A method and system for evaluating the safe operation risk of a power transmission line in a mining influence area

Abstract

The application provides a kind of mining influence area transmission line safe operation risk assessment method and system, belongs to mining influence area transmission line safe operation risk assessment technical field;To realize the risk assessment goal of multi-source heterogeneous data fusion, nonlinear evaluation, weight dynamic self-adaption, uncertainty quantification and digital twin visualization, a hierarchical factor system containing geological mining static factor, tower body static factor, meteorological environment dynamic factor and space-time correlation factor is constructed, and each factor is quantified and normalized;The combination weight is obtained by solving the Nash equilibrium of subjective weight and objective weight using game theory;Calculate the preliminary risk score, and adaptively change the weight of dynamic factors in the high risk interval;Risk level is determined by fuzzy comprehensive evaluation, and evaluation confidence is calculated by extracting cloud model characteristic parameters;Combined with CUSUM algorithm for mutation detection and early warning;The application is applied to mining influence area transmission line safe operation risk assessment.

Description

Technical Field

[0001] This invention provides a method and system for risk assessment of safe operation of transmission lines in mining-affected areas, belonging to the technical field of risk assessment of safe operation of transmission lines in mining-affected areas. Background Technology

[0002] Mining-affected areas are prone to surface movement and deformation caused by underground coal seam mining, which generates additional stress on the foundations of power transmission line towers. In severe cases, this can lead to tower tilting, foundation cracking, or even tower collapse. Therefore, a thorough risk assessment should be conducted before construction begins on the installation sites of power transmission line towers.

[0003] However, the current risk assessment of power transmission lines in mining areas still has the following deficiencies or shortcomings:

[0004] First, existing methods mostly focus on single geological and mining conditions or rely solely on InSAR settlement monitoring, without systematically integrating the structural properties of the tower itself, dynamic meteorological and hydrological conditions, and the mining deformation process. This results in a single dimension of risk assessment, subjective assessment results, and large errors.

[0005] Second, in terms of weight determination, the traditional analytic hierarchy process is highly subjective and difficult to adapt to the dynamic changes of risk control factors at different stages of mining. While the purely objective entropy weight method can reflect the data distribution, it may deviate from the physical mechanism. The existing linear weighted fusion method has failed to effectively handle the conflict and trade-off between subjective and objective weights.

[0006] Third, the existing weighted summation model has a linear structure, which makes it difficult to characterize the nonlinear coupling effect of multiple factors under extreme conditions, such as the "1+1>2" characteristic of risk mutation when heavy rainfall and high displacement rate occur simultaneously.

[0007] Fourth, most assessment methods only establish static assessment models and cannot use newly added hidden danger verification data to reversely optimize weights and thresholds, resulting in a decrease in the accuracy of the model after long-term use.

[0008] Fifth, existing technologies output a single risk score, lacking analysis of uncertainties caused by data noise and model errors, which is not conducive to hierarchical risk management in operation and maintenance decisions.

[0009] In addition, patent CN113902314B discloses a method for assessing geological disaster risks of transmission lines based on expert scoring, but its factor weights are fixed and it is difficult to adapt to dynamic changes under different geological conditions; patent CN114200447B discloses a method for monitoring settlement of transmission lines based on InSAR technology, which focuses on post-event monitoring and lacks pre-event prediction and comprehensive risk quantification models, neither of which can solve the above-mentioned defects. Summary of the Invention

[0010] To address the technical problems existing in the background art, the present invention provides a method for assessing the safe operation risk of transmission lines in mining-affected areas, comprising the following assessment steps:

[0011] Step S1: Construct a hierarchical evaluation factor system that includes geological and mining static factors, tower body static factors, meteorological and environmental dynamic factors, and spatiotemporal correlation factors. Quantify and normalize each factor to obtain the normalized value N. i ;

[0012] Step S2: Determine the subjective weights W using the Analytic Hierarchy Process (AHP). si An improved entropy weight method was used to determine the objective weight W. oi By solving the Nash equilibrium objective function that minimizes the deviation between the combined weights and the subjective and objective weights, the game-theoretic combined weights W are obtained. i ;

[0013] Step S3: Calculate the preliminary risk score. The calculation formula is as follows:

[0014] R _pre =Σ(W i ·N i );

[0015] If R_ pre If the value is ≥0.6, then the weights of each factor in the dynamic factor set will be increased according to the following formula:

[0016] W i '=W i ·(1 + λ·I(R _pre ≥0.6)·I);

[0017] Where i∈Dynamic Factor Set, λ is the variable weight strength coefficient, and I(·) is the indicator function; after boosting, renormalize so that the sum of all weights is 1;

[0018] Step S4: Using the fuzzy comprehensive evaluation method, construct the membership function of each factor to each risk level, calculate the membership vector using the weighted fuzzy synthesis operator, and determine the risk level according to the principle of maximum membership.

[0019] Extracting cloud model feature parameters for risk scoring, including expected value E. x Entropy E n Hyperentropy H e And calculate the evaluation confidence level Conf, the calculation formula is:

[0020] Conf = 1-E n / E n_max ;

[0021] Among them, E n_maxThe maximum permissible entropy;

[0022] Step S5: Use a sliding window for online learning, maintain the most recent M verified samples, and trigger incremental entropy weight calculation every K new samples to update the objective weight W. oi ;

[0023] Perform regular full updates. Every quarter or when the model precision or recall drops below a set threshold, re-execute steps S2 to S4 using the full historical database to reconstruct the weight and membership function parameters.

[0024] Change point detection and early warning are carried out. The CUSUM algorithm is used to monitor the time series of surface displacement change rate and displacement acceleration. When a significant change is detected, the model early warning level is temporarily raised by one level.

[0025] Step S6: Perform spatiotemporal simulation and realize digital twin visualization. Use the Kriging method to estimate the risk of any point outside the tower location, generate a global risk heat map, introduce the time dimension, and form a risk spatiotemporal evolution sequence.

[0026] In step S1:

[0027] The geological and mining static factors include: mining depth to thickness ratio, distance between tower and goaf boundary, stratum dip angle, fault influence coefficient, residual slope layer thickness ratio, and mining fullness coefficient.

[0028] The static factors of the tower body include: tower importance coefficient, foundation type coefficient, initial slope, and tower service life reduction coefficient;

[0029] The meteorological environmental dynamic factors include: effective cumulative rainfall, soil moisture saturation, surface displacement rate, and displacement acceleration.

[0030] The spatiotemporal correlation factors include: the risk coupling coefficient between adjacent towers and the frequency of historical risks.

[0031] The expression for the objective function is:

[0032] min||W-Ws||²+||W-Wo||²;

[0033] The combined weight W is obtained by solving.

[0034] In step S1, the normalization function for the sufficiency coefficient adopts an S-shaped growth curve function, the expression of which is:

[0035] N = 1 / (1+e^(-k·(q-q0)));

[0036] Where q is the ratio of the mining width to the critical mining width.

[0037] The displacement acceleration in step S1 is the second-order difference of the rate of change of surface displacement. When the average acceleration over three consecutive days is positive and exceeds the threshold, a risk enhancement flag is triggered.

[0038] The improved entropy weight method used in step S2 is specifically as follows:

[0039] Introducing the information entropy correction formula:

[0040] e _i =-1 / ln(n)·Σ(p _ij ·ln(p_ ij +ε));

[0041] Where ε is a very small positive number.

[0042] The membership function constructed in step S4 is specifically a Gaussian membership function, with the following expression:

[0043] μ(x) = exp(-(xc)² / (2σ²));

[0044] Where c is the level center value and σ is the level width parameter.

[0045] The system used for risk assessment of the safe operation of transmission lines in mining-affected areas includes:

[0046] The data acquisition and edge preprocessing module, deployed on the end side with GNSS, rain gauge, soil moisture sensor, and inclinometer, and on the edge side with edge computing gateway, is used to acquire dynamic factor data in real time and perform anomaly removal and CUSUM mutation detection.

[0047] The factor quantification and database management module stores geographic information, geological and mining data, tower ledgers, real-time monitoring data and historical risk data, and supports hot-swappable configuration of new factors.

[0048] The game theory combined weight calculation module integrates the analytic hierarchy process expert evaluation interface, the improved entropy weight method automatic calculation unit, and the game theory fusion calculation unit.

[0049] The fuzzy cloud risk assessment engine performs fuzzy comprehensive evaluation and cloud model feature extraction, and outputs the risk level and confidence level.

[0050] The dual evolutionary learning module includes a sliding window manager, periodic retraining triggers, and a change point detector;

[0051] The 3D digital twin visualization and decision-making module is based on a 3D GIS engine to construct risk heat maps, render tower models, and overlay surface deformation fields, and supports risk dashboard display.

[0052] The early warning release and feedback closed-loop module connects with the inspection terminal APP to push verification tasks and receive verification results, generating positive and negative sample feedback.

[0053] The 3D digital twin visualization and decision-making module also includes a multi-scenario early warning push unit, with the following information push rules:

[0054] When the confidence level is ≥0.7 and the risk level is high or very high, a confirmation alert will be sent.

[0055] A warning will be sent when the confidence level is less than 0.7 and the risk score is greater than 0.6.

[0056] When CUSUM mutation detection is triggered, a trend mutation warning will be pushed out.

[0057] The edge computing gateway has a built-in CUSUM algorithm parameter adaptive adjustment unit that automatically calculates the detection threshold based on the variance of historical data.

[0058] The purpose of this invention is to provide a method and system for risk assessment of safe operation of transmission lines in mining-affected areas, capable of multi-source heterogeneous data fusion, nonlinear evaluation, dynamic weight adaptation, uncertainty quantification, and digital twin visualization. Compared with existing technologies, it has the following advantages:

[0059] I. Comprehensive and in-depth factor system: This invention adds a mining sufficiency coefficient, displacement acceleration, adjacent tower risk coupling coefficient and multi-source consistency index, which can characterize the mining deformation process from four dimensions: degree, rate, trend and spatial correlation.

[0060] II. More Scientific Weight Determination: This invention adopts game theory to integrate subjective and objective weights, which solves the subjective problem of choosing the linear weighting coefficient α, and the combined weights are more robust.

[0061] Third, it has the ability to adapt and change weights according to risk level: This invention automatically increases the weights of dynamic factors in high-risk intervals, which solves the problem that conventional fixed weights are not responsive enough to sudden environmental factors.

[0062] IV. Nonlinear evaluation is closer to reality: This invention adopts a combination of fuzzy comprehensive evaluation and cloud model to characterize the nonlinear interaction between factors and evaluation uncertainty, and outputs dual information of level and confidence.

[0063] V. The model has dual evolution capabilities: This invention adopts online updating via sliding window and periodic full reconstruction, which can maintain the model's accuracy over a long period of time, and uses CUSUM mutation detection to improve the response speed to sudden deformations;

[0064] VI. Advanced System Architecture: This invention adopts cloud-edge-device collaboration to reduce cloud computing pressure, and real-time mutation detection at the edge to improve early warning timeliness;

[0065] VII. Richer Decision-Making Information: This invention outputs features such as risk level, confidence level, contribution of main control factors, and spatiotemporal evolution sequence, which can support hierarchical early warning strategies. Attached Figure Description

[0066] The present invention will be further described below with reference to the accompanying drawings:

[0067] Figure 1 This is a flowchart illustrating the steps of the risk assessment method for safe operation of transmission lines in the mining impact zone according to the present invention. Detailed Implementation

[0068] like Figure 1 As shown, the risk assessment method for safe operation of transmission lines in mining-affected areas provided by this invention mainly includes: constructing a hierarchical factor system comprising static geological and mining factors, static tower body factors, dynamic meteorological and environmental factors, and spatiotemporal correlation factors, and quantifying and normalizing each factor; using game theory to solve the Nash equilibrium and integrating subjective and objective weights to obtain combined weights; calculating a preliminary risk score and adaptively adjusting the dynamic factors in high-risk intervals; using fuzzy comprehensive evaluation to determine the risk level and extracting cloud model feature parameters to calculate the assessment confidence level; achieving dual evolution of the model through sliding window online learning and periodic full updates, and combining the CUSUM algorithm for mutation detection and early warning; the assessment system provided by this invention adopts a cloud-edge-device architecture, integrates a three-dimensional digital twin visualization module, and can realize multi-dimensional, nonlinear, and adaptive assessment of the risks of transmission lines in mining-affected areas. The assessment results have quantifiable confidence and the model has evolutionary capabilities.

[0069] Furthermore, the dynamic quantitative assessment method for the operation risk of transmission lines in coal mine mining-affected areas provided by this invention specifically includes the following assessment steps:

[0070] Step S1: For the collection and quantification of multi-source risk factors, including:

[0071] A hierarchical evaluation factor system was constructed, comprising static geological and mining factors, static tower body factors, dynamic meteorological and environmental factors, and spatiotemporal correlation factors. Each factor was quantified and normalized to obtain a normalized value N. i .

[0072] The static factors in geological mining include:

[0073] Mining depth to thickness ratio: based on actual measured values;

[0074] The shortest horizontal distance between the tower and the boundary of the goaf: the normalized value adopts the negative exponential function N = e^(-d / D0), where D0 is the radius of influence;

[0075] Dip angle of formation: Take the measured angle value;

[0076] Fault influence coefficient: The value is taken from a table based on the relative position of the fault and the tower, the fault displacement and the dip angle, and is between 0 and 1.

[0077] Residual slope deposit thickness ratio: The ratio of residual slope deposit thickness to total overburden thickness;

[0078] Fully utilized mining coefficient: the ratio of mining width to critical mining width, normalized using an S-shaped growth curve function.

[0079] N = 1 / (1 + e^(-k·(q - q0)));

[0080] Where q is the ratio of the mining width to the critical mining width.

[0081] The static factors of the tower body include:

[0082] Importance coefficient of tower: assigned according to voltage level and tower type;

[0083] Basic type coefficient: assigned a value based on the resistance to deformation;

[0084] Initial slope: Normalized measured values;

[0085] Tower service life reduction factor: based on the exponential decay model of material fatigue.

[0086] Dynamic meteorological and environmental factors include:

[0087] Effective cumulative rainfall: expressed as P _eff = Σ(k^t·P _t The attenuation coefficient k=0.9;

[0088] Soil moisture saturation: The expression is S = θ / θ _sat ;

[0089] Surface displacement rate: GNSS daily average rate;

[0090] Displacement acceleration: the second difference of the rate of change of surface displacement. When the average acceleration over three consecutive days is positive and exceeds the threshold, a risk enhancement flag is triggered.

[0091] Spatiotemporal correlation factors include:

[0092] Risk coupling coefficient between adjacent towers: risk transmission impact based on spatial interpolation;

[0093] Historical risk frequency: The number of risk warnings issued for the same tower in the past 24 months.

[0094] Step S2: Calculate the combined weights based on game theory, including:

[0095] The subjective weights Wsi were determined using the analytic hierarchy process: experts were invited to compare the relative importance of factors within the same level pairwise, a judgment matrix was constructed, and the subjective weights were obtained after passing the consistency test.

[0096] The objective weight Woi is determined by the improved entropy weight method: historical sample data is collected, and the information entropy of each factor is calculated. The smaller the information entropy, the greater the dispersion of the factor and the greater the objective weight.

[0097] The improved entropy weight method introduces an information entropy correction formula: e _i = -1 / ln(n)·Σ(p _ij ·ln(p _ij +ε)), where ε is a very small positive number to avoid the undefined problem caused by a zero value.

[0098] Treating subjective and objective weights as the strategies of both sides in the game, we seek the Nash equilibrium solution that minimizes the deviation between the combined weights and the subjective and objective weights. The expression is:

[0099] Objective function: min||W-Ws||²+||W-Wo||²;

[0100] The combined weight W is obtained by solving.

[0101] Step S3: Adaptive weighting of risk levels, including:

[0102] Calculate the preliminary risk score using the formula R. _pre =Σ(W i ·N i );

[0103] If R _pre If the value is ≥0.6 (high-risk range or above), then the weights of each factor in the dynamic factor set will be increased according to the following formula:

[0104] W i '=W i ·(1 + λ·I(R _pre ≥0.6)·I(i∈Dynamic Factor Set))

[0105] Where λ is the variable weight intensity coefficient, ranging from 0.2 to 0.5, and I(·) is the indicator function; after boosting, it is renormalized so that the sum of all weights is 1.

[0106] This mechanism reflects the physical principle that environmental disturbances become the main cause under high stress conditions.

[0107] Step S4: Conduct risk assessment based on fuzzy cloud, including:

[0108] The fuzzy comprehensive evaluation method is adopted: A factor set U (all factors) and a comment set V = {low risk, medium risk, high risk, extremely high risk} are established. Membership functions for each factor to each level are constructed, using Gaussian membership functions.

[0109] μ(x) = exp(-(xc)² / (2σ²));

[0110] Where c is the level center value and σ is the level width parameter. The weighted fuzzy synthesis operator B=WR (weighted average type) is used to calculate the membership vector, and the risk level is determined according to the principle of maximum membership.

[0111] To represent the uncertainty of the assessment, cloud model feature parameters for risk scoring are extracted, including:

[0112] Expected E x =R (weighted score);

[0113] Entropy E n : Reflects the ambiguity caused by the dispersion of factor values;

[0114] Hyperentropy H e : Reflects the stability of the evaluation system;

[0115] Calculate the confidence level: Conf = 1 - E n / E n_max E n_max This represents the maximum permissible entropy.

[0116] The final output of this step is the risk level and confidence level.

[0117] Step S5: Employ a dual evolutionary mechanism for the model, including:

[0118] Online learning using a sliding window: Maintain the most recent M verified samples (positive and negative samples). Incremental entropy weight calculation is triggered every K new samples, updating only the objective weight W. oi The update frequency can be daily or weekly.

[0119] Regular full updates: Every quarter, or when the model precision or recall drops by more than 5%, re-execute steps S2 to S4 using the full historical database to reconstruct the weight and membership function parameters.

[0120] Change point detection and early warning: The CUSUM algorithm is used to monitor the time series of surface displacement change rate and displacement acceleration. When a significant change is detected, the model early warning level is temporarily raised by one level, bypassing the regular evaluation cycle.

[0121] The rules for managing positive and negative samples are as follows:

[0122] Positive samples: those identified as high-risk by the model and confirmed by on-site verification;

[0123] Negative samples: The model identifies a high risk, but on-site verification confirms it to be a false alarm;

[0124] Strong negative samples: The model judges the risk to be low, but the actual danger occurs (missed report).

[0125] Step S6: Perform spatiotemporal simulation to achieve digital twin visualization, including using the Kriging method to estimate the risk at any point outside the tower location and generating a global risk heat map. Introduce the time dimension to form a spatiotemporal evolution sequence of risk.

[0126] This invention also provides a risk assessment system for the safe operation of transmission lines in mining-affected areas, implementing the above-described method, which adopts a cloud-edge-device collaborative architecture, including:

[0127] GNSS receivers, rain gauges, soil moisture sensors, and inclinometers are deployed at the end to collect dynamic factor data in real time.

[0128] An edge computing gateway is deployed on the edge to perform data preprocessing, outlier removal, and CUSUM mutation detection. The edge gateway has a built-in adaptive adjustment unit for CUSUM algorithm parameters, which automatically calculates the detection threshold based on the variance of historical data.

[0129] Deploy a risk assessment center on the cloud side, including the following modules:

[0130] Factor Quantification and Database Management Module: Stores geographic information, geological and mining data, tower ledgers, real-time monitoring data, and historical risk data, and supports hot-swappable configuration of new factors.

[0131] Game theory combined weight calculation module: integrates the analytic hierarchy process expert evaluation interface, the improved entropy weight method automatic calculation unit, and the game theory fusion calculation unit.

[0132] Fuzzy Cloud Risk Assessment Engine: Performs fuzzy comprehensive evaluation and cloud model feature extraction, and outputs risk level and confidence level.

[0133] Dual evolutionary learning module: includes a sliding window manager, periodic retraining triggers, and a change point detector.

[0134] The 3D digital twin visualization and decision-making module: Based on a 3D GIS engine, it constructs risk heat maps, renders tower models, and overlays surface deformation fields, supporting risk dashboard display; it includes a multi-scenario early warning push unit, with the following information push rules:

[0135] A confirmation alert will be sent when the confidence level is ≥0.7 and the risk level is high or very high.

[0136] A warning is sent when the confidence level is <0.7 and the risk score is >0.6 (the confidence level is low).

[0137] A trend mutation alert is pushed out when CUSUM mutation detection is triggered.

[0138] Early warning release and feedback closed-loop module: It connects with the inspection terminal APP to push verification tasks and receive verification results, forming positive and negative sample feedback.

[0139] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for assessing the safe operation risk of transmission lines in mining-affected areas, characterized in that: The evaluation steps include the following: Step S1: Construct a hierarchical evaluation factor system that includes geological and mining static factors, tower body static factors, meteorological and environmental dynamic factors, and spatiotemporal correlation factors. Quantify and normalize each factor to obtain the normalized value N. i ; Step S2: Determine the subjective weights W using the Analytic Hierarchy Process (AHP). si An improved entropy weight method was used to determine the objective weight W. oi By solving the Nash equilibrium objective function that minimizes the deviation between the combined weights and the subjective and objective weights, the game-theoretic combined weights W are obtained. i ; Step S3: Calculate the preliminary risk score. The calculation formula is as follows: R _pre =Σ(W i ·N i ); If R_ pre If the value is ≥0.6, then the weights of each factor in the dynamic factor set will be increased according to the following formula: W i '=W i ·(1 + λ·I(R _pre ≥0.6)·I); Where i∈Dynamic Factor Set, λ is the variable weight strength coefficient, and I(·) is the indicator function; after boosting, renormalize so that the sum of all weights is 1; Step S4: Using the fuzzy comprehensive evaluation method, construct the membership function of each factor to each risk level, calculate the membership vector using the weighted fuzzy synthesis operator, and determine the risk level according to the principle of maximum membership. Extracting cloud model feature parameters for risk scoring, including expected value E. x Entropy E n Hyperentropy H e And calculate the evaluation confidence level Conf, the calculation formula is: Conf =1-E n / AND n_max ; Among them, E n_max The maximum permissible entropy; Step S5: Use a sliding window for online learning, maintain the most recent M verified samples, and trigger incremental entropy weight calculation every K new samples to update the objective weight W. oi ; Perform regular full updates. Every quarter or when the model precision or recall drops below a set threshold, re-execute steps S2 to S4 using the full historical database to reconstruct the weight and membership function parameters. Change point detection and early warning are carried out. The CUSUM algorithm is used to monitor the time series of surface displacement change rate and displacement acceleration. When a significant change is detected, the model early warning level is temporarily raised by one level. Step S6: Perform spatiotemporal simulation and realize digital twin visualization. Use the Kriging method to estimate the risk of any point outside the tower location, generate a global risk heat map, introduce the time dimension, and form a risk spatiotemporal evolution sequence.

2. The method for risk assessment of safe operation of transmission lines in mining-affected areas according to claim 1, characterized in that: In step S1: The geological and mining static factors include: mining depth to thickness ratio, distance between tower and goaf boundary, stratum dip angle, fault influence coefficient, residual slope layer thickness ratio, and mining fullness coefficient. The static factors of the tower body include: tower importance coefficient, foundation type coefficient, initial slope, and tower service life reduction coefficient; The meteorological environmental dynamic factors include: effective cumulative rainfall, soil moisture saturation, surface displacement rate, and displacement acceleration. The spatiotemporal correlation factors include: the risk coupling coefficient between adjacent towers and the frequency of historical risks.

3. The method for risk assessment of safe operation of transmission lines in mining-affected areas according to claim 2, characterized in that: The expression for the objective function is: min||W-Ws||²+||W-Wo||²; The combined weight W is obtained by solving.

4. The method for risk assessment of safe operation of transmission lines in mining-affected areas according to claim 3, characterized in that: In step S1, the normalization function for the sufficiency coefficient adopts an S-shaped growth curve function, the expression of which is: N = 1 / (1+e^(-k·(q-q0))); Where q is the ratio of the mining width to the critical mining width.

5. The method for risk assessment of safe operation of transmission lines in mining-affected areas according to claim 4, characterized in that: The displacement acceleration in step S1 is the second-order difference of the rate of change of surface displacement. When the average acceleration over three consecutive days is positive and exceeds the threshold, a risk enhancement flag is triggered.

6. The method for risk assessment of safe operation of transmission lines in mining-affected areas according to claim 5, characterized in that: The improved entropy weight method used in step S2 is specifically as follows: Introducing the information entropy correction formula: e _i =-1 / ln(n)·Σ(p _ij ·ln(p_ ij +ε)) Where ε is a very small positive number.

7. The method for risk assessment of safe operation of transmission lines in mining-affected areas according to claim 6, characterized in that: The membership function constructed in step S4 is specifically a Gaussian membership function, with the following expression: μ(x) = exp(-(xc)² / (2σ²)); Where c is the level center value and σ is the level width parameter.

8. The system used to implement the risk assessment method for safe operation of transmission lines in mining-affected areas as described in any one of claims 1 to 7, characterized in that: include: The data acquisition and edge preprocessing module, deployed on the end side with GNSS, rain gauge, soil moisture sensor, and inclinometer, and on the edge side with edge computing gateway, is used to acquire dynamic factor data in real time and perform anomaly removal and CUSUM mutation detection. The factor quantification and database management module stores geographic information, geological and mining data, tower ledgers, real-time monitoring data and historical risk data, and supports hot-swappable configuration of new factors. The game theory combined weight calculation module integrates the analytic hierarchy process expert evaluation interface, the improved entropy weight method automatic calculation unit, and the game theory fusion calculation unit. The fuzzy cloud risk assessment engine performs fuzzy comprehensive evaluation and cloud model feature extraction, and outputs the risk level and confidence level. The dual evolutionary learning module includes a sliding window manager, periodic retraining triggers, and a change point detector; The 3D digital twin visualization and decision-making module is based on a 3D GIS engine to construct risk heat maps, render tower models, and overlay surface deformation fields, and supports risk dashboard display. The early warning release and feedback closed-loop module connects with the inspection terminal APP to push verification tasks and receive verification results, generating positive and negative sample feedback.

9. The risk assessment system for safe operation of transmission lines in mining-affected areas according to claim 8, characterized in that: The 3D digital twin visualization and decision-making module also includes a multi-scenario early warning push unit, with the following information push rules: When the confidence level is ≥0.7 and the risk level is high or very high, a confirmation alert will be sent. A warning will be sent when the confidence level is less than 0.7 and the risk score is greater than 0.

6. When CUSUM mutation detection is triggered, a trend mutation warning will be pushed out.

10. The risk assessment system for safe operation of transmission lines in mining-affected areas according to claim 8, characterized in that: The edge computing gateway has a built-in CUSUM algorithm parameter adaptive adjustment unit that automatically calculates the detection threshold based on the variance of historical data.

Images