A distribution station area risk classification early warning method and system

Abstract

The present invention relates to a distribution substation risk classification warning method and system, comprising: extracting features from target monitoring data of a distribution substation collected at a first data collection time interval to obtain first substation data features; based on the first substation data features, performing correlation analysis on preset feature indicators to obtain first key warning indicators; using the first key warning indicators to fuse feature indicators to obtain first new feature data; based on the first new feature data, using a warning model to predict the warning level of the distribution substation; adjusting the first data collection time interval to obtain a second data collection time interval according to the warning level of the distribution substation; re-collecting target monitoring data based on the second data collection time interval, and re-obtaining the warning level of the distribution substation according to the re-collected target monitoring data. The present invention realizes flexible adjustment of the collection interval of the distribution substation, and improves the accuracy and reliability of the risk warning of the distribution substation.

Description

Technical Field

[0001] The present invention belongs to the technical field of distribution network operation monitoring, and in particular relates to a distribution network area risk classification early warning method and system. Background Art

[0002] As the number of electric vehicles increases significantly, larger electric vehicle charging stations are gradually connected to the distribution network, providing convenient energy services for electric vehicle charging and discharging. Since the operating status monitoring equipment and usage methods of some distribution network substations still have a certain lag, the current operating monitoring capabilities of distribution substations containing electric vehicle charging stations are still insufficient, and it is difficult to meet the needs of efficient monitoring of distribution substations containing electric vehicle charging stations. In addition, during the operation of electric vehicle charging stations, due to the centralized charging of a large number of electric vehicles, the substations where they are located may experience operating status degradation problems such as heavy overload and voltage deviation. At the same time, the distribution network lacks efficient methods for monitoring and early warning of distribution substations, and it is difficult to meet the monitoring and early warning needs to ensure the high-quality operation of the distribution network.

[0003] At present, there have been studies on the problem of operating status monitoring of distribution network substations. The conventional method is to collect relevant electrical quantities of the substation at intervals of 15 minutes, analyze and warn about the operating status. This method can monitor and manage distribution substations with relatively low operating status requirements, but it is difficult to achieve more timely warning of operating status degradation when solving the monitoring and early warning problems of distribution substations containing electric vehicle charging stations. If the operating status of distribution substations containing electric vehicle charging stations is not monitored more efficiently and more accurately, it will seriously affect the operating status of distribution substations containing electric vehicle charging stations, and restrict the energy supply level and reliability of the substations where the charging stations are located. Summary of the invention

[0004] In order to overcome the problems existing in the above-mentioned related technologies, the present application provides a distribution station area risk classification early warning method and system.

[0005] According to a first aspect of an embodiment of the present application, a distribution station area risk classification warning method is provided, comprising:

[0006] Extracting features of the target monitoring data of the distribution station area collected based on the first data collection time interval to obtain data features of the first station area;

[0007] Based on the data characteristics of the first substation area, a correlation analysis is performed on the preset characteristic indicators to obtain a first key early warning indicator;

[0008] Using the first key early warning indicator to perform feature indicator fusion to obtain first new feature data;

[0009] Based on the first new feature data, using the early warning model to predict the early warning level of the distribution station area;

[0010] According to the warning level of the distribution station area, the first data collection time interval is adjusted to obtain a second data collection time interval;

[0011] The target monitoring data is re-collected based on the second data collection time interval, and the warning level of the distribution station area is re-acquired according to the re-collected target monitoring data.

[0012] Preferably, the step of extracting features from the target monitoring data of the distribution station area collected based on the first data collection time interval to obtain the first station area data features includes:

[0013] Performing data cleaning on the target monitoring data;

[0014] Data corresponding to preset characteristic indicators are extracted from the target monitoring data after data cleaning to obtain the data characteristics of the first station area.

[0015] Preferably, the first key early warning indicator is obtained by performing a correlation analysis on the preset characteristic indicators based on the first station area data characteristics, including:

[0016] Based on the data characteristics of the first substation area, a grayscale correlation analysis method is used to perform correlation analysis on the preset characteristic indicators to obtain the first key early warning indicator.

[0017] Preferably, based on the first station area data feature, a grayscale correlation analysis method is used to perform correlation analysis on the preset characteristic indicators to obtain the first key warning indicator, including:

[0018] Constructing a first comparison sequence using the first station area data feature;

[0019] Constructing a first reference series using the collected first historical monitoring data;

[0020] Performing standardization processing on both the first comparison number series and the first reference number series;

[0021] Calculate the absolute difference between the first comparison series after the normalization processing and the first reference series after the normalization processing corresponding to each preset characteristic index;

[0022] Determine the grayscale correlation between the first comparison series after the normalization processing and the first reference series after the normalization processing corresponding to each preset characteristic index by using the absolute difference;

[0023] A preset characteristic indicator corresponding to the grayscale correlation degree that meets the preset conditions is selected as the first key warning indicator.

[0024] Preferably, the step of constructing a first reference number series using the collected first historical monitoring data includes:

[0025] Performing data cleaning on the first historical monitoring data;

[0026] Performing standardization processing on the first historical monitoring data after data cleaning corresponding to the preset characteristic indicator to obtain the first historical monitoring data after standardization processing corresponding to the preset characteristic indicator;

[0027] Let the minimum value of the first historical monitoring data after the standardization processing corresponding to each preset characteristic index be the optimal value, and the maximum value of the first historical monitoring data after the standardization processing corresponding to each preset characteristic index be the worst value;

[0028] The first reference sequence is constructed using the optimal value and the worst value corresponding to the preset characteristic index.

[0029] Preferably, the step of using the absolute difference to determine the grayscale correlation between the first comparison series after standardization and the first reference series after standardization corresponding to each preset characteristic index includes:

[0030] According to the absolute difference, determine the maximum absolute difference and the minimum absolute difference between the first comparison series after the normalization processing and the first reference series after the normalization processing corresponding to each preset characteristic index;

[0031] The maximum absolute difference and the minimum absolute difference are used to calculate the grayscale correlation between the first comparison series after the normalization processing and the first reference series after the normalization processing corresponding to each preset characteristic index.

[0032] Preferably, the step of using the first key early warning indicator to fuse the feature indicators to obtain the first new feature data includes:

[0033] Extracting data corresponding to the first key early warning indicator from the target monitoring data after data cleaning, and constructing a first feature set matrix;

[0034] Based on the first feature set matrix, the first new feature data is constructed using a principal component analysis method.

[0035] Preferably, constructing the first new feature data based on the first feature set matrix using a principal component analysis method includes:

[0036] Using the first feature set matrix, a first covariance matrix is ​​calculated;

[0037] Calculate the cumulative contribution rate of the first p features in the first covariance matrix;

[0038] When the cumulative contribution rate of the first p features is greater than or equal to the expected value of the cumulative contribution rate, constructing the first new feature data using the first p features in the first covariance matrix;

[0039] Where p is a positive integer, p≤k, and k is the total number of features in the first covariance matrix.

[0040] Preferably, the first data collection time interval is adjusted according to the warning level of the distribution station area to obtain the second data collection time interval, including:

[0041] If the warning level is the fourth warning level or the number of electric vehicles entering the distribution station area reaches the first percentage of the charging station capacity, the second data collection time interval is the first multiple of the first data collection time interval;

[0042] If the warning level is the third warning level or the number of electric vehicles entering the distribution station area reaches the second percentage of the charging station capacity, the second data collection time interval is the second multiple of the first data collection time interval;

[0043] If the warning level is lower than or equal to the second warning level, and the number of electric vehicles entering the distribution station area is less than the second percentage of the charging station capacity, the second data collection time interval is the third multiple of the first data collection time interval;

[0044] If the warning level is lower than or equal to the first warning level, and the number of electric vehicles entering the distribution station area is less than the second percentage of the charging station capacity, the second data collection time interval is the first data collection time interval.

[0045] Preferably, the process of establishing the early warning model includes:

[0046] Collect the second historical monitoring data of the distribution station area;

[0047] Extracting features from the second historical monitoring data to obtain second area data features;

[0048] Based on the data characteristics of the second substation area, a correlation analysis is performed on the preset characteristic indicators to obtain a second key early warning indicator;

[0049] Using the second key early warning indicator to fuse the characteristic indicators, to obtain second new characteristic data;

[0050] Determine the historical warning level of the power distribution station area corresponding to the second new characteristic data;

[0051] A data set is constructed using the second new feature data and the historical warning level of the distribution station area, and the long-short neural network model is trained using the data set to obtain the warning model.

[0052] Preferably, the extracting features of the second historical monitoring data to obtain the second area data features includes:

[0053] Performing data cleaning on the second historical monitoring data;

[0054] Data corresponding to preset characteristic indicators are extracted from the second historical monitoring data after data cleaning to obtain the data characteristics of the second station area.

[0055] Preferably, the correlation analysis of preset characteristic indicators is performed based on the second station area data characteristics to obtain the second key early warning indicator, including:

[0056] Based on the data characteristics of the second substation area, a grayscale correlation analysis method is used to perform correlation analysis on the preset characteristic indicators to obtain the second key warning indicator.

[0057] Preferably, based on the second station area data feature, a grayscale correlation analysis method is used to perform correlation analysis on preset characteristic indicators to obtain the second key warning indicator, including:

[0058] Constructing a second comparison sequence using the second station area data feature;

[0059] Constructing a second reference series using the collected third historical monitoring data;

[0060] performing standardization processing on both the second comparison number series and the second reference number series;

[0061] Calculate the absolute difference between the second comparison series after the normalization processing and the second reference series after the normalization processing corresponding to each preset characteristic index;

[0062] Determine the grayscale correlation between the second comparison series after the normalization processing and the second reference series after the normalization processing corresponding to each preset characteristic index by using the absolute difference;

[0063] Selecting a characteristic indicator corresponding to the grayscale correlation degree that meets the preset conditions as the second key early warning indicator;

[0064] Wherein, the collection time of the third historical monitoring data is earlier than the collection time of the second historical monitoring data.

[0065] Preferably, the step of constructing a second reference number series using the collected third historical monitoring data includes:

[0066] Performing data cleaning on the third historical monitoring data;

[0067] Performing standardization processing on the third historical monitoring data after data cleaning corresponding to the preset characteristic indicator to obtain the third historical monitoring data after standardization processing corresponding to the preset characteristic indicator;

[0068] Let the minimum value of the third historical monitoring data after the standardization processing corresponding to each preset characteristic index be the optimal value, and the maximum value of the third historical monitoring data after the standardization processing corresponding to each preset characteristic index be the worst value;

[0069] The second reference sequence is constructed using the optimal value and the worst value corresponding to the preset characteristic index.

[0070] Preferably, the step of using the absolute difference to determine the grayscale correlation between the second comparison series after standardization and the second reference series after standardization corresponding to each preset characteristic index includes:

[0071] According to the absolute difference, determine the maximum absolute difference and the minimum absolute difference between the second comparison series after the normalization processing and the second reference series after the normalization processing corresponding to each preset characteristic index;

[0072] The maximum absolute difference and the minimum absolute difference are used to calculate the grayscale correlation between the second comparison series after the normalization processing and the second reference series after the normalization processing corresponding to each preset characteristic index.

[0073] Preferably, the step of using the second key early warning indicator to fuse the characteristic indicator to obtain the second new characteristic data includes:

[0074] Extracting data corresponding to the second key early warning indicator from the second historical monitoring data after data cleaning, and constructing a second feature set matrix;

[0075] Based on the second feature set matrix, the second new feature data is constructed by using a principal component analysis method.

[0076] Preferably, constructing the second new feature data based on the second feature set matrix by using a principal component analysis method includes:

[0077] Using the second feature set matrix, a second covariance matrix is ​​calculated;

[0078] Calculate the cumulative contribution rate of the first p' features in the second covariance matrix;

[0079] When the cumulative contribution rate of the first p' features is greater than or equal to the expected value of the cumulative contribution rate, the second new feature data is constructed using the first p' features in the second covariance matrix;

[0080] Wherein, p' is a positive integer, p'≤k', and k' is the total number of features in the second covariance matrix.

[0081] Preferably, the determining of the historical warning level of the distribution station area corresponding to the second new characteristic data includes:

[0082] Extract valid data from the second new feature data according to the preset standard limit value of the normal operation of the substation and the preset standard limit value of the fault operation of the substation corresponding to the second key warning indicator, and obtain the processed second new feature data;

[0083] Based on the second key warning indicator, a K-means clustering method is used to perform feature clustering on the processed second new feature data to obtain clustered second new feature data;

[0084] Based on the clustered second new feature data, an interval prediction method is used to analyze and obtain the change extreme value corresponding to the second key warning indicator;

[0085] Based on the clustered second new feature data and the extreme value of change corresponding to the second key warning indicator, a Bootstrap algorithm is used to predict the sample variance of the Bootstrap sample corresponding to the second key warning indicator;

[0086] Using the sample variance corresponding to the second key warning indicator, calculate the upper limit value and the lower limit value of the second key warning indicator;

[0087] According to the upper limit value and the lower limit value of the second key warning indicator, the historical warning level of the distribution station area corresponding to the second new characteristic data is determined.

[0088] Preferably, determining the historical warning level of the distribution station area corresponding to the second new feature data according to the upper limit value and the lower limit value of the second key warning indicator includes:

[0089] When the upper limit value and the lower limit value of the second key warning indicator belong to the fluctuation range corresponding to the first confidence threshold value, the historical warning level of the distribution station area is the first warning level;

[0090] When the upper limit value and the lower limit value of the second key warning indicator belong to the fluctuation range corresponding to the second confidence threshold, the historical warning level of the distribution station area is the second warning level;

[0091] When the upper limit value and the lower limit value of the second key warning indicator belong to the fluctuation range corresponding to the third confidence threshold value, the historical warning level of the distribution station area is the third warning level;

[0092] When the upper limit value and the lower limit value of the second key warning indicator belong to the fluctuation range corresponding to the fourth confidence threshold value, the historical warning level of the distribution station area is the fourth warning level.

[0093] Preferably, the using the data set to train the long-short neural network model to obtain the early warning model includes:

[0094] Dividing the data set into a training set, a validation set, and a test set;

[0095] Using the training set to train the long-short neural network model to obtain a trained long-short neural network model;

[0096] The trained long-short neural network model is verified using the verification set. If the verification is successful, the performance of the trained long-short neural network model is evaluated using the test set. If the verification fails, the hyperparameters of the long-short neural network model are adjusted and overfitting is monitored, and the long-short neural network model is retrained until the verification is successful.

[0097] The performance of the trained long and short neural network model is evaluated using the test set. If the performance evaluation of the trained long and short neural network model is successful, the trained long and short neural network model is the early warning model. If the performance evaluation of the trained long and short neural network model fails, the second historical monitoring data is recollected, and an early warning model is established using the recollected second historical monitoring data.

[0098] Preferably, the calculation formula of the first covariance matrix includes:

[0099] B=X T X

[0100] The calculation formula for the cumulative contribution rate of the first p features in the first covariance matrix includes:

[0101]

[0102] In the above formula, X is the first feature set matrix, T is the transpose, and B is the first covariance matrix; , k is the total number of features in the first covariance matrix, p≤k; is the cumulative contribution rate of the first p features in the first covariance matrix, is the eigenvalue of the i-th feature in the first covariance matrix.

[0103] Preferably, the calculation formula of the second covariance matrix includes:

[0104] B'=X' T X'

[0105] The calculation formula of the cumulative contribution rate of the first p' features in the second covariance matrix includes:

[0106]

[0107] In the above formula, X' is the second feature set matrix, T is the transpose, and B' is the second covariance matrix; i'∈ [ 1,k' ], k' is the total number of features in the second covariance matrix, p'≤k'; is the cumulative contribution rate of the first p' features in the second covariance matrix, is the eigenvalue of the i'th feature in the second covariance matrix.

[0108] Preferably, the calculation formula for the upper limit value of the second key warning indicator includes:

[0109]

[0110] The calculation formula for the lower limit value of the second key warning indicator includes:

[0111]

[0112] In the above formula, is the upper limit value of the second key warning indicator, is the lower limit of the second key warning indicator, is the mean value of the second new feature data after clustering corresponding to the second key warning indicator, is the sample variance corresponding to the second key early warning indicator, is a standard normal distribution 100·(1- α )% corresponding critical value, α is the adjustment factor.

[0113] Preferably, the power distribution station area is a power distribution station area including an electric vehicle charging station.

[0114] According to a second aspect of an embodiment of the present application, a distribution station area risk classification warning system is provided, including:

[0115] An extraction unit, used for extracting features from the target monitoring data of the distribution station area collected based on the first data collection time interval to obtain data features of the first station area;

[0116] A first acquisition unit is used to perform a correlation analysis on a preset characteristic indicator based on the first station area data feature to obtain a first key warning indicator;

[0117] A second acquisition unit is used to use the first key warning indicator to perform feature indicator fusion to obtain first new feature data;

[0118] A third acquisition unit is used to predict the warning level of the power distribution station area by using the warning model based on the first new feature data;

[0119] An adjusting unit, configured to adjust the first data collection time interval according to the warning level of the distribution station area to obtain a second data collection time interval;

[0120] The fourth acquisition unit is used to re-collect the target monitoring data based on the second data collection time interval, and re-acquire the warning level of the distribution station area according to the re-collected target monitoring data.

[0121] Preferably, the extraction unit comprises:

[0122] A data cleaning subunit, used for cleaning the target monitoring data;

[0123] The first acquisition subunit is used to extract data corresponding to preset characteristic indicators from the target monitoring data after data cleaning to obtain the first station area data characteristics.

[0124] Preferably, the first acquisition unit includes:

[0125] The second acquisition subunit is used to perform correlation analysis on the preset characteristic indicators based on the first station area data characteristics by using a grayscale correlation analysis method to obtain the first key warning indicator.

[0126] Preferably, the second acquisition subunit includes:

[0127] A first construction module, used to construct a first comparison sequence using the first station area data feature;

[0128] A second construction module is used to construct a first reference number series using the collected first historical monitoring data;

[0129] A processing module, configured to perform standardization processing on both the first comparison number series and the first reference number series;

[0130] A first calculation module, used for calculating the absolute difference between the first comparison series after standardization and the first reference series after standardization corresponding to each preset characteristic index;

[0131] A determination module, used to determine the grayscale correlation between the first comparison series after standardization and the first reference series after standardization corresponding to each preset characteristic index by using the absolute difference;

[0132] A selection module is used to select a preset characteristic indicator corresponding to a grayscale correlation degree that meets a preset condition as the first key warning indicator.

[0133] Preferably, the second building block is specifically used for:

[0134] Performing data cleaning on the first historical monitoring data;

[0135] Performing standardization processing on the first historical monitoring data after data cleaning corresponding to the preset characteristic indicator to obtain the first historical monitoring data after standardization processing corresponding to the preset characteristic indicator;

[0136] Let the minimum value of the first historical monitoring data after the standardization processing corresponding to each preset characteristic index be the optimal value, and the maximum value of the first historical monitoring data after the standardization processing corresponding to each preset characteristic index be the worst value;

[0137] The first reference sequence is constructed using the optimal value and the worst value corresponding to the preset characteristic index.

[0138] Preferably, the determining module is specifically used to:

[0139] According to the absolute difference, determine the maximum absolute difference and the minimum absolute difference between the first comparison series after the normalization processing and the first reference series after the normalization processing corresponding to each preset characteristic index;

[0140] The maximum absolute difference and the minimum absolute difference are used to calculate the grayscale correlation between the first comparison series after the normalization processing and the first reference series after the normalization processing corresponding to each preset characteristic index.

[0141] Preferably, the second acquisition unit includes:

[0142] A first construction subunit is used to extract data corresponding to the first key early warning indicator from the target monitoring data after data cleaning, and construct a first feature set matrix;

[0143] The second construction subunit is used to construct the first new feature data based on the first feature set matrix by using a principal component analysis method.

[0144] Preferably, the second construction subunit comprises:

[0145] A second calculation module, used to calculate a first covariance matrix using the first feature set matrix;

[0146] A third calculation module, used to calculate the cumulative contribution rate of the first p features in the first covariance matrix;

[0147] A third construction module is used to construct the first new feature data using the first p features in the first covariance matrix when the cumulative contribution rate of the first p features is greater than or equal to the expected value of the cumulative contribution rate;

[0148] Where p is a positive integer, p≤k, and k is the total number of features in the first covariance matrix.

[0149] Preferably, the adjustment unit is specifically used for:

[0150] If the warning level is the fourth warning level or the number of electric vehicles entering the distribution station area reaches the first percentage of the charging station capacity, the second data collection time interval is the first multiple of the first data collection time interval;

[0151] If the warning level is the third warning level or the number of electric vehicles entering the distribution station area reaches the second percentage of the charging station capacity, the second data collection time interval is the second multiple of the first data collection time interval;

[0152] If the warning level is lower than or equal to the second warning level, and the number of electric vehicles entering the distribution station area is less than the second percentage of the charging station capacity, the second data collection time interval is the third multiple of the first data collection time interval;

[0153] If the warning level is lower than or equal to the first warning level, and the number of electric vehicles entering the distribution station area is less than the second percentage of the charging station capacity, the second data collection time interval is the first data collection time interval.

[0154] Preferably, the power distribution station area is a power distribution station area including an electric vehicle charging station.

[0155] According to a third aspect of an embodiment of the present application, there is provided an electronic device, comprising: at least one processor and a memory; the memory and the processor are connected via a bus;

[0156] The memory is used to store one or more programs;

[0157] When the one or more programs are executed by the at least one processor, the distribution station area risk classification warning method is implemented.

[0158] According to a fourth aspect of an embodiment of the present application, a readable storage medium is provided, on which an execution program is stored. When the execution program is executed, the distribution station area risk classification warning method is implemented.

[0159] The technical solution provided by the present invention has the following beneficial effects:

[0160] The present invention provides a distribution station area risk classification warning method and system, which extracts features from target monitoring data of the distribution station area collected based on a first data collection time interval to obtain first area data features, performs correlation analysis on preset feature indicators based on the first area data features to obtain first key warning indicators, obtains first new feature data by using the first key warning indicators to fuse feature indicators, obtains warning levels of the distribution station area by using warning model prediction based on the first new feature data, adjusts the first data collection time interval according to the warning level of the distribution station area to obtain a second data collection time interval, re-collects target monitoring data based on the second data collection time interval, and re-acquires the warning level of the distribution station area according to the re-collected target monitoring data, which not only realizes flexible adjustment of the collection interval of the distribution station area, reduces the pressure on the data collection resources of the distribution station area, reduces the task intensity of data transmission and reading, but also improves the accuracy and reliability of the risk warning of the distribution station area, improves the convenience of the power supply of the distribution station area including the electric vehicle charging station to serve the transportation energy, and thus improves the safe and stable attributes of the operation of the distribution network system. BRIEF DESCRIPTION OF THE DRAWINGS

[0161] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. 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 creative work.

[0162] Figure 1 It is a flow chart of a distribution station area risk classification early warning method provided by an embodiment of the present invention;

[0163] Figure 2 It is a structural block diagram of a risk classification warning system for distribution station areas provided by an embodiment of the present invention;

[0164] Figure 3 It is a structural block diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0165] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the following embodiments are 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 ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0166] Embodiment 1

[0167] The present invention provides a distribution station area risk classification early warning method, such as Figure 1 As shown, the following steps are included:

[0168] Step 11: extracting features from the target monitoring data of the distribution station area collected based on the first data collection time interval to obtain data features of the first station area;

[0169] Step 12: Based on the data characteristics of the first substation area, a correlation analysis is performed on the preset characteristic indicators to obtain the first key early warning indicator;

[0170] Step 13: Use the first key early warning indicator to fuse the feature indicators to obtain the first new feature data;

[0171] Step 14: Based on the first new feature data, the warning level of the distribution station area is predicted by using the warning model;

[0172] Step 15: According to the warning level of the distribution station area, the first data collection time interval is adjusted to obtain a second data collection time interval;

[0173] Step 16: re-collecting the target monitoring data based on the second data collection time interval, and re-obtaining the warning level of the distribution station area according to the re-collected target monitoring data.

[0174] In some embodiments, after obtaining the warning information, ie, the warning level of the distribution station area, the warning information may be stored as a reference for handling.

[0175] Furthermore, the distribution station area is a distribution station area including an electric vehicle charging station.

[0176] The present invention realizes more efficient monitoring and more accurate early warning of the operating status of the distribution station area containing the electric vehicle charging station, thereby improving the energy supply level and reliability of the distribution station area where the charging station is located.

[0177] Further, step 11 includes:

[0178] Step 111: Cleaning the target monitoring data;

[0179] Step 112: extracting data corresponding to preset characteristic indicators from the target monitoring data after data cleaning to obtain the data characteristics of the first station area.

[0180] Further, step 12 includes:

[0181] Step 121: Based on the data characteristics of the first substation, a grayscale correlation analysis method is used to perform correlation analysis on the preset characteristic indicators to obtain the first key warning indicator.

[0182] In some embodiments, the first key warning indicator may include, but is not limited to: severe overload of the substation, three-phase voltage imbalance, voltage deviation, and frequency deviation.

[0183] Further, step 121 includes:

[0184] Step 1211: construct a first comparison sequence using the data features of the first station area;

[0185] Step 1212: construct a first reference number series using the collected first historical monitoring data;

[0186] Step 1213: performing standardization processing on both the first comparison number sequence and the first reference number sequence;

[0187] Step 1214: Calculate the absolute difference between the first comparison series after the normalization processing and the first reference series after the normalization processing corresponding to each preset characteristic index;

[0188] Step 1215: using the absolute difference, determining the grayscale correlation between the first comparison series after the normalization processing and the first reference series after the normalization processing corresponding to each preset characteristic index;

[0189] Step 1216: Select a preset characteristic indicator corresponding to the grayscale correlation degree that meets the preset conditions as the first key warning indicator.

[0190] It should be noted that the present invention does not limit the "preset conditions", which can be set by those skilled in the art according to test data, expert experience or engineering needs. In some embodiments, the preset conditions can be, but are not limited to: arranging the grayscale correlation degrees in order from large to small, and selecting the characteristic indicator with the largest grayscale correlation degree as the first key warning indicator.

[0191] Further, step 1212 includes:

[0192] Step 1212a: performing data cleaning on the first historical monitoring data;

[0193] Step 1212b: performing standardization processing on the first historical monitoring data after data cleaning corresponding to the preset characteristic index, to obtain the first historical monitoring data after standardization processing corresponding to the preset characteristic index;

[0194] Step 1212c: The minimum value of the first historical monitoring data after the standardization processing corresponding to each preset characteristic index is set as the optimal value, and the maximum value of the first historical monitoring data after the standardization processing corresponding to each preset characteristic index is set as the worst value;

[0195] Step 1212d: construct a first reference sequence using the optimal value and the worst value corresponding to the preset characteristic index.

[0196] Further, step 1215 includes:

[0197] Step 1215a: Determine the maximum absolute difference and the minimum absolute difference between the first comparison number series after the normalization processing and the first reference number series after the normalization processing corresponding to each preset characteristic index according to the absolute difference;

[0198] Step 1215b: using the maximum absolute difference and the minimum absolute difference, calculate the grayscale correlation between the first comparison series after the normalization processing and the first reference series after the normalization processing corresponding to each preset characteristic index.

[0199] Further, step 13 includes:

[0200] Step 131: extracting data corresponding to the first key early warning indicator from the target monitoring data after data cleaning, and constructing a first feature set matrix;

[0201] Step 132: Based on the first feature set matrix, a principal component analysis method is used to construct first new feature data.

[0202] Further, step 132 includes:

[0203] Step 1321: using the first feature set matrix, calculate and obtain a first covariance matrix;

[0204] Specifically, the calculation formula of the first covariance matrix includes:

[0205] B=X T X

[0206] In the above formula, X is the first feature set matrix and T is the transpose;

[0207] Step 1322: Calculate the cumulative contribution rate of the first p features in the first covariance matrix;

[0208] Specifically, the calculation formula for the cumulative contribution rate of the first p features in the first covariance matrix includes:

[0209]

[0210] In the above formula, X is the first feature set matrix, T is the transpose, and B is the first covariance matrix; , k is the total number of features in the first covariance matrix, p≤k; is the cumulative contribution rate of the first p features in the first covariance matrix, is the eigenvalue of the i-th feature in the first covariance matrix;

[0211] Step 1323: when the cumulative contribution rate of the first p features is greater than or equal to the expected value of the cumulative contribution rate, construct the first new feature data using the first p features in the first covariance matrix;

[0212] Where p is a positive integer, p≤k, and k is the total number of features in the first covariance matrix.

[0213] Further, step 15 includes:

[0214] Step 151: If the warning level is the fourth warning level or the number of electric vehicles entering the distribution station area reaches the first percentage of the charging station capacity, the second data collection time interval is the first multiple of the first data collection time interval;

[0215] Step 152: If the warning level is the third warning level or the number of electric vehicles entering the distribution station area reaches the second percentage of the charging station capacity, the second data collection time interval is the second multiple of the first data collection time interval;

[0216] Step 153: If the warning level is lower than or equal to the second warning level, and the number of electric vehicles entering the distribution station area is less than the second percentage of the charging station capacity, the second data collection time interval is the third multiple of the first data collection time interval;

[0217] Step 154: If the warning level is lower than or equal to the first warning level, and the number of electric vehicles entering the distribution station area is less than the second percentage of the charging station capacity, the second data collection time interval is the first data collection time interval.

[0218] For example, suppose the target monitoring data collection time interval is t 0, the first warning level is the green warning level, the second warning level is the blue warning level, the third warning level is the orange warning level, and the fourth warning level is the red warning level;

[0219] If the warning level reaches the red warning level, or the number of electric vehicles entering the charging station reaches 90% of the charging station capacity, the collection time interval of the target monitoring data will be adjusted to t 0 / 10.

[0220] If the warning level reaches the orange warning level, or the number of electric vehicles entering the charging station reaches 80% of the charging station capacity, the collection time interval of the target monitoring data will be adjusted to t 0 / 8.

[0221] If the warning levels are not higher than the blue warning level, and the number of electric vehicles entering the charging station is less than 80% of the charging station capacity, the collection time interval of the target monitoring data will be adjusted to t 0 / 5.

[0222] If the warning level is not higher than the green warning level, and the number of electric vehicles entering the charging station is less than 80% of the charging station capacity, the target monitoring data collection time interval is maintained at t 0.

[0223] The risk classification and early warning method for distribution station areas based on adjustment of data monitoring time intervals provided by the present invention performs operation status analysis and early warning based on monitoring data of distribution station areas containing electric vehicle charging stations and historical data, and flexibly adjusts the area collection time interval based on the early warning results and collection time adjustment criteria, thereby realizing flexible monitoring and graded early warning under different operating conditions in the operation of distribution station areas containing electric vehicle charging stations.

[0224] Furthermore, the method further comprises: Step 10: establishing an early warning model;

[0225] Step 10 includes:

[0226] Step 101: Collect the second historical monitoring data of the distribution station area;

[0227] Step 102: extracting features from the second historical monitoring data to obtain features of the second area data;

[0228] Step 103: Based on the data characteristics of the second substation area, a correlation analysis is performed on the preset characteristic indicators to obtain a second key early warning indicator;

[0229] Step 104: using the second key early warning indicator to fuse the characteristic indicators to obtain second new characteristic data;

[0230] Step 105: Determine the historical warning level of the distribution station area corresponding to the second new characteristic data;

[0231] Step 106: construct a data set using the second new feature data and the historical warning level of the distribution station area, and use the data set to train the long-short neural network model to obtain a warning model.

[0232] Further, step 102 includes:

[0233] Step 1021: performing data cleaning on the second historical monitoring data;

[0234] Step 1022: extract data corresponding to the preset characteristic index from the second historical monitoring data after data cleaning to obtain the second station area data characteristics.

[0235] Further, step 103 includes:

[0236] Step 1031: Based on the data characteristics of the second substation area, a grayscale correlation analysis method is used to perform correlation analysis on the preset characteristic indicators to obtain a second key warning indicator.

[0237] Further, step 1031 includes:

[0238] Step 1031a: construct a second comparison sequence using the data features of the second station area;

[0239] Step 1031b: constructing a second reference series using the collected third historical monitoring data;

[0240] Step 1031c: performing standardization processing on both the second comparison number sequence and the second reference number sequence;

[0241] Step 1031d: Calculate the absolute difference between the second comparison series after the normalization processing and the second reference series after the normalization processing corresponding to each preset characteristic index;

[0242] Step 1031e: using the absolute difference, determining the grayscale correlation between the second comparison series after the normalization processing and the second reference series after the normalization processing corresponding to each preset characteristic index;

[0243] Step 1031f: Select the characteristic index corresponding to the grayscale correlation degree that meets the preset conditions as the second key warning index;

[0244] Among them, the collection time of the third historical monitoring data is earlier than the collection time of the second historical monitoring data.

[0245] It should be noted that the present invention does not limit the "preset conditions", which can be set by those skilled in the art according to test data, expert experience or engineering needs. In some embodiments, the preset conditions can be, but are not limited to: arranging the grayscale correlation degrees in order from large to small, and selecting the characteristic indicator with the largest grayscale correlation degree as the second key warning indicator.

[0246] Further, step 1031b includes:

[0247] Perform data cleaning on the third historical monitoring data;

[0248] Performing standardization processing on the third historical monitoring data after data cleaning corresponding to the preset characteristic index to obtain the third historical monitoring data after standardization processing corresponding to the preset characteristic index;

[0249] Let the minimum value of the third historical monitoring data after the standardization processing corresponding to each preset characteristic index be the optimal value, and the maximum value of the third historical monitoring data after the standardization processing corresponding to each preset characteristic index be the worst value;

[0250] The second reference sequence is constructed using the optimal value and the worst value corresponding to the preset characteristic index.

[0251] Further, step 1031e includes:

[0252] Determine the maximum absolute difference and the minimum absolute difference between the second comparison series after the normalization processing and the second reference series after the normalization processing corresponding to each preset characteristic index according to the absolute difference;

[0253] The grayscale correlation between the standardized second comparison series and the standardized second reference series corresponding to each preset characteristic index is calculated by using the maximum absolute difference and the minimum absolute difference.

[0254] Further, step 104 includes:

[0255] Step 1041: extracting data corresponding to the second key early warning indicator from the second historical monitoring data after data cleaning, and constructing a second feature set matrix;

[0256] Step 1042: Based on the second feature set matrix, a principal component analysis method is used to construct second new feature data.

[0257] Further, step 1042 includes:

[0258] Step 1042a: using the second feature set matrix, calculate and obtain a second covariance matrix;

[0259] Specifically, the calculation formula of the second covariance matrix includes:

[0260] B'=X' T X'

[0261] In the above formula, X' is the second feature set matrix, and T is the transpose;

[0262] Step 1042b: Calculate the cumulative contribution rate of the first p' features in the second covariance matrix;

[0263] Specifically, the calculation formula for the cumulative contribution rate of the first p' features in the second covariance matrix includes:

[0264]

[0265] In the above formula, X' is the second feature set matrix, T is the transpose, and B' is the second covariance matrix; i'∈ [ 1,k' ], k' is the total number of features in the second covariance matrix, p'≤k'; is the cumulative contribution rate of the first p' features in the second covariance matrix, is the eigenvalue of the i'th feature in the second covariance matrix;

[0266] Step 1042c: when the cumulative contribution rate of the first p' features is greater than or equal to the expected value of the cumulative contribution rate, construct the second new feature data using the first p' features in the second covariance matrix;

[0267] Wherein, p' is a positive integer, p'≤k', and k' is the total number of features in the second covariance matrix.

[0268] Further, step 105 includes:

[0269] Step 1051: extracting valid data from the second new feature data according to the preset standard limit value of the normal operation of the substation and the preset standard limit value of the fault operation of the substation corresponding to the second key warning indicator, and obtaining the processed second new feature data;

[0270] Step 1052: Based on the second key warning indicator, a K-means clustering method is used to perform feature clustering on the processed second new feature data to obtain clustered second new feature data;

[0271] Step 1053: Based on the clustered second new feature data, an interval prediction method is used to analyze and obtain the change extreme value corresponding to the second key warning indicator;

[0272] For example, assuming that the key warning indicators are heavy overload, three-phase voltage imbalance, voltage deviation and frequency deviation, the valid data in the second new feature data can be extracted according to the national standard limit values ​​of the normal operation and fault operation of the substation corresponding to these four key warning indicators, and the processed second new feature data can be obtained. The K-means clustering method is used to perform feature clustering on the warning objects (i.e., heavy overload, three-phase voltage imbalance, voltage deviation, and frequency deviation), and the indicator changes and their extreme values ​​are analyzed through interval prediction.

[0273] Step 1054: Based on the clustered second new feature data and the extreme value of change corresponding to the second key warning indicator, the Bootstrap algorithm is used to predict the sample variance of the Bootstrap sample corresponding to the second key warning indicator;

[0274] Step 1055: using the sample variance corresponding to the second key warning indicator, calculate the upper limit value and the lower limit value of the second key warning indicator;

[0275] Specifically, the calculation formula for the upper limit value of the second key warning indicator includes:

[0276]

[0277] The calculation formula for the lower limit value of the second key warning indicator includes:

[0278]

[0279] In the above formula, is the upper limit value of the second key warning indicator, is the lower limit of the second key warning indicator, is the mean value of the second new feature data after clustering corresponding to the second key warning indicator, is the sample variance corresponding to the second key early warning indicator, is a standard normal distribution 100·(1- α )% corresponding critical value, α is the adjustment factor.

[0280] Step 1056: Determine the historical warning level of the distribution station area corresponding to the second new characteristic data according to the upper limit value and the lower limit value of the second key warning indicator.

[0281] It should be noted that the method of "using the Bootstrap algorithm to predict the sample variance of the Bootstrap sample corresponding to the second key warning indicator" involved in the embodiments of the present invention is well known to those skilled in the art, so its specific implementation method will not be described in detail. In some embodiments, the calculation formula of the sample variance includes:

[0282]

[0283]

[0284] In the above formula, , B is the total number of Bootstrap samples; is the sample variance, is the b-th Bootstrap sample, is the sample mean.

[0285] Further, step 1056 includes:

[0286] Step 1056a: When the upper limit value and the lower limit value of the second key warning indicator belong to the fluctuation range corresponding to the first confidence threshold, the historical warning level of the distribution station area is the first warning level;

[0287] Step 1056b: When the upper limit value and the lower limit value of the second key warning indicator belong to the fluctuation range corresponding to the second confidence threshold, the historical warning level of the distribution station area is the second warning level;

[0288] Step 1056c: When the upper limit value and the lower limit value of the second key warning indicator belong to the fluctuation range corresponding to the third confidence threshold, the historical warning level of the distribution station area is the third warning level;

[0289] Step 1056d: When the upper limit value and the lower limit value of the second key warning indicator belong to the fluctuation range corresponding to the fourth confidence threshold, the historical warning level of the distribution station area is the fourth warning level.

[0290] It should be noted that the present invention does not limit the "first confidence threshold, the second confidence threshold, the third confidence threshold and the fourth confidence threshold", which can be set by technical personnel in this field according to test data, expert experience or engineering needs.

[0291] For example, the 98% confidence interval is defined as Level 1 (i.e. the first warning level), the 95% confidence interval is defined as Level 2 (i.e. the second warning level), the 90% confidence interval is defined as Level 3 (i.e. the third warning level), and the 86% confidence interval is defined as Level 4 (i.e. the fourth warning level). Then, according to the severity of the risk warning, it is divided into four warning levels from low to high, namely green (Level 1), blue (Level 2), orange (Level 3) and red (Level 4), thus completing the warning level division.

[0292] Furthermore, in step 106, the long-short neural network model is trained using the data set to obtain an early warning model, including:

[0293] Step 1061: Divide the data set into a training set, a validation set, and a test set;

[0294] Step 1062: training the long-short neural network model using the training set to obtain a trained long-short neural network model;

[0295] Step 1063: Use the validation set to validate the trained long and short neural network model. If the validation is successful, use the test set to evaluate the performance of the trained long and short neural network model. If the validation fails, adjust the hyperparameters of the long and short neural network model and monitor overfitting, and retrain the long and short neural network model until the validation is successful.

[0296] Step 1064: Use the test set to perform performance evaluation on the trained long-short neural network model. If the performance evaluation of the trained long-short neural network model is successful, the trained long-short neural network model is an early warning model. If the performance evaluation of the trained long-short neural network model fails, re-collect the second historical monitoring data, and use the re-collected second historical monitoring data to establish an early warning model.

[0297] It is understandable that the failure of the performance evaluation of the trained long-short neural network model indicates that the characteristics of the second warning indicator in the trained long-short neural network model have changed. For example, if an early warning model is established based on the second historical monitoring data of 2020, and the performance evaluation of the long-short neural network model fails after training, it indicates that the characteristics of the second early warning indicator of the substation may have changed due to changes in the source load level or topological architecture, and the second historical monitoring data needs to be recollected. However, the recollected second historical monitoring data cannot be from 2020, but can be, but not limited to, data from 2021, thereby re-establishing a new early warning model that characterizes the characteristics of the second early warning indicator.

[0298] It should be noted that the method of "using a test set to perform performance evaluation on the trained long and short neural network model" involved in the embodiment of the present invention is well known to those skilled in the art, and therefore, its specific implementation method will not be described in detail.

[0299] In some embodiments, the statistical operation data of nearly three years are divided into training set, validation set and test set by year. The first year's data is used as a training set to train the neural network model; the second year's data is used as a validation set to adjust the model's hyperparameters and monitor overfitting; the third year's data is used as a test set to evaluate the model performance; based on the aforementioned new features as historical data input features, the valid historical data is trained, and the risk prediction results of each indicator in the next period are output to determine the charging safety risk and distribution network risk; when the warning value is greater than the warning boundary value, a warning is issued; if it is less than the boundary value, the system is in normal operation and no warning is issued, thereby completing the graded warning.

[0300] To further illustrate the above-mentioned distribution station area risk classification early warning method, the present invention provides a specific example, including the following steps:

[0301] S100: The distribution station area that needs to be monitored and warned is monitored at an initial time interval t 0 is the data collection time interval of the transformer in the substation. Data is collected for the monitoring and early warning substation to obtain the target monitoring data. The target monitoring data may include but is not limited to: voltage, current and power of each substation, equipment status (for example, whether the transformer is operating normally), fault recording, electric vehicle charging power and number of connected electric vehicles N ev Among them, the distribution station area that needs to be monitored and warned includes electric vehicle charging stations.

[0302] S200: preprocessing the target monitoring data and extracting the data features of the first station area;

[0303] Specifically, the preprocessing of target monitoring data includes: using a generative adversarial network for a small amount of missing data, constructing generator and discriminator networks in sequence, defining the data completion loss function, adversarial network training, missing value generation, and generated value consistency evaluation to complete the missing data of the data set;

[0304] The first area data feature extraction includes but is not limited to: overall form, fluctuation, change trend, correlation, date type, voltage deviation, power factor, three-phase unbalance, harmonic current and other features;

[0305] Furthermore, for the overall form of the data in the substation area, the daily and monthly electricity consumption rate is used to reflect the overall change in electricity consumption, the peak-to-valley difference rate reflects the magnitude of the change in electricity consumption, and the quarterly electricity consumption ratio reflects the distribution of electricity consumption;

[0306] In view of the power fluctuation of the area, the power consumption dispersion coefficient is used to indicate the degree of deviation of the power consumption curve of the charging area relative to the average value curve, so as to reflect the fluctuation of the power consumption of the charging area. The difference between the first and last values ​​of the monthly power consumption data is used as the basis to reflect the overall fluctuation of the power consumption of the charging area.

[0307] According to the trend characteristics of the area data m t , using the simple moving average method, the daily power consumption sequence of the charging station area is calculated according to the following formula:

[0308]

[0309] In the above formula, n is the number of data in the time period, t is the time, m t-1 is the trend characteristic of the area data at the t-1th moment, m t-2 is the trend characteristic of the area data at the t-2th moment, m t-n is the trend characteristic of the area data at the tnth moment; S t is the daily power consumption of the charging station area at time t;

[0310] Compare each charging station area one by one t Time series M t and sequence S t ,Will M t Greater than S t The point is recorded as: a 1 ,a 2 ,…,a e ,…,a u}, that is, record the data points where the actual value is larger than the predicted value, where a 1 is the first M t Greater than S t point, a 2 is the second M t Greater than S t point, a e For the e indivual M t Greater than S t point, a u For the u indivual M t Greater than S t point; M t Less than S t The point is recorded as: b 1 , b 2 ,…, b w ,…,b v}, that is, record the data points where the actual value is smaller than the predicted value, where b 1 is the first M t Less than S t point, b 2 is the second M t Less than S t point, b w For the w indivual M t Less than S t point, b v For the v indivual M t Less than S t point; then, the upward trend is calculated according to the following formula tra and a downward trend trb :

[0311]

[0312]

[0313] In the above formula, u for M t Greater than S t The total number of points; v for M t Less than S t The total number of points.

[0314] S300: Analyze the correlation of the above characteristic indicators, select key characteristics of the area data and perform characteristic indicator fusion:

[0315] The correlation degree calculation adopts the grayscale correlation analysis method, uses the data characteristics of the first substation to construct the first comparison series, and uses the collected second historical monitoring data to construct the first reference series; both the first comparison series and the first reference series are standardized to eliminate the dimension difference; the absolute difference between the first comparison series after standardization and the first reference series after standardization corresponding to each preset characteristic index is calculated, and the maximum value and the minimum value in the absolute difference are listed to obtain the maximum absolute difference and the minimum absolute difference; the grayscale correlation degree between the first comparison series after standardization and the first reference series after standardization corresponding to each preset characteristic index is calculated using the maximum absolute difference and the minimum absolute difference, and the correlation coefficient between the series can be obtained by the average value calculation method;

[0316] Arrange the grayscale correlation from large to small, select the characteristic indicator with the largest correlation as the first key warning indicator, and obtain the first key warning indicator including: heavy overload in the substation, three-phase voltage imbalance, voltage deviation, and frequency deviation for graded warning;

[0317] The first feature set matrix formed by the first key early warning indicator is set to X (it can be understood that each row of the feature set matrix is ​​a record), and the first new feature data is constructed by using the principal component analysis method;

[0318] In some embodiments, constructing the first new feature data using the principal component analysis method includes: constructing a central matrix with a column mean of 0, and calculating the eigenvalues ​​of the covariance matrix B according to the following three formulas { λ 1, λ 2, λ 3,..., λ k}, and find the corresponding standard orthogonal vector {y 1 ,y 2 ,y 3 ,... ,y k}:

[0319]

[0320] In the above formula, λ 1 is the first eigenvalue, λ 2 is the second eigenvalue, λ 3 is the third eigenvalue, λ k is the kth eigenvalue, V is the standard orthogonal vector of the covariance matrix B, T is the transpose, is the eigenvalue of the covariance matrix B, y 1 for λ 1 is the standard orthogonal vector, y 2 for λ 2 is the standard orthogonal vector, y 3 for λ 3 standard orthogonal vectors, y k for λ k The standard orthogonal vectors of ;

[0321] According to the following formula, the front P The cumulative contribution rate of new features, setting the expected value of the cumulative contribution rate a 0, let a p > a 0 determines the number of new features, and the new features are :

[0322]

[0323] In the above formula, , k is the total number of features in the first covariance matrix, p≤k; is the cumulative contribution rate of the first p features in the first covariance matrix, is the eigenvalue of the i-th feature in the first covariance matrix; y j for λ j The standard orthogonal vectors of λ j is the jth eigenvalue.

[0324] S400: using the first new feature data as an input of the warning model, and outputting the predicted warning level of the distribution station area;

[0325] S500: Combine the warning levels of the four first key warning indicators of heavy overload, three-phase voltage imbalance, voltage deviation and frequency deviation with the number of electric vehicles entering the site, adjust the judgment criteria according to the data collection time interval, and adjust the data collection time interval t 0 for adjustment;

[0326] The criteria for adjusting the data collection time interval are as follows:

[0327] 1) If one of the four first key warning indicators reaches the red warning level, or the number of electric vehicles entering the charging station reaches 90% of the charging station capacity, the data collection interval will be adjusted to t 0 / 10.

[0328] 2) If one of the four first key warning indicators reaches the orange warning level, or the number of electric vehicles entering the charging station reaches 80% of the charging station capacity, the data collection interval will be adjusted to t 0 / 8.

[0329] 3) If the four first key warning indicators are not higher than the blue warning level, and the number of electric vehicles entering the charging station is less than 80% of the charging station capacity, the data collection interval is adjusted to t 0 / 5.

[0330] 4) If the four first key warning indicators are not higher than the green warning level, and the number of electric vehicles entering the charging station is less than 80% of the charging station capacity, the data collection time interval is maintained at t 0.

[0331] S600: Send the updated data collection time interval to the substation data collection terminal, adjust the substation data collection frequency, and store the warning information as a reference for handling.

[0332] The present invention provides a distribution station area risk classification warning method, which adopts data feature analysis and feature indicator correlation analysis methods to calculate the substation operation monitoring data and historical data, obtain the operation status classification warning threshold, and realize a more detailed evaluation of the operation status of the distribution station area; the present invention proposes a collection time adjustment criterion based on the operation status analysis and the number of electric vehicles, and flexibly adjusts the collection interval of the distribution station area, thereby reducing the pressure on the data collection resources of the distribution station area, and at the same time reduces the task intensity of data transmission and reading, and improves the accuracy and reliability of the risk warning of the distribution station area; the distribution station area containing the electric vehicle charging station in the present invention improves the convenience of electric energy serving transportation energy consumption, and the engineering practice is increasing. The present invention can be applied to the field of distribution network operation monitoring to improve the safe and stable attributes of the operation of the distribution network system.

[0333] Embodiment 2

[0334] The present invention also provides a distribution station area risk classification early warning system, such as Figure 2 As shown, including:

[0335] An extraction unit, used for extracting features from the target monitoring data of the distribution station area collected based on the first data collection time interval to obtain data features of the first station area;

[0336] A first acquisition unit is used to perform a correlation analysis on a preset characteristic index based on the data characteristics of the first station area to obtain a first key early warning index;

[0337] A second acquisition unit is used to use the first key warning indicator to fuse the characteristic indicator to obtain first new characteristic data;

[0338] A third acquisition unit is used to predict the warning level of the distribution station area using the warning model based on the first new feature data;

[0339] An adjustment unit, configured to adjust the first data collection time interval according to the warning level of the distribution station area to obtain a second data collection time interval;

[0340] The fourth acquisition unit is used to re-collect the target monitoring data based on the second data collection time interval, and re-acquire the warning level of the distribution station area according to the re-collected target monitoring data.

[0341] Furthermore, the extraction unit comprises:

[0342] A data cleaning subunit is used to clean the target monitoring data;

[0343] The first acquisition subunit is used to extract data corresponding to preset characteristic indicators from the target monitoring data after data cleaning to obtain the first station area data characteristics.

[0344] Furthermore, the first acquisition unit includes:

[0345] The second acquisition subunit is used to perform correlation analysis on preset characteristic indicators based on the data characteristics of the first substation area by using a grayscale correlation analysis method to obtain a first key warning indicator.

[0346] Furthermore, the second acquisition subunit includes:

[0347] A first construction module, used to construct a first comparison sequence using the data characteristics of the first station area;

[0348] A second construction module is used to construct a first reference number series using the collected first historical monitoring data;

[0349] A processing module, used for performing standardization processing on both the first comparison number series and the first reference number series;

[0350] A first calculation module, used for calculating the absolute difference between the first comparison series after standardization and the first reference series after standardization corresponding to each preset characteristic index;

[0351] A determination module, used to determine the grayscale correlation between the first comparison series after standardization and the first reference series after standardization corresponding to each preset characteristic index by using the absolute difference;

[0352] The selection module is used to select a preset characteristic indicator corresponding to the grayscale correlation degree that meets the preset conditions as the first key warning indicator.

[0353] Furthermore, the second building block is specifically used for:

[0354] Performing data cleaning on the first historical monitoring data;

[0355] Performing standardization processing on the first historical monitoring data after data cleaning corresponding to the preset characteristic index to obtain the first historical monitoring data after standardization processing corresponding to the preset characteristic index;

[0356] Let the minimum value of the first historical monitoring data after the standardization processing corresponding to each preset characteristic index be the optimal value, and the maximum value of the first historical monitoring data after the standardization processing corresponding to each preset characteristic index be the worst value;

[0357] The first reference sequence is constructed using the optimal value and the worst value corresponding to the preset characteristic index.

[0358] Further, the module is determined, specifically for:

[0359] Determine the maximum absolute difference and the minimum absolute difference between the first comparison number series after the normalization processing and the first reference number series after the normalization processing corresponding to each preset characteristic index according to the absolute difference;

[0360] The grayscale correlation between the first comparison series after the normalization processing and the first reference series after the normalization processing corresponding to each preset characteristic index is calculated by using the maximum absolute difference and the minimum absolute difference.

[0361] Furthermore, the second acquisition unit includes:

[0362] A first construction subunit is used to extract data corresponding to the first key early warning indicator from the target monitoring data after data cleaning, and construct a first feature set matrix;

[0363] The second construction subunit is used to construct the first new feature data based on the first feature set matrix by adopting the principal component analysis method.

[0364] Furthermore, the second construction subunit includes:

[0365] A second calculation module is used to calculate a first covariance matrix using the first feature set matrix;

[0366] The third calculation module is used to calculate the cumulative contribution rate of the first p features in the first covariance matrix;

[0367] A third construction module is used to construct first new feature data using the first p features in the first covariance matrix when the cumulative contribution rate of the current p features is greater than or equal to the expected value of the cumulative contribution rate;

[0368] Where p is a positive integer, p≤k, and k is the total number of features in the first covariance matrix.

[0369] Furthermore, the adjustment unit is specifically used for:

[0370] If the warning level is the fourth warning level or the number of electric vehicles entering the distribution station area reaches the first percentage of the charging station capacity, the second data collection time interval is the first multiple of the first data collection time interval;

[0371] If the warning level is the third warning level or the number of electric vehicles entering the distribution station area reaches the second percentage of the charging station capacity, the second data collection time interval is the second multiple of the first data collection time interval;

[0372] If the warning level is lower than or equal to the second warning level, and the number of electric vehicles entering the distribution station area is less than the second percentage of the charging station capacity, the second data collection time interval is the third multiple of the first data collection time interval;

[0373] If the warning level is lower than or equal to the first warning level, and the number of electric vehicles entering the distribution station area is less than the second percentage of the charging station capacity, the second data collection time interval is the first data collection time interval.

[0374] Furthermore, the device also includes: an establishment unit for establishing an early warning model; the establishment unit includes:

[0375] A collection subunit, used for collecting the second historical monitoring data of the distribution station area;

[0376] The third acquisition subunit is used to extract features from the second historical monitoring data to obtain data features of the second station area;

[0377] The fourth acquisition subunit is used to perform a correlation analysis on the preset characteristic indicators based on the data characteristics of the second station area to obtain a second key early warning indicator;

[0378] A fifth acquisition subunit is used to use the second key early warning indicator to fuse the characteristic indicators to obtain second new characteristic data;

[0379] A determination subunit, used to determine the historical warning level of the distribution station area corresponding to the second new characteristic data;

[0380] The sixth acquisition subunit is used to construct a data set using the second new feature data and the historical warning level of the distribution station area, and use the data set to train the long-short neural network model to obtain a warning model.

[0381] Furthermore, the third acquisition subunit includes:

[0382] A data cleaning module, used for cleaning the second historical monitoring data;

[0383] The first acquisition module is used to extract data corresponding to the preset characteristic index from the second historical monitoring data after data cleaning to obtain the second station area data characteristics.

[0384] Furthermore, the fourth acquisition subunit includes:

[0385] The second acquisition module is used to perform correlation analysis on preset characteristic indicators based on the data characteristics of the second substation area by using a grayscale correlation analysis method to obtain a second key early warning indicator.

[0386] Furthermore, the second acquisition module includes:

[0387] A first construction submodule is used to construct a second comparison sequence using the data characteristics of the second station area;

[0388] A second construction submodule, used to construct a second reference number series using the collected third historical monitoring data;

[0389] A standardization submodule, used for performing standardization processing on both the second comparison number series and the second reference number series;

[0390] A calculation submodule, used for calculating the absolute difference between the second comparison series after the normalization processing and the second reference series after the normalization processing corresponding to each preset characteristic index;

[0391] A determination submodule, used to determine the grayscale correlation between the second comparison series after standardization and the second reference series after standardization corresponding to each preset characteristic index by using the absolute difference;

[0392] A selection submodule is used to select a characteristic indicator corresponding to a grayscale correlation degree that meets a preset condition as a second key early warning indicator;

[0393] Among them, the collection time of the third historical monitoring data is earlier than the collection time of the second historical monitoring data.

[0394] Furthermore, the second construction submodule is specifically used for:

[0395] Perform data cleaning on the third historical monitoring data;

[0396] Performing standardization processing on the third historical monitoring data after data cleaning corresponding to the preset characteristic index to obtain the third historical monitoring data after standardization processing corresponding to the preset characteristic index;

[0397] Let the minimum value of the third historical monitoring data after the standardization processing corresponding to each preset characteristic index be the optimal value, and the maximum value of the third historical monitoring data after the standardization processing corresponding to each preset characteristic index be the worst value;

[0398] The second reference sequence is constructed using the optimal value and the worst value corresponding to the preset characteristic index.

[0399] Further, submodules are determined, specifically for:

[0400] Determine the maximum absolute difference and the minimum absolute difference between the second comparison series after the normalization processing and the second reference series after the normalization processing corresponding to each preset characteristic index according to the absolute difference;

[0401] The grayscale correlation between the standardized second comparison series and the standardized second reference series corresponding to each preset characteristic index is calculated by using the maximum absolute difference and the minimum absolute difference.

[0402] Furthermore, the fifth acquisition subunit includes:

[0403] The third construction submodule is used to extract data corresponding to the second key early warning indicator from the second historical monitoring data after data cleaning, and construct a second feature set matrix;

[0404] The fourth construction submodule is used to construct the second new feature data based on the second feature set matrix by using the principal component analysis method.

[0405] Furthermore, the fourth construction submodule is specifically used for:

[0406] Using the second feature set matrix, a second covariance matrix is ​​calculated;

[0407] Calculate the cumulative contribution rate of the first p' features in the second covariance matrix;

[0408] When the cumulative contribution rate of the first p' features is greater than or equal to the expected value of the cumulative contribution rate, the first p' features in the second covariance matrix are used to construct the second new feature data;

[0409] Wherein, p' is a positive integer, p'≤k', and k' is the total number of features in the second covariance matrix.

[0410] Further, subunits are determined, including:

[0411] A third acquisition module is used to extract valid data from the second new feature data according to the preset standard limit value of normal operation of the substation and the preset standard limit value of fault operation of the substation corresponding to the second key warning indicator, so as to obtain the processed second new feature data;

[0412] A fourth acquisition module is used to perform feature clustering on the processed second new feature data by using a K-means clustering method based on the second key warning indicator to obtain clustered second new feature data;

[0413] A fifth acquisition module is used to analyze and obtain the change extreme value corresponding to the second key warning indicator by using an interval prediction method based on the clustered second new feature data;

[0414] A prediction module, used to predict the sample variance of the Bootstrap sample corresponding to the second key warning indicator by using the Bootstrap algorithm based on the clustered second new feature data and the change extreme value corresponding to the second key warning indicator;

[0415] A fourth calculation module, used to calculate the upper limit value and the lower limit value of the second key warning indicator by using the sample variance corresponding to the second key warning indicator;

[0416] The sixth acquisition module is used to determine the historical warning level of the distribution station area corresponding to the second new feature data according to the upper limit value and the lower limit value of the second key warning indicator.

[0417] Furthermore, the sixth acquisition module is specifically used for:

[0418] When the upper limit value and the lower limit value of the second key warning indicator belong to the fluctuation range corresponding to the first confidence threshold value, the historical warning level of the distribution station area is the first warning level;

[0419] When the upper limit value and the lower limit value of the second key warning indicator belong to the fluctuation range corresponding to the second confidence threshold, the historical warning level of the distribution station area is the second warning level;

[0420] When the upper limit value and the lower limit value of the second key warning indicator belong to the fluctuation range corresponding to the third confidence threshold value, the historical warning level of the distribution station area is the third warning level;

[0421] When the upper limit value and the lower limit value of the second key warning indicator belong to the fluctuation range corresponding to the fourth confidence threshold, the historical warning level of the distribution station area is the fourth warning level.

[0422] Furthermore, the sixth acquisition subunit is specifically used for:

[0423] Divide the dataset into training, validation and test sets;

[0424] Using the training set to train the long and short neural network model, to obtain a trained long and short neural network model;

[0425] Use the validation set to validate the trained long and short neural network model. If the validation is successful, use the test set to evaluate the performance of the trained long and short neural network model. If the validation fails, adjust the hyperparameters of the long and short neural network model and monitor overfitting, and retrain the long and short neural network model until the validation is successful.

[0426] The performance of the trained long-short neural network model is evaluated using the test set. If the performance evaluation of the trained long-short neural network model is successful, the trained long-short neural network model is an early warning model. If the performance evaluation of the trained long-short neural network model fails, the second historical monitoring data is recollected, and the early warning model is established using the recollected second historical monitoring data.

[0427] Furthermore, the calculation formula of the first covariance matrix includes:

[0428] B=X T X

[0429] The calculation formula for the cumulative contribution rate of the first p features in the first covariance matrix includes:

[0430]

[0431] In the above formula, X is the first feature set matrix, T is the transpose, and B is the first covariance matrix; , k is the total number of features in the first covariance matrix, p≤k; is the cumulative contribution rate of the first p features in the first covariance matrix, is the eigenvalue of the i-th feature in the first covariance matrix.

[0432] Furthermore, the calculation formula of the second covariance matrix includes:

[0433] B'=X' T X'

[0434] The calculation formula for the cumulative contribution rate of the first p' features in the second covariance matrix includes:

[0435]

[0436] In the above formula, X' is the second feature set matrix, T is the transpose, and B' is the second covariance matrix; i'∈ [ 1,k' ], k' is the total number of features in the second covariance matrix, p'≤k'; is the cumulative contribution rate of the first p' features in the second covariance matrix, is the eigenvalue of the i'th feature in the second covariance matrix.

[0437] Furthermore, the calculation formula for the upper limit value of the second key warning indicator includes:

[0438]

[0439] The calculation formula for the lower limit value of the second key warning indicator includes:

[0440]

[0441] In the above formula, is the upper limit value of the second key warning indicator, is the lower limit of the second key warning indicator, is the mean value of the second new feature data after clustering corresponding to the second key warning indicator, is the sample variance corresponding to the second key early warning indicator, is a standard normal distribution 100·(1- α )% corresponding critical value, α is the adjustment factor.

[0442] Furthermore, the distribution station area is a distribution station area including an electric vehicle charging station.

[0443] It can be understood that the system embodiment provided above corresponds to the method embodiment described above, and the corresponding specific contents can be referenced to each other and will not be repeated here.

[0444] It can be understood that the same or similar parts of the above embodiments can be referenced to each other, and the contents not described in detail in some embodiments can refer to the same or similar contents in other embodiments.

[0445] Embodiment 3

[0446] like Figure 3As shown, the present invention also provides an electronic device, which may be a computer device, a single-chip device, an intelligent mobile device, etc. The electronic device in this embodiment may include a processor, a memory, a transceiver component, etc. The memory, the processor, and the transceiver component are connected via a bus; the memory may be used to store an execution program, and an exemplary execution program may include instructions; the processor is used to execute the instructions stored in the memory. The memory may also be used to store data, which may be called and / or modified when the instructions are executed.

[0447] The processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. It is the computing core and control core of the terminal, which is suitable for implementing one or more instructions, specifically suitable for loading and executing one or more instructions in a storage medium to implement the corresponding method flow or corresponding functions, so as to implement the steps of a distribution station area risk classification warning method in the above embodiment.

[0448] Embodiment 4

[0449] Based on the same inventive concept, the present invention also provides a readable storage medium, specifically an electronic device readable storage medium (Memory), which is a memory device in an electronic device for storing programs and data. It can be understood that the storage medium here can include both built-in storage media in electronic devices and extended storage media supported by electronic devices. The storage medium provides a storage space, which stores the operating system of the terminal. In addition, one or more instructions suitable for being loaded and executed by a processor are also stored in the storage space, and these instructions can be one or more execution programs (including program codes). It should be noted that the storage medium here can be a high-speed RAM memory or a non-volatile memory, such as at least one disk memory. The processor loads and executes one or more instructions stored in the storage medium, which can implement the steps of a distribution station area risk classification warning method in the above embodiment.

[0450] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0451] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0452] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0453] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0454] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the relevant field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A distribution station area risk classification early warning method, characterized in that: include: Extracting features of the target monitoring data of the distribution station area collected based on the first data collection time interval to obtain data features of the first station area; Based on the data characteristics of the first substation area, a correlation analysis is performed on the preset characteristic indicators to obtain a first key early warning indicator; Using the first key early warning indicator to perform feature indicator fusion to obtain first new feature data; Based on the first new feature data, using the early warning model to predict the early warning level of the distribution station area; According to the warning level of the distribution station area, the first data collection time interval is adjusted to obtain a second data collection time interval; Re-collecting the target monitoring data based on the second data collection time interval, and re-obtaining the warning level of the distribution station area according to the re-collected target monitoring data; The process of establishing the early warning model includes: Collect the second historical monitoring data of the distribution station area; Extracting features from the second historical monitoring data to obtain second area data features; Based on the data characteristics of the second substation area, a correlation analysis is performed on the preset characteristic indicators to obtain a second key early warning indicator; Using the second key early warning indicator to fuse the characteristic indicators, to obtain second new characteristic data; Determine the historical warning level of the power distribution station area corresponding to the second new characteristic data; Using the second new feature data and the historical warning level of the power distribution station area to construct a data set, and using the data set to train a long-short neural network model to obtain the warning model; The determining of the historical warning level of the distribution station area corresponding to the second new characteristic data includes: Extract valid data from the second new feature data according to the preset standard limit value of the normal operation of the substation and the preset standard limit value of the fault operation of the substation corresponding to the second key warning indicator, and obtain the processed second new feature data; Based on the second key warning indicator, a K-means clustering method is used to perform feature clustering on the processed second new feature data to obtain clustered second new feature data; Based on the clustered second new feature data, an interval prediction method is used to analyze and obtain the change extreme value corresponding to the second key warning indicator; Based on the clustered second new feature data and the extreme value of change corresponding to the second key warning indicator, a Bootstrap algorithm is used to predict the sample variance of the Bootstrap sample corresponding to the second key warning indicator; Using the sample variance corresponding to the second key warning indicator, calculate the upper limit value and the lower limit value of the second key warning indicator; According to the upper limit value and the lower limit value of the second key warning indicator, the historical warning level of the distribution station area corresponding to the second new characteristic data is determined.

2. The method according to claim 1, characterized in that: The step of extracting features from the target monitoring data of the distribution station area collected based on the first data collection time interval to obtain the first station area data features includes: Performing data cleaning on the target monitoring data; Data corresponding to preset characteristic indicators are extracted from the target monitoring data after data cleaning to obtain the data characteristics of the first station area.

3. The method according to claim 1, characterized in that: The method of performing correlation analysis on preset characteristic indicators based on the first station area data characteristics to obtain a first key early warning indicator includes: Based on the data characteristics of the first substation area, a grayscale correlation analysis method is used to perform correlation analysis on the preset characteristic indicators to obtain the first key early warning indicator.

4. The method according to claim 3, characterized in that: Based on the first station area data feature, a grayscale correlation analysis method is used to perform correlation analysis on the preset feature indicators to obtain the first key early warning indicator, including: Constructing a first comparison sequence using the first station area data feature; Constructing a first reference series using the collected first historical monitoring data; Performing standardization processing on both the first comparison number series and the first reference number series; Calculate the absolute difference between the first comparison series after the normalization processing and the first reference series after the normalization processing corresponding to each preset characteristic index; Determine the grayscale correlation between the first comparison series after the normalization processing and the first reference series after the normalization processing corresponding to each preset characteristic index by using the absolute difference; A preset characteristic indicator corresponding to the grayscale correlation degree that meets the preset conditions is selected as the first key warning indicator.

5. The method according to claim 4, characterized in that The step of constructing a first reference number series using the collected first historical monitoring data includes: Performing data cleaning on the first historical monitoring data; Performing standardization processing on the first historical monitoring data after data cleaning corresponding to the preset characteristic indicator to obtain the first historical monitoring data after standardization processing corresponding to the preset characteristic indicator; Let the minimum value of the first historical monitoring data after the standardization processing corresponding to each preset characteristic index be the optimal value, and the maximum value of the first historical monitoring data after the standardization processing corresponding to each preset characteristic index be the worst value; The first reference sequence is constructed using the optimal value and the worst value corresponding to the preset characteristic index.

6. The method according to claim 4, characterized in that The step of using the absolute difference to determine the grayscale correlation between the first comparison series after standardization and the first reference series after standardization corresponding to each preset characteristic index includes: According to the absolute difference, determine the maximum absolute difference and the minimum absolute difference between the first comparison series after the normalization processing and the first reference series after the normalization processing corresponding to each preset characteristic index; The maximum absolute difference and the minimum absolute difference are used to calculate the grayscale correlation between the first comparison series after the normalization processing and the first reference series after the normalization processing corresponding to each preset characteristic index.

7. The method according to claim 2, characterized in that The step of using the first key early warning indicator to fuse the feature indicators to obtain the first new feature data includes: Extracting data corresponding to the first key early warning indicator from the target monitoring data after data cleaning, and constructing a first feature set matrix; Based on the first feature set matrix, the first new feature data is constructed using a principal component analysis method.

8. The method according to claim 7, characterized in that The constructing the first new feature data based on the first feature set matrix by using a principal component analysis method includes: Using the first feature set matrix, a first covariance matrix is ​​calculated; Calculate the cumulative contribution rate of the first p features in the first covariance matrix; When the cumulative contribution rate of the first p features is greater than or equal to the expected value of the cumulative contribution rate, constructing the first new feature data using the first p features in the first covariance matrix; Where p is a positive integer, p≤k, and k is the total number of features in the first covariance matrix.

9. The method according to claim 2, characterized in that: The adjusting the first data collection time interval according to the warning level of the distribution station area to obtain the second data collection time interval includes: If the warning level is the fourth warning level or the number of electric vehicles entering the distribution station area reaches the first percentage of the charging station capacity, the second data collection time interval is the first multiple of the first data collection time interval; If the warning level is the third warning level or the number of electric vehicles entering the distribution station area reaches the second percentage of the charging station capacity, the second data collection time interval is the second multiple of the first data collection time interval; If the warning level is lower than or equal to the second warning level, and the number of electric vehicles entering the distribution station area is less than the second percentage of the charging station capacity, the second data collection time interval is the third multiple of the first data collection time interval; If the warning level is lower than or equal to the first warning level, and the number of electric vehicles entering the distribution station area is less than the second percentage of the charging station capacity, the second data collection time interval is the first data collection time interval.

10. The method according to claim 1, characterized in that The extracting features of the second historical monitoring data to obtain the second area data features includes: Performing data cleaning on the second historical monitoring data; Data corresponding to preset characteristic indicators are extracted from the second historical monitoring data after data cleaning to obtain the data characteristics of the second station area.

11. The method according to claim 1, characterized in that Based on the second station area data feature, the preset feature indicators are analyzed for correlation to obtain the second key early warning indicator, including: Based on the data characteristics of the second substation area, a grayscale correlation analysis method is used to perform correlation analysis on the preset characteristic indicators to obtain the second key warning indicator.

12. The method according to claim 11, characterized in that Based on the second station area data feature, a grayscale correlation analysis method is used to perform correlation analysis on the preset feature indicators to obtain the second key early warning indicator, including: Constructing a second comparison sequence using the second station area data feature; Constructing a second reference series using the collected third historical monitoring data; performing standardization processing on both the second comparison number series and the second reference number series; Calculate the absolute difference between the second comparison series after the normalization processing and the second reference series after the normalization processing corresponding to each preset characteristic index; Determine the grayscale correlation between the second comparison series after the normalization processing and the second reference series after the normalization processing corresponding to each preset characteristic index by using the absolute difference; Selecting a characteristic indicator corresponding to the grayscale correlation degree that meets the preset conditions as the second key early warning indicator; Wherein, the collection time of the third historical monitoring data is earlier than the collection time of the second historical monitoring data.

13. The method according to claim 12, characterized in that The method of constructing a second reference number series by using the collected third historical monitoring data includes: Performing data cleaning on the third historical monitoring data; Performing standardization processing on the third historical monitoring data after data cleaning corresponding to the preset characteristic indicator to obtain the third historical monitoring data after standardization processing corresponding to the preset characteristic indicator; Let the minimum value of the third historical monitoring data after the standardization processing corresponding to each preset characteristic index be the optimal value, and the maximum value of the third historical monitoring data after the standardization processing corresponding to each preset characteristic index be the worst value; The second reference sequence is constructed using the optimal value and the worst value corresponding to the preset characteristic index.

14. The method according to claim 12, characterized in that The step of using the absolute difference to determine the grayscale correlation between the second comparison series after the normalization processing and the second reference series after the normalization processing corresponding to each preset characteristic index includes: According to the absolute difference, determine the maximum absolute difference and the minimum absolute difference between the second comparison series after the normalization processing and the second reference series after the normalization processing corresponding to each preset characteristic index; The maximum absolute difference and the minimum absolute difference are used to calculate the grayscale correlation between the second comparison series after the normalization processing and the second reference series after the normalization processing corresponding to each preset characteristic index.

15. The method according to claim 10, characterized in that The step of using the second key early warning indicator to fuse the feature indicators to obtain the second new feature data includes: Extracting data corresponding to the second key early warning indicator from the second historical monitoring data after data cleaning, and constructing a second feature set matrix; Based on the second feature set matrix, the second new feature data is constructed by using a principal component analysis method.

16. The method according to claim 15, characterized in that The constructing the second new feature data based on the second feature set matrix by using a principal component analysis method includes: Using the second feature set matrix, a second covariance matrix is ​​calculated; Calculate the cumulative contribution rate of the first p' features in the second covariance matrix; When the cumulative contribution rate of the first p' features is greater than or equal to the expected value of the cumulative contribution rate, the second new feature data is constructed using the first p' features in the second covariance matrix; Wherein, p' is a positive integer, p'≤k', and k' is the total number of features in the second covariance matrix.

17. The method according to claim 1, characterized in that Determining the historical warning level of the distribution station area corresponding to the second new feature data according to the upper limit value and the lower limit value of the second key warning indicator includes: When the upper limit value and the lower limit value of the second key warning indicator belong to the fluctuation range corresponding to the first confidence threshold value, the historical warning level of the distribution station area is the first warning level; When the upper limit value and the lower limit value of the second key warning indicator belong to the fluctuation range corresponding to the second confidence threshold, the historical warning level of the distribution station area is the second warning level; When the upper limit value and the lower limit value of the second key warning indicator belong to the fluctuation range corresponding to the third confidence threshold value, the historical warning level of the distribution station area is the third warning level; When the upper limit value and the lower limit value of the second key warning indicator belong to the fluctuation range corresponding to the fourth confidence threshold value, the historical warning level of the distribution station area is the fourth warning level.

18. The method according to claim 1, characterized in that The method of using the data set to train the long-short neural network model to obtain the early warning model includes: Dividing the data set into a training set, a validation set, and a test set; Using the training set to train the long-short neural network model to obtain a trained long-short neural network model; The trained long-short neural network model is verified using the verification set. If the verification is successful, the performance of the trained long-short neural network model is evaluated using the test set. If the verification fails, the hyperparameters of the long-short neural network model are adjusted and overfitting is monitored, and the long-short neural network model is retrained until the verification is successful. The performance of the trained long and short neural network model is evaluated using the test set. If the performance evaluation of the trained long and short neural network model is successful, the trained long and short neural network model is the early warning model. If the performance evaluation of the trained long and short neural network model fails, the second historical monitoring data is recollected, and an early warning model is established using the recollected second historical monitoring data.

19. The method according to claim 8, characterized in that The calculation formula of the first covariance matrix includes: B=X T X The calculation formula for the cumulative contribution rate of the first p features in the first covariance matrix includes: In the above formula, X is the first feature set matrix, T is the transpose, and B is the first covariance matrix; , k is the total number of features in the first covariance matrix, p≤k; is the cumulative contribution rate of the first p features in the first covariance matrix, is the eigenvalue of the i-th feature in the first covariance matrix.

20. The method according to claim 16, characterized in that The calculation formula of the second covariance matrix includes: B'=X' T X' The calculation formula of the cumulative contribution rate of the first p' features in the second covariance matrix includes: In the above formula, X' is the second feature set matrix, T is the transpose, and 'B is the second covariance matrix; i'∈ [ 1,k' ], k' is the total number of features in the second covariance matrix, p'≤k'; is the cumulative contribution rate of the first p' features in the second covariance matrix, is the eigenvalue of the i'th feature in the second covariance matrix.

21. The method according to claim 1, characterized in that The calculation formula for the upper limit value of the second key warning indicator includes: The calculation formula for the lower limit value of the second key warning indicator includes: In the above formula, is the upper limit value of the second key warning indicator, is the lower limit of the second key warning indicator, is the mean value of the second new feature data after clustering corresponding to the second key warning indicator, is the sample variance corresponding to the second key early warning indicator, is a standard normal distribution 100·(1- α )% corresponding critical value, α is the adjustment factor.

22. The method according to claim 1, characterized in that The distribution station area is a distribution station area including an electric vehicle charging station.

23. A distribution station area risk classification warning system, characterized in that: include: An extraction unit, used for extracting features from the target monitoring data of the distribution station area collected based on the first data collection time interval to obtain data features of the first station area; A first acquisition unit is used to perform a correlation analysis on a preset characteristic indicator based on the first station area data feature to obtain a first key warning indicator; A second acquisition unit is used to use the first key warning indicator to perform feature indicator fusion to obtain first new feature data; A third acquisition unit is used to predict the warning level of the power distribution station area by using the warning model based on the first new feature data; An adjusting unit, configured to adjust the first data collection time interval according to the warning level of the distribution station area to obtain a second data collection time interval; a fourth acquisition unit, configured to re-collect the target monitoring data based on the second data collection time interval, and re-acquire the warning level of the distribution station area according to the re-collected target monitoring data; The system further comprises: an establishing unit for establishing an early warning model; the establishing unit comprises: A collection subunit, used for collecting the second historical monitoring data of the distribution station area; The third acquisition subunit is used to extract features from the second historical monitoring data to obtain data features of the second station area; The fourth acquisition subunit is used to perform a correlation analysis on the preset characteristic indicators based on the data characteristics of the second station area to obtain a second key early warning indicator; A fifth acquisition subunit is used to use the second key early warning indicator to fuse the characteristic indicators to obtain second new characteristic data; A determination subunit, used to determine the historical warning level of the distribution station area corresponding to the second new characteristic data; The sixth acquisition subunit is used to construct a data set using the second new feature data and the historical warning level of the distribution station area, and use the data set to train the long-short neural network model to obtain a warning model; The determining subunit comprises: A third acquisition module is used to extract valid data from the second new feature data according to the preset standard limit value of normal operation of the substation and the preset standard limit value of fault operation of the substation corresponding to the second key warning indicator, so as to obtain the processed second new feature data; A fourth acquisition module is used to perform feature clustering on the processed second new feature data by using a K-means clustering method based on the second key warning indicator to obtain clustered second new feature data; A fifth acquisition module is used to analyze and obtain the change extreme value corresponding to the second key warning indicator by using an interval prediction method based on the clustered second new feature data; A prediction module, used to predict the sample variance of the Bootstrap sample corresponding to the second key warning indicator by using the Bootstrap algorithm based on the clustered second new feature data and the change extreme value corresponding to the second key warning indicator; A fourth calculation module, used to calculate the upper limit value and the lower limit value of the second key warning indicator by using the sample variance corresponding to the second key warning indicator; The sixth acquisition module is used to determine the historical warning level of the distribution station area corresponding to the second new feature data according to the upper limit value and the lower limit value of the second key warning indicator.

24. The system according to claim 23, characterized in that The extraction unit comprises: A data cleaning subunit, used for cleaning the target monitoring data; The first acquisition subunit is used to extract data corresponding to preset characteristic indicators from the target monitoring data after data cleaning to obtain the first station area data characteristics.

25. The system according to claim 23, characterized in that The first acquisition unit includes: The second acquisition subunit is used to perform correlation analysis on the preset characteristic indicators based on the first station area data characteristics by using a grayscale correlation analysis method to obtain the first key warning indicator.

26. The system according to claim 25, characterized in that The second acquisition subunit includes: A first construction module, used to construct a first comparison sequence using the first station area data feature; A second construction module is used to construct a first reference number series using the collected first historical monitoring data; A processing module, configured to perform standardization processing on both the first comparison number series and the first reference number series; A first calculation module, used for calculating the absolute difference between the first comparison series after standardization and the first reference series after standardization corresponding to each preset characteristic index; A determination module, used to determine the grayscale correlation between the first comparison series after standardization and the first reference series after standardization corresponding to each preset characteristic index by using the absolute difference; A selection module is used to select a preset characteristic indicator corresponding to a grayscale correlation degree that meets a preset condition as the first key warning indicator.

27. The system according to claim 26, characterized in that The second building block is specifically used for: Performing data cleaning on the first historical monitoring data; Performing standardization processing on the first historical monitoring data after data cleaning corresponding to the preset characteristic indicator to obtain the first historical monitoring data after standardization processing corresponding to the preset characteristic indicator; Let the minimum value of the first historical monitoring data after the standardization processing corresponding to each preset characteristic index be the optimal value, and the maximum value of the first historical monitoring data after the standardization processing corresponding to each preset characteristic index be the worst value; The first reference sequence is constructed using the optimal value and the worst value corresponding to the preset characteristic index.

28. The system according to claim 26, characterized in that The determining module is specifically used for: According to the absolute difference, determine the maximum absolute difference and the minimum absolute difference between the first comparison series after the normalization processing and the first reference series after the normalization processing corresponding to each preset characteristic index; The maximum absolute difference and the minimum absolute difference are used to calculate the grayscale correlation between the first comparison series after the normalization processing and the first reference series after the normalization processing corresponding to each preset characteristic index.

29. The system according to claim 24, characterized in that The second acquisition unit includes: A first construction subunit is used to extract data corresponding to the first key early warning indicator from the target monitoring data after data cleaning, and construct a first feature set matrix; The second construction subunit is used to construct the first new feature data based on the first feature set matrix by using a principal component analysis method.

30. The system according to claim 29, wherein the second construction subunit comprises: A second calculation module, used to calculate a first covariance matrix using the first feature set matrix; A third calculation module, used to calculate the cumulative contribution rate of the first p features in the first covariance matrix; A third construction module is used to construct the first new feature data using the first p features in the first covariance matrix when the cumulative contribution rate of the first p features is greater than or equal to the expected value of the cumulative contribution rate; Where p is a positive integer, p≤k, and k is the total number of features in the first covariance matrix.

31. The system according to claim 23, characterized in that The adjustment unit is specifically used for: If the warning level is the fourth warning level or the number of electric vehicles entering the distribution station area reaches the first percentage of the charging station capacity, the second data collection time interval is the first multiple of the first data collection time interval; If the warning level is the third warning level or the number of electric vehicles entering the distribution station area reaches the second percentage of the charging station capacity, the second data collection time interval is the second multiple of the first data collection time interval; If the warning level is lower than or equal to the second warning level, and the number of electric vehicles entering the distribution station area is less than the second percentage of the charging station capacity, the second data collection time interval is the third multiple of the first data collection time interval; If the warning level is lower than or equal to the first warning level, and the number of electric vehicles entering the distribution station area is less than the second percentage of the charging station capacity, the second data collection time interval is the first data collection time interval.

32. The system according to claim 23, characterized in that The distribution station area is a distribution station area including an electric vehicle charging station.

33. An electronic device, characterized in that: include: at least one processor and memory; The memory and the processor are connected via a bus; The memory is used to store one or more programs; When the one or more programs are executed by the at least one processor, the distribution station area risk classification warning method as described in any one of claims 1 to 22 is implemented.

34. A readable storage medium, characterized in that: An execution program is stored thereon, and when the execution program is executed, the distribution station area risk classification warning method as described in any one of claims 1 to 22 is implemented.

Images