CN-121981708-A - High-voltage cable intelligent operation and maintenance management method and system based on big data analysis

CN121981708ACN 121981708 ACN121981708 ACN 121981708ACN-121981708-A

Abstract

The invention relates to the technical field of data processing, and provides a high-voltage cable intelligent operation and maintenance management method and system based on big data analysis. The method comprises the steps of extracting characteristics of a multi-source heterogeneous data set of a high-voltage cable in a first time window to obtain key characteristic parameter values corresponding to all time nodes, calculating a temperature prediction result of the high-voltage cable in a second time window based on the key characteristic parameter values corresponding to all time nodes and the temperature parameter values of the high-voltage cable, evaluating the temperature prediction result of the high-voltage cable in the second time window through a buffer analysis algorithm to obtain a risk early warning event, maintaining the risk early warning event based on real-time power flow and a topological structure of a power grid to generate a first operation and maintenance management scheme, and optimizing the first operation and maintenance management scheme through a neural network algorithm to obtain a target operation and maintenance management scheme. The embodiment of the invention improves the reliability of operation and maintenance management of the high-voltage cable and the operation safety of the cable.

Inventors

GUO ZHIGANG
LI SHAOBIN
YAN FENG
YANG LINQING
LIU PENGYUE
QU TONG
Yao Jiachi
WU QIAN
JIA ZHANHAO
WU JINHAO

Assignees

国网陕西省电力有限公司西安供电公司

Dates

Publication Date: 20260505
Application Date: 20260205

Claims (10)

1. The intelligent operation and maintenance management method for the high-voltage cable based on big data analysis is characterized by comprising the following steps of: Performing feature extraction on the multi-source heterogeneous data set of the high-voltage cable in a first time window to obtain key feature parameter values corresponding to all time nodes; Calculating a temperature prediction result of the high-voltage cable in a second time window based on key characteristic parameter values corresponding to all the time nodes and the temperature parameter values of the high-voltage cable; evaluating a temperature prediction result of the high-voltage cable in a second time window through a buffer area analysis algorithm to obtain a risk early warning event; Based on the real-time tide and the topological structure of the power grid, maintaining the risk early warning event to generate a first operation and maintenance management scheme; And optimizing the first operation and maintenance management scheme through a neural network algorithm to obtain a target operation and maintenance management scheme.
2. The method of claim 1, wherein the feature extraction of the multi-source heterogeneous data set of the high-voltage cable in the first time window to obtain key feature parameter values corresponding to each time node comprises: Performing feature extraction on each feature parameter in the multi-source heterogeneous data set in the first time window to obtain a feature value of each feature parameter; Calculating a correlation coefficient between the characteristic value of the characteristic parameter and the temperature parameter value of the key part of the cable in the first time window by a Pearson correlation coefficient algorithm aiming at each characteristic parameter to obtain a first correlation coefficient value; extracting each characteristic parameter based on each first correlation coefficient value to obtain each key characteristic parameter; And arranging the characteristic values of the key characteristic parameters according to the sequence of each time node in the first time window to obtain the key characteristic parameter values corresponding to each time node.
3. The method according to claim 1, wherein calculating the temperature prediction result of the high voltage cable in the second time window based on the key feature parameter value corresponding to each time node and the temperature parameter value of the high voltage cable comprises: Calculating correlation coefficients between each key characteristic parameter and the cable conductor layer based on the key characteristic parameter value corresponding to each time node and the temperature parameter value of the cable conductor layer in the high-voltage cable to obtain each second correlation value; Calculating correlation coefficients between each key characteristic parameter and the cable shielding layer based on key characteristic parameter values corresponding to each time node and temperature parameter values of the cable shielding layer in the high-voltage cable to obtain each third phase relation value; according to a preset period dividing rule, carrying out period division on key characteristic parameter values corresponding to each time node to obtain each period parameter set; comparing the average value of the key characteristic parameter values in the same time period in each periodic parameter set to obtain the periodic variation rule of each key characteristic parameter; According to the periodic variation rule of each key characteristic parameter and the time attribute of the second time window, determining the predicted value of each key characteristic parameter in the second time window; And obtaining a temperature prediction result of the high-voltage cable in a second time window according to each second correlation coefficient, each third correlation value, the predicted value of each key characteristic parameter in the second time window and the preset basic temperature of the high-voltage cable.
4. A method according to claim 3, wherein the obtaining a temperature prediction result of the high voltage cable in the second time window according to each second correlation coefficient, each third correlation value, a predicted value of each key feature parameter in the second time window, and a preset base temperature of the high voltage cable includes: Summing products of the second correlation numbers and predicted values of the key characteristic parameters under the time nodes to obtain a first total contribution value; determining a predicted temperature of the cable conductor layer at the time node based on a sum of the first total contribution value and a base temperature of the cable conductor layer in the preset base temperature; Summing products of the third relation numbers and predicted values of the key characteristic parameters under the time nodes to obtain second total contribution values; Determining a predicted temperature of the cable shielding layer at the time node based on a sum of the second total contribution value and a base temperature of the cable shielding layer in the preset base temperature; And determining a temperature prediction result of the high-voltage cable in a second time window based on the predicted temperatures of the cable conductor layer and the cable shielding layer in each time node.
5. The method of claim 1, wherein the evaluating, by a buffer zone analysis algorithm, the temperature prediction of the high voltage cable within a second time window to obtain a risk early warning event comprises: comparing the temperature prediction result of the high-voltage cable in a second time window with a preset short-term safety threshold and a preset long-term safety threshold respectively time by time node to identify temperature points larger than the short-term safety threshold or the long-term safety threshold, and obtaining abnormal temperature points and abnormal types of the abnormal temperature points; Classifying the abnormal temperature points according to the abnormal temperature difference of the abnormal temperature points to obtain abnormal grades of the abnormal temperature points; Determining the basic buffer distance of each abnormal temperature point according to a preset basic buffer distance standard and each abnormal grade; Based on the abnormal type of each abnormal temperature point, adjusting the basic buffer distance of each abnormal temperature point to obtain the target buffer distance of each abnormal temperature point; Constructing a multi-stage buffer area by taking the physical position of each abnormal temperature point as the center according to the target buffer distance of the abnormal temperature point to obtain each multi-stage buffer area; Calculating a risk grade index based on the length of the high-voltage cable section in each multi-stage buffer area, the abnormal temperature difference of each abnormal temperature point, the number of associated devices and the load importance degree parameter; and packaging the risk grade index, each multi-stage buffer area, each abnormal temperature point, the abnormal type of the abnormal temperature point and the abnormal grade to obtain the risk early warning event.
6. The method of claim 5, wherein the calculating a risk level index based on the high voltage cable segment length in each of the multi-level buffer areas, the abnormal temperature difference for each of the abnormal temperature points, the number of associated devices, and the load importance parameter comprises: acquiring the length of a high-voltage cable section of each buffer area in each multi-level buffer area; Summing the lengths of the high-voltage cable sections of each buffer area in the multi-stage buffer areas to obtain the total cable length; summing the number of associated devices in each buffer area in each multi-level buffer area to obtain the total number of associated devices; determining importance degree parameters of each load according to the load types of each device in each multi-level buffer area; summing the load importance degree parameters to obtain a total load parameter; Summing the abnormal temperature differences of the abnormal temperature points to obtain a total abnormal temperature difference; and based on preset weight coefficients of all dimensions, weighting and summing the total high-voltage cable length, the total associated equipment quantity, the total load parameters and the total abnormal temperature difference to obtain the risk grade index.
7. The method of claim 1, wherein the maintaining the risk early warning event based on the real-time power flow and topology of the power grid generates a first operation and maintenance management scheme comprising: carrying out structural analysis on the risk early warning event to obtain a risk key parameter; Mapping the risk grade index in the risk key parameters with a preset operation and maintenance priority interval to obtain a target operation and maintenance priority; Determining a regulation period based on the abnormal duration in the risk key parameter; Determining a line to be adjusted based on the topology structure, the length of the high-voltage cable section in the risk key parameter and the number of associated devices; Determining a load adjustment amount based on the load amount required to be transferred by the line to be adjusted and the real-time power flow of the power grid; and generating the first operation and maintenance management scheme based on the target operation and maintenance priority, the regulation and control period, the line to be regulated and the load regulation amount.
8. The method of claim 1, wherein optimizing the first operation and maintenance management scheme by a neural network algorithm to obtain a target operation and maintenance management scheme comprises: carrying out standardization and coding treatment on each parameter in the first operation and maintenance management scheme to obtain a first input feature vector; Taking the first input feature vector as the input of a pre-trained first neural network model to predict each second operation and maintenance management scheme of the first operation and maintenance management scheme under different adjustment strategies; based on preset power grid operation constraint conditions, verifying operation state parameters in each second operation and maintenance management scheme, and taking each second operation and maintenance management scheme passing the verification as a third operation and maintenance management scheme respectively; Aiming at each third operation and maintenance management scheme, encoding parameters in the third operation and maintenance management scheme to obtain a second input feature vector; taking the second input feature vector as the input of a pre-trained second neural network model to predict sub-dimension evaluation values of the third operation and maintenance management scheme under three evaluation indexes of system stability, load balance degree and risk control result, so as to obtain each sub-dimension evaluation value; carrying out weighted summation on each sub-dimension evaluation value to obtain a comprehensive evaluation score of the third operation and maintenance management scheme; And taking the third operation and maintenance management scheme corresponding to the maximum value in the comprehensive evaluation scores of the third operation and maintenance management schemes as the target operation and maintenance management scheme.
9. The method of claim 1, wherein the optimizing the first operation and maintenance management scheme by the neural network algorithm results in the target operation and maintenance management scheme, and further comprising: analyzing the target operation and maintenance management scheme to obtain an operation execution plan; converting the operation execution plan into a control instruction sequence and issuing the control instruction sequence to a monitoring automation control device so that the monitoring automation control device executes load adjustment operation to obtain a monitoring data set; Receiving a monitoring data set sent by the monitoring automation control device, and evaluating a risk control target based on the monitoring data set to obtain a load adjustment verification result; and generating a detailed overhaul plan according to the load adjustment verification result, the maintenance requirement in the target operation and maintenance management scheme and the historical operation state data of the high-voltage cable.
10. High-voltage cable intelligence fortune dimension management system based on big data analysis, characterized by, include: the characteristic extraction module is used for extracting characteristics of the multi-source heterogeneous data set of the high-voltage cable in the first time window to obtain key characteristic parameter values corresponding to all time nodes; the temperature prediction result calculation module is used for calculating a temperature prediction result of the high-voltage cable in a second time window based on key characteristic parameter values corresponding to all the time nodes and the temperature parameter values of the high-voltage cable; The evaluation module is used for evaluating the temperature prediction result of the high-voltage cable in the second time window through a buffer area analysis algorithm to obtain a risk early warning event; The operation and maintenance management scheme generation module is used for maintaining the risk early warning event based on the real-time tide and the topological structure of the power grid to generate a first operation and maintenance management scheme; and the operation and maintenance management scheme optimizing module is used for optimizing the first operation and maintenance management scheme through a neural network algorithm to obtain a target operation and maintenance management scheme.

Description

High-voltage cable intelligent operation and maintenance management method and system based on big data analysis Technical Field The invention relates to the technical field of data processing, in particular to a high-voltage cable intelligent operation and maintenance management method and system based on big data analysis. Background Along with the continuous expansion of the power grid scale and the continuous improvement of the load level, the high-voltage cable is used as an important carrier for power transmission, and the running state of the high-voltage cable is directly related to the safety and stability of the power grid. The existing operation and maintenance management strategies mostly depend on preset threshold alarming before the occurrence of the cable faults, maintenance of the cable faults according to alarming results, statistics of the faults afterwards and establishment of operation and maintenance management schemes corresponding to fault types according to statistics results. Because fault sample data under various fault types is difficult to exhaust the actual working conditions, and the preset model or threshold is often set based on ideal or typical working conditions, deviation exists between the preset model or threshold and the actual conditions of the scene with various changes, and risk early warning is delayed or false alarm is frequent. Under the condition, an operation and maintenance management scheme or an emergency treatment scheme established based on one-sided or lagged information is difficult to accurately match with actual accident potential or failure modes, so that the reliability of operation and maintenance management of the voltage cable and the operation safety of the cable are difficult to meet actual requirements. Disclosure of Invention The invention provides a high-voltage cable intelligent operation and maintenance management method and system based on big data analysis, which can solve at least one technical problem. In a first aspect, an embodiment of the present invention provides a method for intelligent operation and maintenance management of a high-voltage cable based on big data analysis, including: Performing feature extraction on the multi-source heterogeneous data set of the high-voltage cable in a first time window to obtain key feature parameter values corresponding to all time nodes; Calculating a temperature prediction result of the high-voltage cable in a second time window based on key characteristic parameter values corresponding to all the time nodes and the temperature parameter values of the high-voltage cable; evaluating a temperature prediction result of the high-voltage cable in a second time window through a buffer area analysis algorithm to obtain a risk early warning event; Based on the real-time tide and the topological structure of the power grid, maintaining the risk early warning event to generate a first operation and maintenance management scheme; And optimizing the first operation and maintenance management scheme through a neural network algorithm to obtain a target operation and maintenance management scheme. In a second aspect, an embodiment of the present invention provides a high-voltage cable intelligent operation and maintenance management system based on big data analysis, including: the characteristic extraction module is used for extracting characteristics of the multi-source heterogeneous data set of the high-voltage cable in the first time window to obtain key characteristic parameter values corresponding to all time nodes; the temperature prediction result calculation module is used for calculating a temperature prediction result of the high-voltage cable in a second time window based on key characteristic parameter values corresponding to all the time nodes and the temperature parameter values of the high-voltage cable; The evaluation module is used for evaluating the temperature prediction result of the high-voltage cable in the second time window through a buffer area analysis algorithm to obtain a risk early warning event; The operation and maintenance management scheme generation module is used for maintaining the risk early warning event based on the real-time tide and the topological structure of the power grid to generate a first operation and maintenance management scheme; and the operation and maintenance management scheme optimizing module is used for optimizing the first operation and maintenance management scheme through a neural network algorithm to obtain a target operation and maintenance management scheme. In a third aspect, an embodiment of the present invention further provides an electronic device, including at least one processor, and a memory communicatively coupled to the at least one processor, where the memory stores instructions executable by the at least one processor, the instructions being executable by the at least one processor to enable the at least one processor to perform th