Power transmission service automatic operation and maintenance management system based on big data

Abstract

The invention relates to the technical field of intelligent operation and maintenance, in particular to a power transmission service automatic operation and maintenance management system based on big data, which comprises a node tension monitoring module, a risk pressure assessment module, a fracture evolution staging inference module, a convergence center positioning module and a fracture evolution early warning module. According to the method, the inflection point change is extracted from the tension change tracks of the steel tower contact and the strain section fitting node of the power transmission line, standardized integration and comparison are carried out based on the node position correlation potential mutation amplitude and the microclimate wind speed increment, the risk pressure distribution of the node is accurately identified, and early detection of the hidden danger node can be realized. The node fracture evolution staging is realized by detecting the tension change rate of the node of which the pressure change exceeds a threshold value and dividing stages according to the rate change characteristic. And the hidden danger center is further screened through the pressure level change trend, and the convergence center is locked based on the fluctuation amplitude difference, so that the positioning accuracy of the hidden danger gathering area is improved.

Description

Technical Field

[0001] The present invention relates to the field of intelligent operation and maintenance technology, and in particular to an automated operation and maintenance management system for power transmission services based on big data. Background Art

[0002] The field of intelligent operations and maintenance (O&M) encompasses systematic management methods and approaches for real-time monitoring, analysis, diagnosis, and processing of the operational status of equipment and facilities. Core elements of this technology include collecting large amounts of operational data and applying data mining, intelligent decision-making, and automated execution to identify and assess potential equipment failures. This allows for optimized O&M decision-making and automated workflows throughout the facility's lifecycle.

[0003] Among them, the automated operation and maintenance management system for power transmission business refers to the inspection, monitoring, maintenance and other business links of power transmission lines and their ancillary facilities. It obtains line operation data by deploying sensing terminal equipment, and uses specific processing methods such as data cleaning, feature extraction, status identification, fault location, risk assessment, and maintenance assignment to uniformly handle and command matters such as data management, hidden danger detection, and operation scheduling in the power transmission business operation and maintenance process.

[0004] Although the existing automated operation and maintenance management system for power transmission services collects and processes data through sensing terminal devices, it has limitations in node risk identification and potential hazard expansion deduction. Tension data is mostly collected continuously, lacking fine-grained identification of key inflection points in the node tension change trajectory. This leads to the omission of early signs of potential hazard development, reducing the sensitivity of detection. The identification of node potential hazards is usually based on single indicator assessment, failing to effectively integrate multi-source heterogeneous data such as node potential changes and wind speed increments, resulting in insufficient comprehensiveness and accuracy in risk assessment. In the process of potential hazard development trend analysis, existing technologies are mostly based on static threshold judgments, lacking dynamic tracking of node tension change rates and stage-by-stage evolution trends, and unable to accurately divide potential hazard evolution stages, resulting in delayed potential hazard expansion predictions and difficulty in efficiently responding to operation and maintenance resource allocation. For example, in the scenario of sudden local wind speed changes, the failure to timely perceive tension changes exacerbates the risk of node fracture evolution, increasing the suddenness and uncontrollability of overall line failures. In terms of hidden danger location, only relying on single-point data of pressure level or node anomaly, ignoring the fluctuation relationship between the hidden danger center and surrounding nodes, resulting in deviations in the locking of the hidden danger center, affecting the accuracy of subsequent early warning and emergency repair scheduling. Summary of the Invention

[0005] The purpose of the present invention is to solve the shortcomings of the existing technology and propose an automated operation and maintenance management system for power transmission business based on big data.

[0006] In order to achieve the above objectives, the present invention adopts the following technical solutions: The automatic operation and maintenance management system for power transmission business based on big data includes:

[0007] The node tension monitoring module obtains the tension records of the steel tower joints and tension section hardware nodes in the transmission line, detects the change trajectory of the node tension within the time window, and generates the node tension inflection point change;

[0008] The risk pressure assessment module is based on the node position corresponding to each inflection point change in the node tension inflection point change, combined with the node potential mutation amplitude and micro-meteorological wind speed increment, and then normalized and superimposed for comparison to obtain the node risk pressure distribution;

[0009] The fracture evolution stage inference module obtains the node position where the pressure change exceeds the change threshold in the node risk pressure distribution, traces back the change of the tension inflection point, detects the tension change rate between consecutive nodes, and divides the node fracture evolution stage interval according to the rate and span change characteristics;

[0010] The convergence center positioning module obtains the nodes classified as the accelerated fission stage and the critical fracture stage in the node fracture evolution stage interval, determines the node hidden danger center, and generates the hidden danger convergence center positioning result;

[0011] The fracture evolution warning module simulates the hazard expansion path and node fracture trend distribution based on the hazard center position and surrounding node set in the hazard convergence center positioning result, and generates a transmission line fracture risk situation analysis result.

[0012] As a further solution of the present invention, the node tension inflection point change includes the distribution of the number of inflection points, the tension change amplitude range, and the tension change inflection point position; the node risk pressure distribution includes the node pressure extreme value position, the pressure change rate range, and the pressure change gradient trend; the node fracture evolution stage interval includes the slow decay stage node set, the accelerated fission stage node set, and the critical fracture stage node set; the hidden danger convergence center positioning result includes the hidden danger center node position, the risk distribution status around the hidden danger center, and the hidden danger center pressure change amplitude; the transmission line fracture risk situation analysis result includes the risk expansion path trajectory, the warning target node position, and the fracture risk distribution situation.

[0013] As a further solution of the present invention, the node tension monitoring module includes:

[0014] The tension data acquisition submodule acquires the tension record data of the steel tower joints and tension section hardware nodes in the transmission line based on the big data platform, detects the continuous change trajectory of the tension at each node within the time window, collects the node tension change sequence, extracts the continuous interval segmentation of the node tension change sequence, and generates a continuous record of the node tension;

[0015] The continuous trajectory processing submodule performs interval averaging processing on the continuous tension change interval of each node based on the continuous record of the node tension, obtains a tension average value sequence of the continuous change interval, performs time smoothing processing based on the tension average value sequence, establishes a node continuous tension change trend curve, and generates a node tension continuous change trend based on the change direction of the node continuous tension change trend curve;

[0016] The change inflection point extraction submodule performs piecewise linear fitting on the change trend curve based on the continuous change trend of the node tension, detects the change inflection point position in the fitting curve, extracts the turning characteristics of the tension change trend before and after the inflection point, and generates the node tension inflection point change by combining the number of inflection points and the distribution of change amplitude.

[0017] As a further solution of the present invention, the risk pressure assessment module includes:

[0018] The inflection point change processing submodule obtains the node position corresponding to each inflection point change in the node tension inflection point change, detects the insulator surface potential mutation amplitude data and micrometeorological wind speed increment data corresponding to the node position, performs normalization processing on the node tension inflection point change, potential mutation amplitude, and wind speed increment data, and generates a standardized change set;

[0019] Based on the standardized change set, the potential mutation analysis submodule integrates and compares the data amplitudes of the corresponding positions of the node tension inflection point change, potential mutation amplitude, and wind speed increment standardized data according to the node position, calls the compared data group, identifies the amplitude trend of continuous change according to the node arrangement order, and obtains the node change trend amount;

[0020] The risk pressure distribution generation submodule detects the difference between the change degree of each node and the change threshold based on the node change trend, calls the node position data whose change degree exceeds the change threshold, filters it as the pressure reference set, and establishes the node risk pressure distribution.

[0021] As a further solution of the present invention, the fault evolution stage inference module includes:

[0022] The pressure node extraction submodule obtains the node positions in the node risk pressure distribution where the pressure change exceeds the change threshold, traces back the node tension inflection point change of the corresponding node, calls the node tension inflection point change, detects the change interval range corresponding to the change according to the node sequence, selects nodes that match the pressure change characteristics, and generates a pressure characteristic node group;

[0023] The tension rate detection submodule constructs a continuous node sequence based on the pressure characteristic node group and the change amount of the tension inflection point between nodes, detects the tension change rate between adjacent nodes, calls the change rate data, and aggregates the rate intervals according to the span characteristics of the node sequence to obtain the tension change rate interval;

[0024] The stage interval division submodule divides the node stages according to the tension change rate interval and the rate change trend characteristics, classifies the nodes into slow decay stage, accelerated fission stage and critical fracture stage according to the trend, counts the proportion of nodes in each stage, and establishes the node fracture evolution stage interval.

[0025] As a further solution of the present invention, the convergence center positioning module includes:

[0026] The hidden danger node screening submodule obtains the nodes in the accelerated fission stage and critical fracture stage in the node fracture evolution stage interval, screens the position data of the corresponding nodes, calls the node position data to obtain the node risk pressure distribution, detects the pressure level change trend between the screened nodes, determines the node sequence fluctuation amplitude according to the change amplitude, and generates a pressure change trend group;

[0027] Based on the pressure change trend group, the center candidate extraction submodule extracts the node with the largest pressure level fluctuation amplitude as the hidden danger center candidate location, calls the risk pressure level data of the nodes surrounding the candidate node, and performs a corresponding change amplitude detection based on the fluctuation amplitude of the candidate location and the fluctuation amplitude of the surrounding nodes to obtain the hidden danger center difference sequence;

[0028] The convergence center locking submodule determines the pressure change amplitude difference between the surrounding nodes and the candidate nodes according to the hidden danger center difference sequence, selects the node with the smallest amplitude difference as the final hidden danger center locking point, and establishes the hidden danger convergence center positioning result.

[0029] As a further solution of the present invention, the fracture evolution early warning module includes:

[0030] The expansion trend deduction submodule obtains the hidden danger center position and the surrounding node set in the hidden danger convergence center positioning result, calls the risk pressure change speed data of the hidden danger center position and the surrounding nodes, detects the time series process of risk expansion based on the relationship between the change speeds between nodes, and deduces based on the synchronization characteristics of the expansion speed change and the stage division state to generate the risk expansion evolution trend;

[0031] Based on the risk expansion evolution trend, the early warning target screening submodule extracts the node sequence where the risk level has a sudden change during the expansion process, detects whether the corresponding evolution stage of the node has a stage transition, and selects nodes with both risk level mutation and stage transition characteristics as early warning targets to obtain a stage transition node set;

[0032] The risk situation assessment submodule detects the spatial distribution of warning target nodes in the transmission line based on the set of stage transition nodes, derives the overall line fracture risk distribution situation based on the relationship between node distribution density and position, and establishes the transmission line fracture risk situation analysis results, which are used in the automated operation and maintenance management of transmission business to determine the expansion path range of the hidden danger center, the node fracture risk distribution density and the overall risk zoning status of the line.

[0033] Compared with the prior art, the advantages and positive effects of the present invention are:

[0034] In the present invention, by extracting the inflection point changes from the tension change trajectory of the transmission line steel tower joints and the tension section hardware nodes, and performing standardized integrated comparison based on the node position-related potential mutation amplitude and the micro-meteorological wind speed increment, the risk pressure distribution of the node is accurately identified, which can achieve early detection of hidden danger nodes. By detecting the tension change rate of nodes where the pressure change exceeds the threshold, and dividing the stages according to the rate change characteristics, the node fracture evolution is staged, and the visualization and fine-grained analysis of the fault evolution process are enhanced. Further, the hidden danger center is screened by the pressure level change trend, and the convergence center is locked based on the difference in fluctuation amplitude, which improves the accuracy of locating the hidden danger concentration area. The expansion path is deduced by combining the risk pressure change rate of the hidden danger center and the surrounding nodes, and the node fracture risk distribution situation is dynamically evaluated, so that the hidden danger expansion trend warning of the transmission line has the characteristics of stage, evolution, and density. Through the whole process of data standardization processing, continuous trajectory analysis, change trend modeling and phased deduction, the overall accuracy and real-time performance of various links such as transmission line hidden danger detection, evolution warning, and risk zoning have been improved, providing high-value and high-reliability data support for operation and maintenance decision-making, and significantly enhancing the intelligence level and proactive prevention and control capabilities of operation and maintenance management. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 is a system flow chart of the present invention;

[0036] Figure 2 This is a flow chart of the node tension monitoring module of the present invention;

[0037] Figure 3 This is a flow chart of the risk pressure assessment module of the present invention;

[0038] Figure 4 This is a flow chart of the fault evolution stage inference module of the present invention;

[0039] Figure 5 This is a flow chart of the convergence center positioning module of the present invention;

[0040] Figure 6 This is a flow chart of the fracture evolution early warning module of the present invention. DETAILED DESCRIPTION

[0041] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0042] In the description of the present invention, it should be understood that the terms "length," "width," "up," "down," "front," "back," "left," "right," "vertical," "horizontal," "top," "bottom," "inside," "outside," and the like, indicating positions or relationships, are based on the positions or relationships shown in the accompanying drawings and are intended only to facilitate the description of the present invention and simplify the description. They do not indicate or imply that the devices or elements referred to must have a specific orientation, be constructed, or operate in a specific orientation. Therefore, they should not be construed as limiting the present invention. Furthermore, in the description of the present invention, "plurality" means two or more, unless otherwise expressly and specifically defined.

[0043] See also Figure 1 The big data-based automated operation and maintenance management system for power transmission services includes:

[0044] The node tension monitoring module obtains the tension records of the steel tower joints and tension section hardware nodes in the transmission line based on the big data platform. It detects the continuous change trajectory of the tension of each node within the time window, averages the continuous interval, performs piecewise linear fitting on the continuous average tension sequence, extracts the inflection point position in the change trajectory, integrates the turning characteristics of the tension change trend between inflection points, and generates the node tension inflection point change;

[0045] The risk pressure assessment module obtains the surface potential mutation amplitude of the node insulator and the micro-meteorological wind speed increment data based on the node position corresponding to each inflection point change in the node tension inflection point change. It then performs standardization processing on the tension inflection point change, potential mutation amplitude, and wind speed increment data. The three standardized data are superimposed and compared according to the node position. The basic risk pressure level of each node is determined based on the change trend after the superposition of the data. Nodes with a change degree exceeding the change threshold are selected as the pressure reference set to obtain the node risk pressure distribution.

[0046] The fracture evolution stage inference module obtains the node positions where the pressure change exceeds the change threshold in the node risk pressure distribution, traces back the change amount of the node tension inflection point of the corresponding node, performs tension change rate detection between consecutive nodes for the change amount sequence, groups and classifies them according to the change rate and span change characteristics, and divides them into slow decay stage, accelerated fission stage and critical fracture stage according to the change trend. The proportion of nodes in each stage is counted to generate the node fracture evolution stage interval;

[0047] The convergence center positioning module obtains nodes classified as the accelerated fission stage and critical fracture stage in the node fracture evolution stage, filters the location data of the corresponding nodes, obtains the node risk pressure distribution of the filtered nodes, detects the pressure level change trend between the filtered nodes, extracts the node with the largest pressure level fluctuation in the change trend as the candidate location of the hidden danger center, calls the risk pressure level of the nodes around the candidate location, compares the pressure change amplitude of the surrounding nodes with the pressure level change amplitude of the candidate location itself, selects the node with the smallest difference amplitude as the hidden danger center locking point, and generates the hidden danger convergence center positioning result;

[0048] The fracture evolution warning module is based on the hidden danger center location and surrounding node set in the hidden danger convergence center positioning results, and the relationship between the risk pressure change speed of the hidden danger center location and the surrounding nodes. It deduces the temporal evolution process of risk expansion, simulates the development path of the hidden danger center outward expansion and the node fracture trend distribution according to the synchronous change characteristics of the risk expansion speed and the stage division state, selects nodes with sudden changes in risk levels and transitions in evolution stages during the expansion process as warning targets, and evaluates the overall transmission line fracture risk distribution situation based on the distribution status of the warning target nodes, generating transmission line fracture risk situation analysis results;

[0049] The changes in node tension inflection points include the distribution of inflection point numbers, tension change amplitude range, and tension change inflection point positions. The node risk pressure distribution includes the node pressure extreme value position, pressure change rate range, and pressure change gradient trend. The node fracture evolution stage interval includes the slow decay stage node set, the accelerated fission stage node set, and the critical fracture stage node set. The hidden danger convergence center positioning results include the hidden danger center node position, the risk distribution status around the hidden danger center, and the pressure change amplitude of the hidden danger center. The transmission line fracture risk situation analysis results include the risk expansion path trajectory, the warning target node position, and the fracture risk distribution situation.

[0050] See also Figure 2 , the node tension monitoring module includes:

[0051] The tension data acquisition submodule acquires the tension record data of the steel tower joints and tension section hardware nodes in the transmission line based on the big data platform, detects the continuous change trajectory of the tension at each node within the time window, collects the node tension change sequence, extracts the continuous interval segmentation of the node tension change sequence, and generates a continuous record of the node tension;

[0052] First, set the time window for tension data collection, set the window range to 60 minutes, and record the node tension data every 5 minutes. Taking node A as an example, the tension values recorded at 0 minutes, 5 minutes, 10 minutes, 15 minutes, 20 minutes, and 25 minutes are 120.2kN, 121.5kN, 119.8kN, 122.1kN, 121.9kN, and 122.5kN respectively. Detect the continuous change trajectory of the tension value of each node within the time window, and draw the node tension change curve by connecting the tension values of adjacent time points. Then collect the node tension change sequence, that is, arrange and number all tension values in chronological order, and extract the continuous interval segmentation for the node tension change sequence. Set the tension change amount less than 3kN as As the interval division standard, in the data of node A, the changes in tension at adjacent time points were 1.3kN, -1.7kN, 2.3kN, -0.2kN, and 0.6kN, respectively, all less than 3kN. Therefore, the entire 60-minute data can be classified into the same continuous interval, generating a continuous record of node tension. The 3kN change threshold is set based on the following: the normal tension range of steel tower joints and tension section hardware nodes is generally 120kN to 140kN. Considering that short-term tension fluctuations caused by environmental vibration and wind loads usually do not exceed 2%-4% of the static tension of the node, taking the 120kN node static tension as the benchmark, 2% is 2.4kN, 4% is 4.8kN, and the median value of 3.6kN is rounded down to 3kN as the change threshold for continuous interval judgment.

[0053] The continuous trajectory processing submodule is based on the continuous recording of node tension. It performs interval averaging processing on the continuous tension change interval of each node, obtains the tension average value sequence of the continuous change interval, performs time smoothing processing based on the tension average value sequence, establishes the node continuous tension change trend curve, and generates the node tension continuous change trend according to the change direction of the node continuous tension change trend curve;

[0054] Based on the continuous recording of node tension, interval averaging is performed for the continuous tension change interval of each node. Taking the recorded data of node A as an example, the tension records from 0 to 25 minutes are 120.2kN, 121.5kN, 119.8kN, 122.1kN, 121.9kN, and 122.5kN. The corresponding interval average calculation formula is: ,in, It represents the average tensile force value of all nodes in the continuous interval, in kN. n represents the number of data points recorded in the current continuous interval, which is an integer. i represents the index number of each data point, which is numbered continuously starting from 1. i Indicates the tensile force corresponding to the data point with index i, in kN. ∑ is used to sum all the tensile forces within the specified range. Substituting the actual value: The average value of the tension in the continuous variation interval is 121.33 kN. Then, time smoothing is performed based on the tension average sequence, using a three-point sliding average. If the interval average values of node B at time 1, 2, and 3 are 119.6 kN, 120.2 kN, and 120.9 kN, respectively, the smoothed value at time 2 is: in, This represents the smoothed tension value (kN) calculated based on three adjacent points at the node at time 2. A complete smoothing sequence is generated in this way, establishing a continuous tension trend curve for the node. When the smoothed values of three consecutive points increase or decrease monotonically, the trend direction is determined. If the consecutive smoothed values are 120.0kN, 120.4kN, and 120.9kN, the trend is increasing. This ultimately generates a continuous tension trend for the node, providing the input basis for subsequent inflection point extraction.

[0055] The inflection point extraction submodule performs piecewise linear fitting on the trend curve based on the continuous change trend of the node tension, detects the position of the change inflection point in the fitting curve, extracts the turning characteristics of the tension change trend before and after the inflection point, and generates the node tension inflection point change by combining the number of inflection points and the distribution of the change amplitude;

[0056] Based on the continuous change trend of the node tension, a piecewise linear fitting is performed on the change trend curve. In the fitting process, every five consecutive sampling points are selected as an interval, and the least squares method is used for linear fitting. The slope change of the curve after fitting is used for inflection point detection. The inflection point is determined as the slope change rate exceeding 30%. The specific fitting equation is: F(t) = a×t+b, where F(t) represents the tension value of the node at time t, in kN, a represents the slope of the node tension change trend curve in the current section, in kN / min, b represents the intercept term of the node tension change trend curve in the current section, in kN, and t represents the time point, in minutes. The calculation formula of the slope a is: Where n represents the number of data points used for fitting, i represents the i-th sampling point, index number, numbered in chronological order, from 1 to n consecutively, t i Indicates the time value corresponding to the i-th time point, in minutes, F i Indicates the tension value corresponding to the i-th time point, in kN, ∑(t i F i ) represents the sum of the products of all i-th time points multiplied by their corresponding tension values, ∑(t i ) represents the sum of the time values of all i-th time points, ∑(F i ) represents the sum of the tension values corresponding to all i-th time points, It represents the sum of the squares of the values at all i-th time points. Taking the data recorded at five time points of node C as an example, assuming that the time points t are 0min, 5min, 10min, 15min, and 20min, and the tension values are 120kN, 121kN, 123kN, 125kN, and 126kN respectively, calculate the parameters: ∑t i

[0057] =0+5+10+15+20=50,∑F i =120+121+123+125+126=615,∑(t i F i )=(0×120)+(5×121)+(10×123)+(15×125)+(20×126)

[0058] =0+605+1230+1875+2520=6230, Substituting into the slope formula we get: That is, the slope a = 0.32 kN / min. If the slope obtained by fitting in the next section is a′ = -0.18 kN / min, the slope change rate is calculated as: That is, the rate of change is 156.25%, which is much larger than the set 30% change rate threshold. Therefore, an inflection point is detected at this node. The change characteristics before and after the inflection point are from rising to falling. The inflection point change amplitude threshold is set to 5kN. The setting basis is: when the transmission line is in operation, the tension change of the node under normal load usually fluctuates no more than 4%-6% of the node rated tension. Taking the node rated tension of 120kN as the benchmark, 4% is 4.8kN and 6% is 7.2kN. The lower limit standard is taken and rounded up to 5kN as the inflection point change amplitude threshold. If the tension difference before and after the inflection point exceeds 5kN, the inflection point is further confirmed to be valid. Assuming that the tension value before the inflection point is 126kN and the tension value after the inflection point is 121kN, the difference is 5kN, which meets the set standard. Finally, the effective inflection point information is extracted and the node tension inflection point change is generated as an important basis for fracture evolution inference.

[0059] See also Figure 3 , the risk pressure assessment module includes:

[0060] The inflection point change processing submodule obtains the node position corresponding to each inflection point change in the node tension inflection point change, detects the insulator surface potential mutation amplitude data and micro-meteorological wind speed increment data corresponding to the node position, performs standardization processing on the node tension inflection point change, potential mutation amplitude, and wind speed increment data, and generates a standardized change set;

[0061] First, obtain the node number and node coordinates, and record the node tension inflection point change and node position one by one. If the node numbers are set to A, B, and C, the corresponding changes are 3.5kN, 2.8kN, and 4.2kN respectively. At the same time, collect the surface potential mutation amplitude data of the insulator corresponding to the node position. Assume that the surface potential change amplitude of node A is 1.2kV, node B is 1.0kV, and node C is 1.5kV. At the same time, collect the wind speed increment data at the node under micrometeorological conditions. The wind speed increment at node A is 0.8m / s, the wind speed increment at node B is 0.6m / s, and the wind speed increment at node C is 1.0m / s. Then, perform standardization on the node tension inflection point change, potential mutation amplitude, and wind speed increment data respectively. The standardization process adopts the range standardization method. The specific standardization formula is: Among them, x′ is the standardized data value, the unit is consistent with the original data, x is the original sampling data, which represents the actual measurement value collected by a single node, and x min is the minimum value of the corresponding measurement values of all nodes in the sample set, x max is the maximum value of the corresponding measurement values of all nodes in the sample set. Taking the change in the inflection point of the node tension as an example, the maximum value of the sample is 4.2kN and the minimum value is 2.8kN. The normalized change of node A is: The potential mutation amplitude and wind speed increment data are standardized in the same way. After standardization, a standardized value set of all nodes in three dimensions is generated, and finally a standardized change set is formed.

[0062] The potential mutation analysis submodule is based on the standardized change set. It integrates and compares the data amplitudes of the corresponding positions of the node tension inflection point change, potential mutation amplitude, and wind speed increment standardized data according to the node position. It calls the compared data group and identifies the amplitude trend of continuous changes according to the node arrangement order to obtain the node change trend quantity.

[0063] Based on the standardized change set, the node tension inflection point change, potential mutation amplitude, and wind speed increment standardized data are integrated and compared according to the node position. The specific integration and comparison method is to calculate the comprehensive amplitude average of the standardized values of the three indicators at each node. The calculation formula for the comprehensive amplitude average is: Among them, M k is the average comprehensive amplitude of the kth node, indicating the comprehensive change intensity, S 1k is the change in the tensile inflection point of the kth node after normalization, and the unit is the normalized value (dimensionless), S 2k is the normalized potential mutation amplitude of the kth node, and the unit is the normalized value (dimensionless), S 3kis the normalized wind speed increment at the kth node, in units of normalized values (dimensionless), k is the node index number, numbered sequentially from 1 to N, and N is the total number of nodes. Taking node A as an example, if the three normalized values of node A are 0.5, 0.4, and 0.6 respectively, the average comprehensive amplitude of node A is: After calling the data group with comprehensive amplitude, the continuously changing amplitude trend is identified according to the node arrangement order. The continuous trend judgment rule is: if the comprehensive amplitude average value of three or more consecutive nodes is monotonically increasing or monotonically decreasing, it is defined as a continuously changing trend segment. If the comprehensive amplitudes of nodes A, B, and C are 0.5, 0.6, and 0.7 respectively, it is determined to be an increasing trend segment. If the comprehensive amplitudes of nodes A, B, and C are 0.6, 0.5, and 0.4, it is determined to be a decreasing trend segment. By traversing all node arrangements, the node change trend amount is finally obtained.

[0064] The risk pressure distribution generation submodule detects the difference between the change degree of each node and the change threshold based on the node change trend, calls the node position data whose change degree exceeds the change threshold, filters it as the pressure reference set, and establishes the node risk pressure distribution;

[0065] According to the node change trend, the difference between the change degree of each node and the change threshold is detected, and the change threshold is set to 0.2. The threshold is set based on the following: the standard deviation of the three data types of insulator surface potential mutation amplitude, node tension change, and wind speed increment is usually distributed in the range of 0.3 to 0.7, 0.5 to 1.0, and 0.4 to 0.8. The middle value is selected and scaled by 1 / 2 to set the threshold, that is, 0.2. The specific detection method is to calculate the absolute difference between the average comprehensive amplitude of each node and the average comprehensive amplitude of the adjacent nodes. If the difference exceeds the threshold, it is identified as a node with significant pressure change. For example, the comprehensive amplitude of node D is 0.8 and the comprehensive amplitude of node E is 0.5. The change is: ΔM = |0.8-0.5| = 0.3, where ΔM is the change in the comprehensive amplitude of adjacent nodes, and the unit is a standardized value (dimensionless). Since 0.3>0.2, node D is included in the pressure reference set. The position data of all qualified nodes are further called to screen the pressure reference set. The number of nodes in the pressure reference set is recorded as n. r , the node index number is k r ,Finally, the node risk pressure distribution is established. The distribution information takes the node index as the horizontal coordinate and the comprehensive amplitude change as the vertical coordinate to form a discrete distribution curve.

[0066] See also Figure 4 , the fault evolution stage inference module includes:

[0067] The pressure node extraction submodule obtains the node positions where the pressure change exceeds the change threshold in the node risk pressure distribution, traces back the node tension inflection point change of the corresponding node, calls the node tension inflection point change, detects the change range corresponding to the change according to the node sequence, filters out the nodes that match the pressure change characteristics, and generates a pressure characteristic node group;

[0068] Obtain the node positions where the pressure change exceeds the change threshold in the node risk pressure distribution. First, sort out all node risk pressure change values based on the node number sequence. Set the pressure change values of nodes P, Q, R, and S to 0.18, 0.26, 0.21, and 0.31, respectively. Set the pressure change threshold to 0.2. This threshold is generally around 0.25 based on the standard deviation of node pressure change. According to the 80% interval screening principle, the threshold of 0.2 is taken as the critical judgment limit. Detect the pressure change value of each node. If the node pressure change value is greater than 0.2, extract the node position. Node Q and node S meet the extraction conditions. Then trace back the node tension inflection point change data corresponding to node Q and node S, which are 4.1kN respectively. and 4.5kN, call the node tension inflection point variation, sort the extracted node tension inflection point variation according to the spatial arrangement order of the nodes, renumber them from small to large, node Q is ranked first, node S is ranked second, and for each node, according to its tension inflection point variation, detect the distribution of the variation within the interval, set the variation interval threshold to 1.0kN, according to the transmission line node in the steady state and perturbation state, the tension fluctuation range is allowed to be between 3% and 5%, take the node static tension of 120kN as the benchmark, 5% is 6kN, take one-sixth of it, that is, 1.0kN, as the small interval judgment standard, detect the difference between the tension inflection point variation of node Q and node S, the specific calculation process is: ΔF=|F S -F Q |=|4.5-4.1|=0.4kN, where ΔF is the difference in the change of the node tension inflection point, in kN, F S is the change in the tensile inflection point at node S, in kN, F Q is the change in the tension inflection point of node Q, in kN. ΔF = 0.4 kN, which is less than 1.0 kN. Therefore, nodes Q and S are classified into the same interval. Finally, nodes that match the pressure change characteristics are selected to form a pressure characteristic node group.

[0069] The tension rate detection submodule is based on the pressure characteristic node group. It constructs a continuous node sequence based on the tension inflection point change between nodes, detects the tension change rate between adjacent nodes, calls the change rate data, and aggregates the rate intervals according to the span characteristics of the node sequence to obtain the tension change rate interval.

[0070] Based on the pressure characteristic node group, a continuous node sequence is constructed for the change in the tension inflection point between nodes. The node number is uniformly recorded as j, and the change in the tension inflection point is recorded as F.j The rate of change of tension between adjacent nodes is calculated by the difference in tension change and the difference in distance between nodes. The rate calculation formula is: Among them, V j is the rate of change of tension between nodes j and j+1, in kN / m, F j is the change in the tensile inflection point of the jth node, in kN, F j+1 is the change in the tensile inflection point of the j+1th node, in kN, d j is the cumulative distance of the jth node in the line direction, in meters, d j+1 is the cumulative distance of the j+1th node in the line direction, in meters. Taking nodes Q and S as an example, assuming that the location of node Q is 1000m and the location of node S is 1050m, the change in tension at node Q is 4.1kN, and the change in tension at node S is 4.5kN, substituting into the formula yields: The tension change rate between nodes QS was found to be 0.008 kN / m. After calling all inter-node change rate data, rate intervals were grouped based on the span characteristics of the node sequence. The interval division criteria were set as follows: less than 0.01 kN / m was classified as a slow change interval, 0.01 kN / m to 0.05 kN / m was classified as a moderate change interval, and greater than 0.05 kN / m was classified as a sharp change interval. A node QS rate of 0.008 kN / m was classified as a slow change interval, ultimately resulting in a tension change rate interval dataset.

[0071] The stage interval division submodule divides the node stages according to the tension change rate interval and the rate change trend characteristics, classifies the nodes into slow decay stage, accelerated fission stage and critical fracture stage according to the trend, counts the proportion of nodes in each stage, and establishes the node fracture evolution stage interval;

[0072] According to the tension change rate interval, the node stages are divided according to the rate change trend characteristics. The stage division standard is set as follows: if the rates between consecutive nodes are less than 0.01kN / m, it is classified as the slow decay stage; if the consecutive node rates are in the range of 0.01kN / m to 0.05kN / m, it is classified as the accelerated fission stage; if the rate exceeds 0.05kN / m, it is classified as the critical fracture stage. The node index j is used as the traversal number, and the node is detected node by node. The node sequence tension change rate is set as follows: node QS rate 0.008kN / m, node ST rate 0.045kN / m, node TU rate 0.06kN / m. According to the judgment standard, node QS is classified into the slow decay stage, node ST is classified into the accelerated fission stage, and node TU is classified into the critical fracture stage. Then the number of nodes in each stage is counted, and the number of nodes in the slow decay stage is set as n decay , the number of nodes in the accelerated fission stage is naccelerate , the number of nodes in the critical fracture stage is n critical , the total number of nodes is n total , the calculation formula for the node proportion in each stage is: Among them, R decay is the proportion of nodes in the slow decay stage, R accelerate is the percentage of nodes in the accelerated fission stage, R critical is the percentage of nodes in the critical fracture stage, n decay is the number of nodes in the slow decay phase, n accelerate is the number of nodes in the accelerated fission stage, n critical is the number of nodes in the critical fracture stage, n total is the total number of nodes. Assuming that in a certain data set, the number of nodes is: 2 in the slow decay stage, 3 in the accelerated fission stage, and 1 in the critical fracture stage, and the total number of nodes is 6, then the proportion of each stage is as follows: Finally, based on the above-mentioned zoning divisions and statistical proportions, the node fracture evolution period intervals were established.

[0073] See also Figure 5 , the convergence center positioning module includes:

[0074] The hidden danger node screening submodule obtains the nodes in the accelerated fission stage and critical fracture stage in the node fracture evolution stage interval, screens the location data of the corresponding nodes, calls the node location data to obtain the node risk pressure distribution, detects the pressure level change trend between the screened nodes, determines the node sequence fluctuation amplitude according to the change amplitude, and generates a pressure change trend group;

[0075] Obtain the nodes in the accelerated fission stage and critical fracture stage in the node fracture evolution stage. First, screen out the node number sequence. Assume that nodes A, B, and C belong to the accelerated fission stage, and nodes D and E belong to the critical fracture stage. Screen out the position data corresponding to nodes A to E. The positions of nodes A, B, C, D, and E along the line are 200m, 250m, 300m, 350m, and 400m, respectively. Call the node position data and extract the risk pressure level data corresponding to the node based on the node number. Assume that the risk pressures of nodes A to E are 1.2kPa, 1.5kPa, 2.0kPa, 2.8kPa, and 3.1kPa, respectively. Detect the pressure level change trend between the screened nodes and calculate the pressure change of adjacent nodes. The change calculation formula is: ΔP h =P h+1 -P h , where ΔP h is the pressure change between the hth node and the h+1th node, in kPa, P h is the risk pressure level of the hth node, in kPa, Ph+1 is the risk pressure level of the h+1th node, in kPa, and h is the node sequence index. Taking nodes A and B as an example, calculate: ΔP A =1.5-1.2=0.3kPa, Node B to C: ΔP B

[0076] =2.0-1.5=0.5kPa, node C to D: ΔP C =2.8-2.0=0.8kPa, nodes D to E: ΔP D =3.1-2.8=0.3kPa, then the fluctuation amplitude of the node sequence is judged according to the change amplitude, and the fluctuation amplitude distinction standard is set. Below 0.2kPa is weak fluctuation, 0.2kPa to 0.7kPa is moderate fluctuation, and greater than 0.7kPa is violent fluctuation. The changes of nodes AB, BC, and DE are all in the moderate range, and the change of node CD of 0.8kPa is classified as violent fluctuation. Finally, the pressure change trend group is generated according to the fluctuation characteristics between nodes, and each node with the same continuous change characteristics is combined into a group to form the input data for subsequent hidden danger center candidate extraction.

[0077] The center candidate extraction submodule extracts the node with the largest pressure level fluctuation amplitude as the candidate location of the hidden danger center based on the pressure change trend group. It then calls the risk pressure level data of the nodes surrounding the candidate node and performs a corresponding change amplitude detection based on the fluctuation amplitude of the candidate location and the fluctuation amplitude of the surrounding nodes to obtain the hidden danger center difference sequence.

[0078] Based on the pressure change trend group, the node with the largest pressure level fluctuation amplitude is extracted as the candidate location of the hidden danger center. First, the pressure change trend group is traversed to find the node pair corresponding to the largest fluctuation amplitude. The change of node CD is 0.8kPa, which is the maximum value. Node C is determined to be the candidate location of the hidden danger center. The risk pressure level data of the nodes around node C are called, that is, node B (previous node) and node D (next node). The pressure of node B is 1.5kPa, the pressure of node C is 2.0kPa, and the pressure of node D is 2.8kPa. According to the corresponding detection of the pressure fluctuation amplitude between the candidate location and the surrounding nodes, the change amplitudes are calculated as follows: candidate location and previous node: ΔP C-B =|2.0-1.5|=0.5kPa, candidate position and next node: ΔP C-D =|2.8-2.0|

[0079] =0.8kPa, where ΔP C-B is the pressure change between candidate node C and the previous node B, ΔP C-Dis the pressure change amplitude between candidate node C and the subsequent node D. The two sets of difference data are combined into a hidden danger center difference sequence, recorded as (node B, 0.5kPa) and (node D, 0.8kPa), respectively. This sequence serves as the basis for subsequently identifying the hidden danger center.

[0080] The convergence center locking submodule determines the pressure change amplitude difference between the surrounding nodes and the candidate nodes based on the hidden danger center difference sequence, selects the node with the smallest amplitude difference as the final hidden danger center locking point, and establishes the hidden danger convergence center positioning result;

[0081] According to the hidden danger center difference sequence, the pressure change amplitude difference between the surrounding nodes and the candidate nodes is judged, and the node with the smallest amplitude difference is selected as the final hidden danger center locking point. First, the pressure change amplitude corresponding to each node in the hidden danger center difference sequence is compared. The change amplitude of node B is 0.5kPa, and the change amplitude of node D is 0.8kPa. The judgment benchmark is set as the minimum pressure change amplitude. The difference amplitude of node B is small, so node B is selected as the final hidden danger center locking point, and the hidden danger convergence center positioning result is established. The final hidden danger center positioning node number is B, the positioning position is 250m, and the risk pressure level is 1.5kPa. The positioning result is used for subsequent hidden danger expansion path deduction and fracture trend prediction basic data to complete the hidden danger node locking action.

[0082] See also Figure 6 , the fault evolution early warning module includes:

[0083] The expansion trend deduction submodule obtains the hidden danger center location and the surrounding node set from the hidden danger convergence center positioning result, calls the risk pressure change speed data of the hidden danger center location and the surrounding nodes, detects the time series process of risk expansion based on the relationship between the change speeds of the nodes, and deduces the synchronization characteristics of the expansion speed change and the stage division state to generate the risk expansion evolution trend;

[0084] Obtain the hidden danger center position and the surrounding node set in the hidden danger convergence center positioning result. First, determine the hidden danger center node number and its position. Assume that the hidden danger center node is recorded as C and the position is 320m. The surrounding nodes are recorded as B and D, and the positions are 300m and 340m respectively. Call the risk pressure change speed data of the hidden danger center position and the surrounding nodes. Assume that the risk pressure change speeds of nodes B, C, and D are 0.002kPa / min, 0.006kPa / min, and 0.004kPa / min respectively. According to the relationship between the change speeds of the nodes, detect the time series process of risk expansion, calculate the pressure speed change difference between the node pairs, and use the speed difference calculation formula: ΔV g =|V g+1 -V g |, where ΔV gis the pressure change rate difference between the gth node and the g+1th node, in kPa / min, V g is the pressure change rate of the g-th node, in kPa / min, V g+1 is the pressure change rate of the g+1th node, in kPa / min, g is the node sequence index, sorted by the nodes around the hidden danger center. Calculate the speed change difference between nodes B and C: ΔV B =|0.006-0.002|=0.004kPa / min, calculate the velocity change difference between nodes C and D: ΔV C =|0.004-0.006|=0.002kPa / min. According to the time series expansion judgment standard, the speed difference less than 0.003kPa / min is identified as the stable expansion process segment, and the speed difference greater than or equal to 0.003kPa / min is identified as the expansion change segment. Therefore, the segment from node B to C is classified as the expansion change segment, and the segment from node C to D is classified as the expansion stable segment. Subsequently, according to the expansion speed change and the current evolution stage of the node, its stage state change is synchronously detected. When the node expansion speed is higher than 0.005kPa / min, it is marked as a high expansion state, and when it is lower than 0.005kPa / min, it is marked as a slow expansion state. The speed of node C is 0.006kPa / min, which is marked as a high expansion state, and the speed of node D is 0.004kPa / min, which is marked as a slow expansion state. Finally, the risk expansion evolution trend of the hidden danger center and surrounding nodes is deduced, which is used as the basis for subsequent warning node screening.

[0085] The early warning target screening submodule extracts the node sequence where the risk level mutates during the expansion process based on the risk expansion evolution trend, detects whether the node corresponding to the evolution stage has a stage transition, and selects nodes with both risk level mutation and stage transition characteristics as early warning targets to obtain a stage transition node set;

[0086] Based on the risk expansion evolution trend, the node sequence with sudden changes in risk level during the expansion process is extracted. First, the nodes with sudden changes in risk level change rate during the expansion process are screened out. The mutation judgment standard is set as a rate change greater than 0.003kPa / min. According to the node expansion speed change detection result, the rate change between nodes BC is 0.004kPa / min, which meets the mutation standard. Therefore, nodes B and C enter the initial screening sequence. At the same time, it is detected whether the corresponding node evolution stage has a stage transition. The stage transition standard is defined as the node state changing from a slow expansion state to a high expansion state, or from a high expansion state to a slow expansion state. Node B is in a slow expansion state, and node C is in a high expansion state. A stage transition occurs. Node C meets the dual conditions of mutation + transition. Node D does not have a stage transition and does not meet the conditions. Therefore, node C is finally selected as the warning target node. The warning target node set is recorded as: G warning= {C}, where G warning is the set of nodes that have both phase transitions and sudden changes in risk levels. {} is a set symbol representing the set of node numbers. Node C meets the screening conditions for both sudden changes in risk levels and phase transitions, resulting in the set of phase transition nodes.

[0087] The risk situation assessment submodule detects the spatial distribution of warning target nodes on the transmission line based on the set of stage transition nodes, derives the overall line fracture risk distribution situation based on the relationship between node distribution density and position, and establishes the transmission line fracture risk situation analysis results. These results are used in the automated operation and maintenance management of transmission services to determine the expansion path range of the hidden danger center, the node fracture risk distribution density, and the overall risk zoning status of the line;

[0088] According to the set of stage transition nodes, the spatial distribution of the warning target nodes in the transmission line is detected. First, the spatial position of node C is obtained at 320m. According to the actual distribution range of the node in space, the risk zoning analysis interval standard is set. Every 100m is divided into a spatial segment. 320m falls in the fourth segment (300m-400m segment). The number of stage transition nodes in each segment is counted. If only node C has a transition in the segment, the transition node density is 1 / 100m. The overall line fracture risk distribution situation is derived based on the node distribution density. The distribution density is set to be greater than or equal to 1 / 100m as a high-risk area, less than 1 / 200m as a low-risk area, and between 1 / 100m and 1 / 20 The area between 0m and 10m is a medium-risk area, and the section where node C is located is classified as a high-risk area. At the same time, the hidden danger expansion path is derived according to the node position relationship. The path direction is connected from the hidden danger center to the jump node, and a one-way expansion path is constructed from the node center C to the direction of node C itself. Combined with the node fracture risk density and path direction, the risk zoning status of the entire transmission line is further divided. The high-risk area is defined as the node dense distribution area, the medium-risk area is defined as the node relatively sparse distribution area, and the low-risk area is defined as the non-jump node distribution area. Finally, the transmission line fracture risk situation analysis results are established, which are used in the automated operation and maintenance management of transmission business to determine the expansion path range of the hidden danger center, the node fracture risk distribution density and the overall risk zoning status of the line.

[0089] The above are merely preferred embodiments of the present invention and do not limit the present invention in any other form. Any technician familiar with the profession may use the technical content disclosed above to change or modify it into an equivalent embodiment with equivalent changes and apply it to other fields. However, any simple modification, equivalent change and modification made to the above embodiment based on the technical essence of the present invention without departing from the content of the technical solution of the present invention shall still fall within the scope of protection of the technical solution of the present invention.

Claims

1. The power transmission business automated operation and maintenance management system based on big data is characterized by: The system comprises: The node tension monitoring module obtains the tension records of the steel tower joints and tension section hardware nodes in the transmission line, detects the change trajectory of the node tension within the time window, and generates the node tension inflection point change; The risk pressure assessment module is based on the node position corresponding to each inflection point change in the node tension inflection point change, combined with the node potential mutation amplitude and micro-meteorological wind speed increment, and then normalized and superimposed for comparison to obtain the node risk pressure distribution; The fracture evolution stage inference module obtains the node position where the pressure change exceeds the change threshold in the node risk pressure distribution, traces back the change of the tension inflection point, detects the tension change rate between consecutive nodes, and divides the node fracture evolution stage interval according to the rate and span change characteristics; The convergence center positioning module obtains the nodes classified as the accelerated fission stage and the critical fracture stage in the node fracture evolution stage interval, determines the node hidden danger center, and generates the hidden danger convergence center positioning result; The fracture evolution warning module simulates the hazard expansion path and node fracture trend distribution based on the hazard center position and surrounding node set in the hazard convergence center positioning result, and generates a transmission line fracture risk situation analysis result.

2. The big data-based automated operation and maintenance management system for power transmission services according to claim 1, characterized in that: The node tension inflection point changes include the distribution of inflection point numbers, tension change amplitude range, and tension change inflection point positions; the node risk pressure distribution includes the node pressure extreme value position, pressure change rate range, and pressure change gradient trend; the node fracture evolution stage interval includes the slow decay stage node set, the accelerated fission stage node set, and the critical fracture stage node set; the hidden danger convergence center positioning results include the hidden danger center node position, the risk distribution status around the hidden danger center, and the pressure change amplitude of the hidden danger center; the transmission line fracture risk situation analysis results include the risk expansion path trajectory, the warning target node position, and the fracture risk distribution situation.

3. The big data-based automated operation and maintenance management system for power transmission services according to claim 1, characterized in that: The node tension monitoring module includes: The tension data acquisition submodule acquires the tension record data of the steel tower joints and tension section hardware nodes in the transmission line based on the big data platform, detects the continuous change trajectory of the tension at each node within the time window, collects the node tension change sequence, extracts the continuous interval segmentation of the node tension change sequence, and generates a continuous record of the node tension; The continuous trajectory processing submodule performs interval averaging processing on the continuous tension change interval of each node based on the continuous record of the node tension, obtains a tension average value sequence of the continuous change interval, performs time smoothing processing based on the tension average value sequence, establishes a node continuous tension change trend curve, and generates a node tension continuous change trend based on the change direction of the node continuous tension change trend curve; The change inflection point extraction submodule performs piecewise linear fitting on the change trend curve based on the continuous change trend of the node tension, detects the change inflection point position in the fitting curve, extracts the turning characteristics of the tension change trend before and after the inflection point, and generates the node tension inflection point change by combining the number of inflection points and the distribution of change amplitude.

4. The big data-based automated operation and maintenance management system for power transmission services according to claim 3, characterized in that: The risk pressure assessment module includes: The inflection point change processing submodule obtains the node position corresponding to each inflection point change in the node tension inflection point change, detects the insulator surface potential mutation amplitude data and micrometeorological wind speed increment data corresponding to the node position, performs normalization processing on the node tension inflection point change, potential mutation amplitude, and wind speed increment data, and generates a standardized change set; Based on the standardized change set, the potential mutation analysis submodule integrates and compares the data amplitudes of the corresponding positions of the node tension inflection point change, potential mutation amplitude, and wind speed increment standardized data according to the node position, calls the compared data group, identifies the amplitude trend of continuous change according to the node arrangement order, and obtains the node change trend amount; The risk pressure distribution generation submodule detects the difference between the change degree of each node and the change threshold based on the node change trend, calls the node position data whose change degree exceeds the change threshold, filters it as the pressure reference set, and establishes the node risk pressure distribution.

5. The big data-based automated operation and maintenance management system for power transmission services according to claim 4, characterized in that: The fault evolution stage inference module includes: The pressure node extraction submodule obtains the node positions in the node risk pressure distribution where the pressure change exceeds the change threshold, traces back the node tension inflection point change of the corresponding node, calls the node tension inflection point change, detects the change interval range corresponding to the change according to the node sequence, selects nodes that match the pressure change characteristics, and generates a pressure characteristic node group; The tension rate detection submodule constructs a continuous node sequence based on the pressure characteristic node group and the change amount of the tension inflection point between nodes, detects the tension change rate between adjacent nodes, calls the change rate data, and aggregates the rate intervals according to the span characteristics of the node sequence to obtain the tension change rate interval; The stage interval division submodule divides the node stages according to the tension change rate interval and the rate change trend characteristics, classifies the nodes into slow decay stage, accelerated fission stage and critical fracture stage according to the trend, counts the proportion of nodes in each stage, and establishes the node fracture evolution stage interval.

6. The big data-based automated operation and maintenance management system for power transmission services according to claim 5, characterized in that: The convergence center positioning module includes: The hidden danger node screening submodule obtains the nodes in the accelerated fission stage and critical fracture stage in the node fracture evolution stage interval, screens the position data of the corresponding nodes, calls the node position data to obtain the node risk pressure distribution, detects the pressure level change trend between the screened nodes, determines the node sequence fluctuation amplitude according to the change amplitude, and generates a pressure change trend group; Based on the pressure change trend group, the center candidate extraction submodule extracts the node with the largest pressure level fluctuation amplitude as the hidden danger center candidate location, calls the risk pressure level data of the nodes surrounding the candidate node, and performs a corresponding change amplitude detection based on the fluctuation amplitude of the candidate location and the fluctuation amplitude of the surrounding nodes to obtain the hidden danger center difference sequence; The convergence center locking submodule determines the pressure change amplitude difference between the surrounding nodes and the candidate nodes according to the hidden danger center difference sequence, selects the node with the smallest amplitude difference as the final hidden danger center locking point, and establishes the hidden danger convergence center positioning result.

7. The big data-based automated operation and maintenance management system for power transmission services according to claim 6, characterized in that: The fault evolution early warning module includes: The expansion trend deduction submodule obtains the hidden danger center position and the surrounding node set in the hidden danger convergence center positioning result, calls the risk pressure change speed data of the hidden danger center position and the surrounding nodes, detects the time series process of risk expansion based on the relationship between the change speeds between nodes, and deduces based on the synchronization characteristics of the expansion speed change and the stage division state to generate the risk expansion evolution trend; Based on the risk expansion evolution trend, the early warning target screening submodule extracts the node sequence where the risk level has a sudden change during the expansion process, detects whether the corresponding evolution stage of the node has a stage transition, and selects nodes with both risk level mutation and stage transition characteristics as early warning targets to obtain a stage transition node set; The risk situation assessment submodule detects the spatial distribution of warning target nodes in the transmission line based on the set of stage transition nodes, derives the overall line fracture risk distribution situation based on the relationship between node distribution density and position, and establishes the transmission line fracture risk situation analysis results, which are used in the automated operation and maintenance management of transmission business to determine the expansion path range of the hidden danger center, the node fracture risk distribution density and the overall risk zoning status of the line.

Images