising:

[0006] Acquire multiple measurement data of the distribution network to be evaluated, filter each of the measurement data based on a preset window length, and obtain corresponding mutation measurement data and target measurement data;

[0007] Performing anomaly detection on each of the mutation measurement data based on a preset false data detection model to obtain corresponding target false measurement data;

[0008] Adaptively filtering each of the target false measurement data to obtain a corresponding target state value;

[0009] A state evaluation is performed according to each of the target measurement data and each of the target state values to obtain an evaluation result corresponding to the distribution network to be evaluated.

[0010] Optionally, the step of screening each of the measurement data based on a preset window length to obtain corresponding mutation measurement data and target measurement data includes:

[0011] Calculating the average covariance of the innovation sequence in each of the measurement data based on a preset window length;

[0012] Calculating the total variance of the data to be analyzed in each of the measurement data based on a preset window length;

[0013] Determine whether the total variance of each of the data to be analyzed is greater than the corresponding average covariance;

[0014] If the total variance is greater than the corresponding average covariance, the data to be analyzed is determined to be mutation measurement data;

[0015] If the total variance is less than or equal to the corresponding average covariance, the data to be analyzed is determined as target measurement data;

[0016] The expression of the innovation sequence is specifically:

[0017]

[0018] in, For the innovation sequence, k+2- The difference between the observed value and the theoretical value at time , k+3- The difference between the observed value and the theoretical value at time , is the difference between the observed value and the theoretical value at time k+1, is the window length, and k is the time value.

[0019] Optionally, the false data detection model includes a first detection model and a second detection model, and the step of performing anomaly detection on each of the mutation measurement data based on the preset false data detection model to obtain corresponding target false measurement data includes:

[0020] Acquire a plurality of training measurement data, perform data preprocessing on each of the training measurement data, and generate a measurement feature set;

[0021] Using a first feature set of the measurement feature set to input the first detection model for training to generate a first target model and a first error threshold;

[0022] Using a second feature set of the measurement feature set to input the second detection model for training to generate a second target model and a second error threshold;

[0023] Screening each of the mutation measurement data based on the first target model and the first error threshold to obtain corresponding initial false measurement data;

[0024] The initial false measurement data are screened based on the second target model and the second error threshold to obtain corresponding target false measurement data.

[0025] Optionally, the step of using the first feature set of the measurement feature set to input the first detection model for training to generate a first target model and a first error threshold includes:

[0026] Dividing the first feature set of the measurement feature set according to a preset ratio to obtain a first training feature set and a first test feature set;

[0027] Training the first detection model using the first training feature set to obtain a first target model;

[0028] Inputting the first test feature set into the first target model to obtain a plurality of test steady-state node state values;

[0029] Calculating a first relative prediction error of each of the test steady-state node state values according to the first test feature set;

[0030] Sort the first relative prediction errors from large to small according to their sizes, and select the first relative prediction errors of a preset number of values to form a first sequence;

[0031] The minimum value in the first sequence is selected as the first error threshold.

[0032] Optionally, the step of screening each of the mutation measurement data based on the first target model and the first error threshold to obtain corresponding initial false measurement data includes:

[0033] Inputting each of the mutation measurement data into the first target model to obtain a first prediction value corresponding to each of the mutation measurement data;

[0034] Calculating a first relative error between each of the first prediction values and the corresponding mutation measurement data;

[0035] Determining whether each of the first relative errors is greater than the first error threshold;

[0036] When the first relative error is greater than the first error threshold, the sudden change measurement data corresponding to the first relative error is determined as initial false measurement data.

[0037] Optionally, the step of performing adaptive filtering on each of the target false measurement data to obtain a corresponding target state value includes:

[0038] Based on a preset observation noise proportional function, constructing an observation noise proportional factor matrix corresponding to each target false measurement data;

[0039] Performing Kalman gain processing on each of the observation noise scale factor matrices based on a Kalman filter algorithm to obtain a filter coefficient matrix corresponding to each of the observation noise scale factor matrices;

[0040] The corresponding target false measurement data are filtered and corrected using each of the filter coefficient matrices to obtain a target state value corresponding to each of the target false measurement data.

[0041] Optionally, the step of performing a state evaluation according to each of the target measurement data and each of the target state values to obtain an evaluation result corresponding to the distribution network to be evaluated includes:

[0042] Determining whether each of the target state values is within a preset first threshold range;

[0043] If any of the target state values is not within the first threshold range, determining the evaluation result corresponding to the distribution network to be evaluated as abnormal operation of the distribution network;

[0044] If all of the target state values are within the first threshold range, determining whether the real-time measurement value of each of the target measurement data is within a preset second threshold range;

[0045] If any of the real-time measurement values is not within the second threshold range, determining the evaluation result corresponding to the distribution network to be evaluated as abnormal operation of the distribution network;

[0046] If all of the real-time measurement values are within the second threshold range, the evaluation result corresponding to the distribution network to be evaluated is determined to be that the distribution network is operating normally.

[0047] A second aspect of the present invention provides a distribution network status assessment system, comprising:

[0048] An acquisition module is used to obtain multiple measurement data of the distribution network to be evaluated, and filter each of the measurement data based on a preset window length to obtain corresponding mutation measurement data and target measurement data;

[0049] A detection module, configured to perform anomaly detection on each of the mutation measurement data based on a preset false data detection model to obtain corresponding target false measurement data;

[0050] A filtering module, configured to perform adaptive filtering on each of the target false measurement data to obtain a corresponding target state value;

[0051] An evaluation module is used to perform a state evaluation based on each of the target measurement data and each of the target state values to obtain an evaluation result corresponding to the distribution network to be evaluated.

[0052] A third aspect of the present invention provides an electronic device, comprising a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the steps of the distribution network status assessment method as described in any one of the above items.

[0053] A fourth aspect of the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed, implements the distribution network state assessment method as described in any one of the above items.

[0054] It can be seen from the above technical solutions that the present invention has the following advantages:

[0055] The present invention obtains multiple measurement data of the distribution network to be evaluated, screens each measurement data based on a preset window length to obtain corresponding mutation measurement data and target measurement data, performs anomaly detection on each mutation measurement data based on a preset false data detection model to obtain corresponding target false measurement data, performs adaptive filtering on each target false measurement data to obtain a corresponding target state value, and then performs state evaluation on the distribution network to be evaluated based on each target state value and target measurement data. This overcomes the technical problem that the existing technology mainly evaluates the distribution network state through a robust state estimation method, but usually regards all fluctuation data as false data, resulting in misjudgment of real fluctuation data and reducing the accuracy of distribution network state evaluation. Compared with traditional distribution network evaluation methods, the present invention screens measurement data through a preset window length and a false data detection model, thereby filtering target false measurement data with weak time correlation and spatial correlation, avoiding the correction of all fluctuation data, and improving the accuracy of distribution network state evaluation. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0057] Figure 1 A flow chart of the steps of a distribution network status assessment method provided in Example 1 of the present invention;

[0058] Figure 2 A flow chart of the steps of a distribution network status assessment method provided in the second embodiment of the present invention;

[0059] Figure 3 This is a structural block diagram of a distribution network status assessment system provided in Example 3 of the present invention;

[0060] Figure 4 This is a structural block diagram of a computer device provided in Example 4 of the present invention. DETAILED DESCRIPTION

[0061] The embodiments of the present invention provide a distribution network state assessment method and system for solving the technical problem that the existing technology mainly evaluates the distribution network state through a robust state estimation method, but usually regards all fluctuation data as false data, resulting in misjudgment of the real fluctuation data, thereby reducing the accuracy of the distribution network state assessment.

[0062] In order to make the purpose, features, and advantages of the present invention more obvious and easy to understand, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the embodiments described below 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 making creative work are within the scope of protection of the present invention.

[0063] See also Figure 1 , Figure 1 This is a flowchart of the steps of a distribution network status assessment method provided in Example 1 of the present invention.

[0064] The present invention provides a distribution network status assessment method, comprising:

[0065] Step 101: Acquire multiple measurement data of the distribution network to be evaluated, filter each measurement data based on a preset window length, and obtain corresponding mutation measurement data and target measurement data;

[0066] The distribution network to be evaluated refers to the power distribution network that requires performance analysis, reliability assessment, optimization planning or fault diagnosis.

[0067] Measurement data refers to real-time or historical electrical measurement information collected by sensors, smart meters, monitoring equipment, etc. during the operation of the power distribution system. Measurement data includes, but is not limited to, innovation sequences and unanalyzed data (i.e., raw, unanalyzed measurement data).

[0068] Window length refers to the time span or number of samples of a continuous data segment used for local analysis in signal processing, data analysis, or time series calculations.

[0069] Sudden change measurement data refers to abnormal data in which the measurement value suddenly deviates from the normal range due to system failure, external interference or equipment abnormality during continuous monitoring.

[0070] Target measurement data refers to measurement data within the normal range.

[0071] In an embodiment of the present invention, multiple measurement data of the distribution network to be evaluated are obtained, and the average covariance of the innovation sequence in each measurement data and the total variance of the data to be analyzed in each measurement data are respectively calculated based on a preset window length, and it is determined whether the total variance of each data to be analyzed is greater than the corresponding average covariance. If the total variance is greater than the corresponding average covariance, the data to be analyzed is determined to be mutation measurement data; if the total variance is less than or equal to the corresponding average covariance, the data to be analyzed is determined to be target measurement data.

[0072] Step 102: Perform anomaly detection on each mutation measurement data based on a preset false data detection model to obtain corresponding target false measurement data;

[0073] The false data detection model refers to an algorithm model used to identify outliers in mutation measurement data. The false data detection model includes a first detection model and a second detection model.

[0074] Targeted false measurement data refers to false data in the distribution network that is maliciously constructed or generated by systematic errors, with the purpose of interfering with system state estimation, control decision-making or safety monitoring.

[0075] In an embodiment of the present invention, a plurality of training measurement data are obtained, data preprocessing is performed on each training measurement data, a measurement feature set is generated, a first feature set of the measurement feature set is input into a first detection model for training, and a first target model and a first error threshold are generated, a second feature set of the measurement feature set is input into a second detection model for training, and a second target model and a second error threshold are generated, each mutation measurement data is screened based on the first target model and the first error threshold to obtain corresponding initial false measurement data, and each initial false measurement data is screened based on the second target model and the second error threshold to obtain corresponding target false measurement data.

[0076] Step 103: Adaptively filter the false measurement data of each target to obtain the corresponding target state value;

[0077] The target state value refers to the measured value after filtering and correction in the target false measurement data.

[0078] In an embodiment of the present invention, based on a preset observation noise proportional function, an observation noise proportional factor matrix corresponding to each target false measurement data is constructed, and Kalman gain processing is performed on each observation noise proportional factor matrix based on the Kalman filter algorithm to obtain a filter coefficient matrix corresponding to each observation noise proportional factor matrix. Each filter coefficient matrix is used to filter and correct the corresponding target false measurement data to obtain a target state value.

[0079] Step 104: Perform a status evaluation based on each target measurement data and each target status value to obtain an evaluation result corresponding to the distribution network to be evaluated.

[0080] The evaluation result refers to the current operating status of the distribution network to be evaluated.

[0081] In an embodiment of the present invention, it is determined whether each target state value is within a preset first threshold range, and whether the real-time measurement value of each target measurement data is within a preset second threshold range. If any target state value is not within the first threshold range or any real-time measurement value is not within the second threshold range, the evaluation result corresponding to the distribution network to be evaluated is determined to be abnormal distribution network operation. If all target state values are within the preset first threshold range and all real-time measurement values are within the second threshold range, the evaluation result corresponding to the distribution network to be evaluated is determined to be normal distribution network operation.

[0082] In an embodiment of the present invention, multiple measurement data of the distribution network to be evaluated are obtained, each measurement data is screened based on a preset window length to obtain corresponding mutation measurement data and target measurement data, each mutation measurement data is detected for anomalies based on a preset false data detection model to obtain corresponding target false measurement data, each target false measurement data is adaptively filtered to obtain a corresponding target state value, and then the state of the distribution network to be evaluated is evaluated based on each target state value and target measurement data. This overcomes the technical problem that the existing technology mainly evaluates the state of the distribution network through a robust state estimation method, but usually regards all fluctuation data as false data, resulting in misjudgment of real fluctuation data and reducing the accuracy of the distribution network state evaluation. Compared with the traditional distribution network evaluation method, the present invention screens the measurement data through a preset window length and a false data detection model, thereby filtering the target false measurement data with weak time correlation and spatial correlation, avoiding the correction of all fluctuation data, and improving the accuracy of the distribution network state evaluation.

[0083] See also Figure 2 , Figure 2 This is a flowchart of the steps of a distribution network status assessment method provided in Example 2 of the present invention.

[0084] The present invention provides a distribution network status assessment method, comprising:

[0085] Step 201: Acquire multiple measurement data of the distribution network to be evaluated, filter each measurement data based on a preset window length, and obtain corresponding mutation measurement data and target measurement data;

[0086] Furthermore, step 201 includes the following sub-steps:

[0087] S11. Calculate the average covariance of the innovation sequence in each measurement data based on a preset window length;

[0088] Innovation sequences, a core concept in signal processing, state estimation (e.g., Kalman filtering), and time series analysis, refer to the sequence of residuals between observed and theoretical values. They reflect unforeseen new information or noise in the system model and are an important tool for evaluating model accuracy and detecting anomalies.

[0089] Average covariance refers to the result of statistical averaging of multiple covariance matrices or covariance values.

[0090] In the embodiment of the present invention, each measurement data and a preset window length are respectively input into a preset average covariance function to obtain the average covariance of the innovation sequence in each measurement data.

[0091] It should be noted that the expression of the innovation sequence is specifically:

[0092]

[0093] in, For the innovation sequence, k+2- The difference between the observed value and the theoretical value at time , k+3- The difference between the observed value and the theoretical value at time , is the difference between the observed value and the theoretical value at time k+1, is the window length, and k is the time value. Observed values refer to electrical measurement information collected by sensors, smart meters, monitoring equipment, and the like. The theoretical value refers to the predicted electrical measurement value obtained by inputting the observed value at the previous time into the state estimation model (for example, inputting the observed value at time k-1 into the state estimation model yields the theoretical value at time k). The state estimation model is a pre-trained CNN model, which can be trained using multiple historical electrical measurement data.

[0094] It should be noted that the average covariance function is specifically:

[0095]

[0096] in, is the difference between the observed value and the theoretical value at the k+2-t moment, is the window length, is the average covariance, t is the time, and k is the moment value.

[0097] S12, calculating the total variance of the data to be analyzed in each measurement data based on a preset window length;

[0098] In an embodiment of the present invention, the total variance of the data to be analyzed is calculated for each measurement data item based on a preset window length. For example, for a particular data item to be analyzed, multiple measurement data points are extracted based on the window length. A measurement covariance matrix is constructed using each measurement data point. The measurement covariance matrix is then input into a preset total variance function to obtain the total variance corresponding to the data item to be analyzed.

[0099] It should be noted that the total variance function is specifically:

[0100]

[0101] in, is the total variance, is the measurement covariance matrix of the Zth data to be analyzed.

[0102] S13, respectively judging whether the total variance of each data to be analyzed is greater than the corresponding average covariance;

[0103] S14. If the total variance is greater than the corresponding average covariance, the data to be analyzed is determined as mutation measurement data;

[0104] In the embodiment of the present invention, it is determined whether the total variance of each data to be analyzed is greater than the corresponding average covariance. When the total variance is greater than the corresponding average covariance, the data to be analyzed is determined to be mutation measurement data.

[0105] S15. If the total variance is less than or equal to the corresponding average covariance, the data to be analyzed is determined as target measurement data.

[0106] In the embodiment of the present invention, when the total variance is less than or equal to the corresponding average covariance, the data to be analyzed is determined as target measurement data.

[0107] Step 202: Perform anomaly detection on each mutation measurement data based on a preset false data detection model to obtain corresponding target false measurement data;

[0108] Furthermore, the false data detection model includes a first detection model and a second detection model, and step 202 includes the following sub-steps:

[0109] S21, obtaining a plurality of training measurement data, performing data preprocessing on each training measurement data, and generating a measurement feature set;

[0110] Data preprocessing refers to cleaning, smoothing and labeling the training measurement data, and extracting input features for inputting the model from the modified training measurement data, thereby obtaining a measurement feature set consisting of multiple input features.

[0111] Training measurement data refers to historical or real-time distribution network measurement datasets used to train machine learning models or state estimation algorithms. Training measurement data includes historical real data, historical data from suspicious nodes, and historical data from real nodes.

[0112] In an embodiment of the present invention, a plurality of training measurement data are acquired, and cleaning, smoothing, and labeling operations are performed on each training measurement data to generate a measurement feature set.

[0113] S22, using a first feature set of the measurement feature set to input a first detection model for training, to generate a first target model and a first error threshold;

[0114] Furthermore, S22 includes the following sub-steps:

[0115] S221, dividing the first feature set of the measurement feature set according to a preset ratio to obtain a first training feature set and a first test feature set;

[0116] The first feature set refers to multiple historical real data after data preprocessing (historical real data is historical data of the distribution network in the previous period without false data injection attacks).

[0117] The first training feature set refers to the historical real data in the process of machine learning or deep learning model training.

[0118] The first test feature set refers to the historical real data in the process of machine learning or deep learning model testing.

[0119] In the embodiment of the present invention, the first feature set is divided into a first training feature set and a first testing feature set according to a ratio of 7:3.

[0120] S222: Train the first detection model using the first training feature set to obtain a first target model;

[0121] In an embodiment of the present invention, the first detection model is trained using the first training feature set until a preset iteration condition is satisfied, thereby obtaining a first target model, wherein the first target model is an LSTM model.

[0122] It's important to note that the LSTM model (Long Short-Term Memory) is a special type of recurrent neural network designed to address the long-term dependency issues of traditional recurrent neural networks. Its core is to selectively retain or discard information through a gating mechanism (input gate, forget gate, and output gate), effectively capturing both long-term and short-term patterns in time series.

[0123] S223, inputting the first test feature set into the first target model to obtain a plurality of test steady-state node state values;

[0124] The test steady-state node state value refers to the power flow calculation result of the power system (mainly composed of synchronous generators, loads and transmission network) predicted by the first objective model.

[0125] In an embodiment of the present invention, the first test feature set is input into the first target model to obtain multiple test steady-state node status values. For example, each historical real data in the first test feature set is input into the first target model to obtain the test steady-state node status value corresponding to each historical real data.

[0126] S224, calculating a first relative prediction error of each test steady-state node state value according to the first test feature set;

[0127] The first relative prediction error refers to the deviation between the predicted value obtained by the first target model and the actual value.

[0128] In an embodiment of the present invention, a standard steady-state node state value corresponding to each test steady-state node state value is selected from a first test feature set, and the difference between each test steady-state node state value and the corresponding standard steady-state node state value is calculated to obtain a plurality of first differences. The absolute value of each first difference is calculated to obtain a plurality of first absolute values. Each first absolute value is then compared with the corresponding standard steady-state node state value to obtain a first relative prediction error for each test steady-state node state value.

[0129] It should be noted that the standard steady-state node status value refers to the power flow calculation result of the power system (mainly composed of synchronous generators, loads and transmission networks) calculated through historical real data.

[0130] S225, sorting the first relative prediction errors from large to small according to their sizes, and selecting the first relative prediction errors of a preset number of values to form a first sequence;

[0131] S226. Select the minimum value in the first sequence as the first error threshold.

[0132] The first error threshold refers to a limit value used to determine whether the sudden change measurement data is initial false measurement data.

[0133] The first sequence refers to the sequence consisting of the first 5% relative forecast errors.

[0134] In an embodiment of the present invention, the first relative prediction errors are sorted from large to small according to their sizes, and the top 5% of the first relative prediction errors are selected to form a first sequence, and the minimum value in the first sequence is selected as the first error threshold.

[0135] S23, using a second feature set of the measurement feature set to input a second detection model for training, to generate a second target model and a second error threshold;

[0136] The second feature set refers to the historical data of multiple suspicious nodes after data preprocessing (that is, the operation records of certain nodes in the past period after they were marked as "suspicious" due to abnormal behavior, performance fluctuations or security risks) and the historical data of multiple real nodes (that is, the operation logs, performance indicators, status parameters, etc. recorded by the actual running nodes in the historical time period).

[0137] The second error threshold refers to a critical value used to determine whether the initial false measurement data is the target false measurement data.

[0138] In an embodiment of the present invention, a second feature set of the measurement feature set is input into a second detection model for training to generate a second target model and a second error threshold, wherein the second detection model is a CNN model. For example, A1: The second feature set is divided into a second training feature set and a second test feature set in a ratio of 7:3. A2: The second detection model is trained using the second training feature set until a preset iteration condition is met, thereby obtaining a second target model. A3: The second test feature set is input into the second target model to obtain multiple test fluctuation node state values. A4: A standard fluctuation node state value corresponding to each test fluctuation node state value is selected from the second test feature set, and the difference between each test fluctuation node state value and the corresponding standard fluctuation node state value is calculated to obtain multiple second differences. The absolute value of each second difference is calculated to obtain multiple second absolute values. Each second absolute value is then compared with the corresponding standard fluctuation node state value to obtain a second relative prediction error for each test fluctuation node state value. A5: The second relative prediction errors are sorted from largest to smallest, and the top 5% of the second relative prediction errors are selected to form a second sequence. The minimum value in the second sequence is selected as the second error threshold.

[0139] It should be noted that the test fluctuation node state value refers to the node operating parameters (such as voltage, frequency, load rate, delay, etc.) when the distribution network undergoes non-steady-state changes predicted by the second objective model.

[0140] It should be noted that the standard fluctuation node status value refers to the node operation parameters in the historical data of the suspicious node.

[0141] It should be noted that the second sequence refers to the sequence consisting of the first 5% of the second relative prediction errors.

[0142] It's important to note that the CNN model (or convolutional neural network model) is a deep learning model specifically designed for processing grid-like data (such as images, video, and audio). Its core concept is to automatically extract spatial features through local receptive fields and parameter sharing. At the spatial level, due to the power flow and physical constraints between nodes, there are strong nonlinear correlations between state data. The weight-sharing and local connectivity structure of the CNN model allows for deeper mining of historical data and extraction of high-dimensional features.

[0143] S24, screening each mutation measurement data based on the first target model and the first error threshold to obtain corresponding initial false measurement data;

[0144] Furthermore, S24 includes the following sub-steps:

[0145] S241, inputting each mutation measurement data into a first target model to obtain a first prediction value corresponding to each mutation measurement data;

[0146] The first predicted value refers to the theoretical value obtained by inputting the mutation measurement data into the first target model.

[0147] In an embodiment of the present invention, each mutation measurement data is input into the first target model to obtain a first predicted value corresponding to each mutation measurement data. For example, if the mutation measurement data includes mutation measurement values at five moments, the mutation measurement values at the first four moments are input into the first target model to obtain the first predicted value corresponding to the mutation measurement data.

[0148] S242, calculating a first relative error between each first prediction value and the corresponding mutation measurement data;

[0149] In this embodiment of the present invention, a first relative error is calculated between each first predicted value and the corresponding mutation measurement data. For example, if the mutation measurement data includes mutation measurement values at five moments, the mutation measurement value at the last moment is used as the current measurement value, and the first relative error between the first predicted value corresponding to the mutation measurement data and the current measurement value is calculated.

[0150] S243, determining whether each first relative error is greater than a first error threshold;

[0151] S244: When the first relative error is greater than a first error threshold, the sudden change measurement data corresponding to the first relative error is determined as initial false measurement data.

[0152] In this embodiment of the present invention, it is determined whether each first relative error is greater than a first error threshold. When the first relative error is greater than the first error threshold, it is determined that the mutation measurement data corresponding to the current first relative error is weak time-correlated data, and the mutation measurement data corresponding to the first relative error is determined as initial false measurement data. When the first relative error is less than or equal to the first error threshold, the mutation measurement data corresponding to the first relative error is strongly correlated data.

[0153] S25 , screening each initial false measurement data based on the second target model and the second error threshold to obtain corresponding target false measurement data.

[0154] In an embodiment of the present invention, each initial false measurement data is screened based on a second target model and a second error threshold to obtain corresponding target false measurement data. For example, B1: Each initial false measurement data is input into the second target model to obtain a second predicted value corresponding to each initial false measurement data. B2: A second relative error is calculated between each second predicted value and the corresponding initial false measurement data. B3: A determination is made as to whether each second relative error is greater than the second error threshold. If the second relative error is greater than the second error threshold, it indicates that the initial false measurement data corresponding to the second relative error is spatially uncorrelated, and the initial false measurement data corresponding to the second relative error is determined as the target false measurement data.

[0155] It should be noted that the second prediction value is obtained by inputting the initial false measurement data into the second target model to obtain the theoretical value of the suspicious node.

[0156] It should be noted that the process of calculating the second relative error between each second predicted value and the corresponding initial false measurement data is specifically: selecting the measurement value associated with the second predicted value from the initial false measurement data corresponding to each second predicted value, and calculating the relative error between each second predicted value and the corresponding measurement value.

[0157] Step 203: construct an observation noise proportional factor matrix corresponding to each target false measurement data based on a preset observation noise proportional function;

[0158] In the embodiment of the present invention, an observation noise proportional factor matrix corresponding to each target false measurement data is constructed according to a preset observation noise proportional function.

[0159] It should be noted that the observation noise ratio function is specifically:

[0160]

[0161] in, is the observation noise scale factor matrix, is the target false measurement data, and i is a sample in the target false measurement data.

[0162] Step 204: Perform Kalman gain processing on each observation noise scale factor matrix based on the Kalman filter algorithm to obtain a filter coefficient matrix corresponding to each observation noise scale factor matrix;

[0163] In the embodiment of the present invention, each observation noise scale factor matrix is input into a preset gain function to obtain a filter coefficient matrix corresponding to each observation noise scale factor matrix.

[0164] It should be noted that the gain function is specifically:

[0165]

[0166] in, is the error variance between the measured value and the state prediction value at time k+1, is the measurement prediction error variance matrix, The observation noise scaling factor that is dynamically adjusted for the target false measurement data identification result, Adjust the variance of the disturbance in the measurements for the observation noise scale factor.

[0167] It is worth mentioning that at time k, the state prediction value and prediction error variance matrix of the next time k+1 are predicted by the dynamic model. The dynamic model is specifically:

[0168]

[0169] in, is the state prediction value at time k+1, n is the total number of samples, i is the sample number, is the estimated value of the i-th sample related to the system state at time k to time k+1, is the prediction error variance matrix, is the state prediction value, is the matrix of the state disturbance vector, and T is the transpose of the matrix.

[0170] Step 205 : Use each filter coefficient matrix to perform filtering correction on the corresponding target false measurement data to obtain the target state value corresponding to each target false measurement data.

[0171] In the embodiment of the present invention, each filter coefficient matrix and the corresponding target false measurement data are respectively input into the filter correction function to obtain the target state value corresponding to each target false measurement data.

[0172] It should be noted that the filter correction function is specifically:

[0173]

[0174] in, is the covariance matrix of the state prediction value and the measurement value, is the filter coefficient matrix, is the difference between the observed value and the theoretical value at time k+1, is the target state value.

[0175] It's worth noting that steps 201-205 only correct the target false measurement data, causing the estimated value corresponding to the target false measurement data to bias toward the predicted value, while other estimated values bias toward the measured value. This avoids erroneous correction of real data and fully respects the actual operation of the power grid. Under a false data injection attack, the voltage phase angle and amplitude metrics estimated by the three state estimation algorithms are compared at appropriate scales. In the early stages, the estimation results of the three CKF algorithms are as follows: the estimation performance of the original CKF (i.e., the Cubature Kalman Filter) deteriorates, and the estimation error suddenly increases. The estimation performance of the robust CKF and the correction method of the present invention is unaffected, and the estimation error of the robust CKF remains greater than that of the correction method of the present invention. In the later stages, as the measurement data continues to be attacked, the estimation results of the three CKF algorithms are as follows: the estimation performance of the original CKF continues to deteriorate, and the estimation error continues to increase. The estimation performance of the robust CKF and the correction method of the present invention is largely unaffected by the false data attack, and both algorithms achieve low estimation errors. This is because the original CKF lacks data recognition capabilities. When a false data injection attack occurs in the system, the original CKF will trust the false data, resulting in large errors in state estimation. Robust CKF can resist the effects of false data injection attacks by identifying and correcting mutant data, and its estimation error is low. The correction method of the present invention can also resist false data attacks and use multi-level data recognition technology to identify the real data generated by new energy sources, and its estimation accuracy is higher than that of robust CKF.

[0176] Step 206: Perform a status evaluation based on each target measurement data and each target status value to obtain an evaluation result corresponding to the distribution network to be evaluated.

[0177] Furthermore, step 206 includes the following sub-steps:

[0178] S31, determining whether each target state value is within a preset first threshold range;

[0179] S32: If any target state value is not within the first threshold value range, determining the evaluation result corresponding to the distribution network to be evaluated as abnormal operation of the distribution network;

[0180] The first threshold value interval refers to the rated threshold value interval corresponding to the target state value. For example, when the target state value is the target state value of the voltage measurement data, the rated voltage value interval corresponding to the target state value is used. When the target state value is the target state value of the current measurement data, the rated current value interval corresponding to the target state value is used.

[0181] In an embodiment of the present invention, it is determined whether each target state value is within a preset first threshold range. If any target state value is not within the first threshold range, it indicates that the distribution network is operating abnormally, and the evaluation result corresponding to the distribution network to be evaluated is determined as abnormal distribution network operation.

[0182] S33: If all target state values are within the first threshold range, determine whether the real-time measurement value of each target measurement data is within a preset second threshold range;

[0183] S34. If any real-time measurement value is not within the second threshold range, determining the evaluation result corresponding to the distribution network to be evaluated as abnormal operation of the distribution network;

[0184] The second threshold interval refers to the rated threshold interval corresponding to the real-time measurement value. For example, when the real-time measurement value is a voltage value, the rated threshold interval corresponding to the real-time measurement value is the rated voltage value interval. When the real-time measurement value is a current value, the rated threshold interval corresponding to the real-time measurement value is the rated current value interval.

[0185] In an embodiment of the present invention, if all target state values are within the first threshold range, it is determined whether the real-time measurement value of each target measurement data is within a preset second threshold range. If any real-time measurement value is not within the second threshold range, it indicates that the distribution network is operating abnormally, and the evaluation result corresponding to the distribution network to be evaluated is determined as abnormal distribution network operation.

[0186] S35. If all real-time measurement values are within the second threshold range, the evaluation result corresponding to the distribution network to be evaluated is determined as normal operation of the distribution network.

[0187] In an embodiment of the present invention, if all real-time measurement values are within the second threshold range and all target state values are within the first threshold range, the distribution network operates normally, and the evaluation result corresponding to the distribution network to be evaluated is determined as the distribution network operates normally.

[0188] In an embodiment of the present invention, multiple measurement data of the distribution network to be evaluated are obtained, each measurement data is screened based on a preset window length to obtain corresponding mutation measurement data and target measurement data, each mutation measurement data is detected for anomalies based on a preset false data detection model to obtain corresponding target false measurement data, each target false measurement data is adaptively filtered to obtain a corresponding target state value, and then the state of the distribution network to be evaluated is evaluated based on each target state value and target measurement data. This overcomes the technical problem that the existing technology mainly evaluates the state of the distribution network through a robust state estimation method, but usually regards all fluctuation data as false data, resulting in misjudgment of real fluctuation data and reducing the accuracy of the distribution network state evaluation. Compared with the traditional distribution network evaluation method, the present invention screens the measurement data through a preset window length and a false data detection model, thereby filtering the target false measurement data with weak time correlation and spatial correlation, avoiding the correction of all fluctuation data, and improving the accuracy of the distribution network state evaluation.

[0189] See also Figure 3 , Figure 3 This is a structural block diagram of a distribution network status assessment system provided in Example 3 of the present invention.

[0190] The present invention provides a distribution network status assessment system, comprising:

[0191] The acquisition module 301 is used to obtain multiple measurement data of the distribution network to be evaluated, filter each measurement data based on a preset window length, and obtain corresponding mutation measurement data and target measurement data;

[0192] A detection module 302 is configured to perform anomaly detection on each mutation measurement data based on a preset false data detection model to obtain corresponding target false measurement data;

[0193] The filtering module 303 is used to perform adaptive filtering on the false measurement data of each target to obtain the corresponding target state value;

[0194] The evaluation module 304 is configured to perform a state evaluation based on each target measurement data and each target state value to obtain an evaluation result corresponding to the distribution network to be evaluated.

[0195] Furthermore, the acquisition module 301 includes:

[0196] The average covariance submodule is used to calculate the average covariance of the innovation sequence in each measurement data based on a preset window length;

[0197] The total variance submodule is used to calculate the total variance of the data to be analyzed in each measurement data based on a preset window length;

[0198] The first screening submodule is used to determine whether the total variance of each data to be analyzed is greater than the corresponding average covariance;

[0199] If the total variance is greater than the corresponding average covariance, the data to be analyzed are determined to be mutation measurement data;

[0200] If the total variance is less than or equal to the corresponding average covariance, the data to be analyzed is determined as the target measurement data;

[0201] The expression of the innovation sequence is specifically:

[0202]

[0203] in, For the innovation sequence, k+2- The difference between the observed value and the theoretical value at time , k+3- The difference between the observed value and the theoretical value at time , is the difference between the observed value and the theoretical value at time k+1, is the window length, and k is the time value.

[0204] Furthermore, the false data detection model includes a first detection model and a second detection model, and the detection module 302 includes:

[0205] A preprocessing submodule is used to obtain multiple training measurement data, perform data preprocessing on each training measurement data, and generate a measurement feature set;

[0206] A first training submodule is configured to input a first feature set of the measurement feature set into a first detection model for training to generate a first target model and a first error threshold;

[0207] A second training submodule is configured to use a second feature set of the measurement feature set to input a second detection model for training, thereby generating a second target model and a second error threshold;

[0208] A second screening submodule is used to screen each mutation measurement data based on the first target model and the first error threshold to obtain corresponding initial false measurement data;

[0209] The third screening submodule is configured to screen each initial false measurement data based on the second target model and the second error threshold to obtain corresponding target false measurement data.

[0210] Furthermore, the first training submodule includes:

[0211] a dividing unit, configured to divide the first feature set of the measurement feature set according to a preset ratio to obtain a first training feature set and a first test feature set;

[0212] A first training unit is used to train a first detection model using a first training feature set to obtain a first target model;

[0213] A first testing unit is configured to input a first test feature set into a first target model to obtain a plurality of test steady-state node state values;

[0214] A first selection unit is used to calculate a first relative prediction error of each test steady-state node state value according to the first test feature set;

[0215] Sort the first relative prediction errors from large to small, and select the first relative prediction errors of a preset number of values to form a first sequence;

[0216] The minimum value in the first sequence is selected as the first error threshold.

[0217] Furthermore, the second screening submodule includes:

[0218] A first detection unit is used to input each mutation measurement data into a first target model to obtain a first prediction value corresponding to each mutation measurement data;

[0219] A first analysis unit is used to calculate a first relative error between each first prediction value and the corresponding mutation measurement data;

[0220] Determining whether each first relative error is greater than a first error threshold;

[0221] When the first relative error is greater than the first error threshold, the sudden change measurement data corresponding to the first relative error is determined as initial false measurement data.

[0222] Furthermore, the filtering module 303 includes:

[0223] The first filtering submodule is used to construct an observation noise proportional factor matrix corresponding to each target false measurement data based on a preset observation noise proportional function;

[0224] The second filtering submodule is used to perform Kalman gain processing on each observation noise scale factor matrix based on the Kalman filtering algorithm to obtain a filter coefficient matrix corresponding to each observation noise scale factor matrix;

[0225] The third filtering submodule is configured to perform filtering correction on the corresponding target false measurement data using each filtering coefficient matrix to obtain a target state value corresponding to each target false measurement data.

[0226] Furthermore, the evaluation module 304 includes:

[0227] A first analysis submodule is used to determine whether each target state value is within a preset first threshold range;

[0228] If any target state value is not within the first threshold value interval, the evaluation result corresponding to the distribution network to be evaluated is determined as abnormal operation of the distribution network;

[0229] A second analysis submodule is configured to determine whether the real-time measurement value of each target measurement data is within a preset second threshold range if all target state values are within the first threshold range;

[0230] If any real-time measurement value is not within the second threshold range, the evaluation result corresponding to the distribution network to be evaluated is determined as abnormal operation of the distribution network;

[0231] If all real-time measurement values are within the second threshold range, the evaluation result corresponding to the distribution network to be evaluated is determined to be that the distribution network is operating normally.

[0232] See also Figure 4 , Figure 4 This is a structural block diagram of a computer device provided in Example 4 of the present invention.

[0233] An electronic device according to an embodiment of the present invention includes: a memory 401 and a processor 402, wherein the memory 401 stores a computer program; when the computer program is executed by the processor 402, the processor 402 executes the distribution network status assessment method as described in any of the above embodiments.

[0234] Memory 401 may be an electronic memory such as flash memory, EEPROM (Electrically Erasable Programmable Read-Only Memory), EPROM, hard disk, or ROM. Memory 401 has storage space 403 for program code 413 for executing any of the method steps described above. For example, storage space 403 for program code may include individual program codes 413 for implementing various steps in the method described above. These program codes may be read from or written to one or more computer program products. These computer program products include program code carriers such as hard disks, compact disks (CDs), memory cards, or floppy disks. The program codes may be compressed, for example, in a suitable format. When executed by a processing device, these codes cause the processing device to execute the various steps in the method described above. These program codes may be read from or written to one or more computer program products. These computer program products include program code carriers such as hard disks, compact disks (CDs), memory cards, or floppy disks. The program codes may be compressed, for example, in a suitable format. When these codes are executed by a computing and processing device, they cause the computing and processing device to execute the various steps in the above-described method for evaluating the state of a power distribution network.

[0235] The fifth embodiment of the present invention further provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the distribution network state assessment method as described in any of the above embodiments is implemented.

[0236] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0237] In the several embodiments provided in this application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are merely schematic. For example, the division of units is only a logical function division. In actual implementation, there may be other division methods, such as 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 mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interface, device or unit, which can be electrical, mechanical or other forms.

[0238] Units described as separate components may or may not be physically separate, and components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0239] In addition, the functional units in the various embodiments of the present invention may be integrated into a single processing unit, each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.

[0240] If the integrated unit is implemented as 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, or the portion that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the various embodiments of the method of the present invention. The aforementioned storage medium includes various media that can store program code, such as USB flash drives, mobile hard drives, read-only memories (ROMs), random access memories (RAMs), magnetic disks, or optical disks.

[0241] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A distribution network status assessment method, characterized in that: include: Acquire multiple measurement data of the distribution network to be evaluated, filter each of the measurement data based on a preset window length, and obtain corresponding mutation measurement data and target measurement data; Performing anomaly detection on each of the mutation measurement data based on a preset false data detection model to obtain corresponding target false measurement data; Adaptively filtering each of the target false measurement data to obtain a corresponding target state value; Performing a state evaluation based on each of the target measurement data and each of the target state values to obtain an evaluation result corresponding to the distribution network to be evaluated; The false data detection model includes a first detection model and a second detection model. The step of performing anomaly detection on each of the mutation measurement data based on the preset false data detection model to obtain corresponding target false measurement data includes: Acquire a plurality of training measurement data, perform data preprocessing on each of the training measurement data, and generate a measurement feature set; Using a first feature set of the measurement feature set to input the first detection model for training to generate a first target model and a first error threshold; Using a second feature set of the measurement feature set to input the second detection model for training to generate a second target model and a second error threshold; Screening each of the mutation measurement data based on the first target model and the first error threshold to obtain corresponding initial false measurement data; screening each of the initial false measurement data based on the second target model and the second error threshold to obtain corresponding target false measurement data; The step of using the first feature set of the measurement feature set to input the first detection model for training to generate a first target model and a first error threshold includes: Dividing the first feature set of the measurement feature set according to a preset ratio to obtain a first training feature set and a first test feature set; Training the first detection model using the first training feature set to obtain a first target model; Inputting the first test feature set into the first target model to obtain a plurality of test steady-state node state values; Calculating a first relative prediction error of each of the test steady-state node state values according to the first test feature set; Sort the first relative prediction errors from large to small according to their sizes, and select the first relative prediction errors of a preset number of values to form a first sequence; The minimum value in the first sequence is selected as the first error threshold.

2. The distribution network status assessment method according to claim 1, characterized in that: The step of screening each of the measurement data based on a preset window length to obtain corresponding mutation measurement data and target measurement data includes: Calculating the average covariance of the innovation sequence in each of the measurement data based on a preset window length; Calculating the total variance of the data to be analyzed in each of the measurement data based on a preset window length; Determine whether the total variance of each of the data to be analyzed is greater than the corresponding average covariance; If the total variance is greater than the corresponding average covariance, the data to be analyzed is determined to be mutation measurement data; If the total variance is less than or equal to the corresponding average covariance, the data to be analyzed is determined as target measurement data; The expression of the innovation sequence is specifically: ； in, For the innovation sequence, k+2- The difference between the observed value and the theoretical value at time , k+3- The difference between the observed value and the theoretical value at time , is the difference between the observed value and the theoretical value at time k+1, is the window length, and k is the time value.

3. The distribution network status assessment method according to claim 1, characterized in that: The step of screening each of the mutation measurement data based on the first target model and the first error threshold to obtain corresponding initial false measurement data includes: Inputting each of the mutation measurement data into the first target model to obtain a first prediction value corresponding to each of the mutation measurement data; Calculating a first relative error between each of the first prediction values and the corresponding mutation measurement data; Determining whether each of the first relative errors is greater than the first error threshold; When the first relative error is greater than the first error threshold, the sudden change measurement data corresponding to the first relative error is determined as initial false measurement data.

4. The distribution network status assessment method according to claim 1, characterized in that: The step of performing adaptive filtering on each of the target false measurement data to obtain a corresponding target state value includes: Based on a preset observation noise proportional function, constructing an observation noise proportional factor matrix corresponding to each target false measurement data; Performing Kalman gain processing on each of the observation noise scale factor matrices based on a Kalman filter algorithm to obtain a filter coefficient matrix corresponding to each of the observation noise scale factor matrices; The corresponding target false measurement data are filtered and corrected using each of the filter coefficient matrices to obtain a target state value corresponding to each of the target false measurement data.

5. The distribution network status assessment method according to claim 1, characterized in that: The step of performing a state evaluation according to each of the target measurement data and each of the target state values to obtain an evaluation result corresponding to the distribution network to be evaluated includes: Determining whether each of the target state values is within a preset first threshold range; If any of the target state values is not within the first threshold range, determining the evaluation result corresponding to the distribution network to be evaluated as abnormal operation of the distribution network; If all of the target state values are within the first threshold range, determining whether the real-time measurement value of each of the target measurement data is within a preset second threshold range; If any of the real-time measurement values is not within the second threshold range, determining the evaluation result corresponding to the distribution network to be evaluated as abnormal operation of the distribution network; If all of the real-time measurement values are within the second threshold range, the evaluation result corresponding to the distribution network to be evaluated is determined to be that the distribution network is operating normally.

6. A distribution network status assessment system, characterized in that: include: An acquisition module is used to obtain multiple measurement data of the distribution network to be evaluated, and filter each of the measurement data based on a preset window length to obtain corresponding mutation measurement data and target measurement data; A detection module, configured to perform anomaly detection on each of the mutation measurement data based on a preset false data detection model to obtain corresponding target false measurement data; A filtering module, configured to perform adaptive filtering on each of the target false measurement data to obtain a corresponding target state value; An evaluation module, configured to perform a state evaluation based on each of the target measurement data and each of the target state values, and obtain an evaluation result corresponding to the distribution network to be evaluated; The false data detection model includes a first detection model and a second detection model, and the detection module includes: A preprocessing submodule is used to obtain a plurality of training measurement data, perform data preprocessing on each of the training measurement data, and generate a measurement feature set; A first training submodule is configured to use a first feature set of the measurement feature set to input the first detection model for training, thereby generating a first target model and a first error threshold; A second training submodule is configured to use a second feature set of the measurement feature set to input the second detection model for training, thereby generating a second target model and a second error threshold; a second screening submodule, configured to screen each of the mutation measurement data based on the first target model and the first error threshold to obtain corresponding initial false measurement data; a third screening submodule, configured to screen each of the initial false measurement data based on the second target model and the second error threshold to obtain corresponding target false measurement data; The first training submodule includes: a dividing unit, configured to divide the first feature set of the measurement feature set according to a preset ratio to obtain a first training feature set and a first test feature set; A first training unit is configured to train the first detection model using the first training feature set to obtain a first target model; A first testing unit, configured to input the first test feature set into the first target model to obtain a plurality of test steady-state node state values; A first selection unit, configured to calculate a first relative prediction error of each of the test steady-state node state values according to the first test feature set; Sort the first relative prediction errors from large to small according to their sizes, and select the first relative prediction errors of a preset number of values to form a first sequence; The minimum value in the first sequence is selected as the first error threshold.

7. An electronic device, characterized in that: The method comprises a memory and a processor, wherein 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 status assessment method according to any one of claims 1 to 5.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed, the distribution network status assessment method according to any one of claims 1 to 5 is implemented.

Images