Line surrounding environment monitoring and management method based on hidden danger discharge

Abstract

The invention discloses a line surrounding environment monitoring and management method based on hidden danger discharge, relates to the technical field of hidden danger discharge identification, and solves the problem that the staged development characteristics of tree barrier discharge cannot be identified and dynamic risk management and control cannot be implemented according to the staged development characteristics in the prior art. According to the invention, the fingerprint characteristics (peak value, frequency spectrum, fractal dimension and the like) of the discharge current are captured through the broadband sensor, and the three stages of corona, spark and flashover are accurately identified by using the AI model, so that the early warning period is prolonged; a dynamic risk index R (x, y) is constructed in combination with a strike distance theory, three elements of tree height, distance and discharge intensity are fused in real time, seasonal weights are automatically adjusted, and the risk misjudgment rate is reduced; finally, a closed-loop management process of'monitoring and early warning, work order distribution, pruning verification and efficiency iteration 'is formed, and the fundamental transformation from'blind person groping type' passive first-aid repair to'cloud and day shifting type 'accurate prevention is realized.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of hidden discharge identification, and particularly relates to a line surrounding environment monitoring and management method based on hidden discharge. BACKGROUND

[0002] With the rapid development of UHV AC / DC hybrid power grid and the continuous extension of power transmission corridors to complex ecological regions, tree barrier hidden dangers caused by vegetation growth under and around the lines have become the primary natural factor threatening the safe and stable operation of the power grid. According to statistics, short-circuit tripping accidents caused by tree barrier discharge account for more than 30% of unplanned outages of transmission lines, and show obvious time sequence evolution characteristics - from initial corona discharge to intermittent spark discharge, and finally developing into permanent flashover, the whole process spanning several weeks to several months. The traditional manual inspection mode is limited by terrain, weather and periodicity, and it is difficult to capture early discharge signals; while the existing online monitoring devices can only realize simple current overrun alarm, lack of fine recognition ability of discharge development stage, and cannot fuse electrical characteristics with multi-dimensional information such as spatial distribution of vegetation and growth characteristics of tree species, resulting in static lag of risk assessment and serious dependence on experience for operation and maintenance decision. Under this background, it has become an urgent need to build an intelligent monitoring system that can realize real-time sensing of discharge state and dynamic quantification of risk level and drive active operation and maintenance, which has become an urgent need for the power grid to change from "passive response" to "active prevention" in disaster prevention and reduction.

[0003] The core deficiency of the prior art is that it cannot identify the phased development characteristics of tree barrier discharge and implement dynamic risk control accordingly. The traditional monitoring system only sets a single current threshold, and when the alarm is triggered, it has often entered the spark discharge or even flashover precursor stage, with a warning time window of less than 24 hours, and the operation and maintenance personnel cannot respond in time; at the same time, its risk assessment is only based on static tree-line distance measurement, without considering tree species differences, seasonal growth rate and real-time discharge intensity, resulting in a false alarm rate of more than 40%, with a large number of low-risk trees being overpruned and high-risk hidden dangers being missed.

[0004] Therefore, there is a need for a line surrounding environment monitoring and management method based on hidden discharge. SUMMARY

[0005] To address the problem that existing technologies cannot identify the phased development characteristics of tree-induced discharges and implement dynamic risk management accordingly, this invention provides a method for monitoring and managing the surrounding environment of power lines based on potential discharge hazards. This method can capture the fingerprint characteristics (peak value, spectrum, fractal dimension, etc.) of the discharge current using broadband sensors, and accurately identify the three stages of corona, spark, and flashover using an AI model, extending the early warning period to 7-15 days. A dynamic risk index R(x,y) is constructed by combining strike distance theory, and the three factors of tree height, distance, and discharge intensity are integrated in real time with automatic seasonal weighting, reducing the risk misjudgment rate to below 15%. Ultimately, a closed-loop management process of "monitoring and early warning - work order dispatch - pruning verification - efficiency iteration" is formed, achieving a fundamental shift from passive "blind men and the elephant" repair to precise "clearing the clouds and seeing the sun" prevention. The specific technical solution is as follows: A method for monitoring and managing the environment around power lines based on potential discharge hazards includes the following steps: S1: Map the traverse lines and the three-dimensional coordinates of the trees, and create a file for each tree, including at least the tree species, tree height, and dielectric constant; S2: Continuously collect current waveforms. When the amplitude exceeds the threshold, automatically record the waveform and synchronize meteorological data. Extract discharge fingerprint features from the tower side, compress and upload the data. S3: Classify each discharge in three stages. If the confidence level is lower than the threshold, it will be manually reviewed. Otherwise, the frequency will be automatically counted and the corresponding level of warning will be triggered. S4: Divide the power transmission corridor into several grids and calculate the risk index for each grid. R ( x , y ), mark the risk level on the GIS heat map; S5: Yes R ( x , y Trees exceeding a set threshold are used to generate a cost-optimized pruning list. After the operation, dual verification using sensors and lidar ensures that the discharge is eliminated and that the tree line distance meets the set standards.

[0006] Preferably, the line data obtained from surveying the traverse includes at least the traverse height above the ground. Voltage level Mark the tower coordinates .

[0007] Preferably, step S2 is as follows: S201: Continuously acquires current waveforms, captures discharge pulses exceeding a threshold, and records time window data before and after the pulse. Simultaneously record meteorological parameters: humidity ,temperature Wind speed ; S202: Install edge computing unit on the tower side, and extract discharge feature vector in real time And upload after compressing the data.

[0008] Preferably, the feature vector is represented as follows: ; Wherein, is the basic electrical feature vector; is the spectral feature vector; is the box dimension; is the maximum singular index; is the envelope entropy.

[0009] Preferably, step S3 is as follows: Calculate for each discharge event, construct a feature matrix , and perform standardization ; Import the historical labeled discharge data, and train a three-classification model; After inputting the feature of a new discharge event into the model, if the recognition confidence is less than the set threshold, it is labeled as "unknown" and pushed to manual review, otherwise it is automatically archived and the discharge frequency is counted.

[0010] Preferably, step S4 is as follows: Calculate the risk index for each grid point, update the tree species parameters through annual laser radar scanning, and dynamically adjust the weight , wherein, represents the phase shift, represents the seasonal cycle; Construct a multi-level early warning triggering system; Overlay the risk cloud map on the GIS platform, label the high-risk cluster , and generate a report containing TOP-N risk point ranking.

[0011] Preferably, step S5 includes: S501: Overlay the risk cloud map on the GIS platform, label the high-risk cluster, and generate a report containing TOP-N risk point ranking; S502: Establish a cutting cost model for trees in the III-level area , solve the constraint optimization ; S503: High-frequency monitoring within 24 hours after pruning confirms that the discharge signal disappears, and laser radar re-measurement verifies , update the vegetation database and risk field model.

[0012] Preferably, it further comprises step S6: statistics of false negative rate, false positive rate and early warning time, retraining model parameters with half-year real data, updating tree species knowledge base and operation SOP, as follows: S601: statistics of false negative rate false positive rate and average early warning time if or <7 days, triggering system optimization; S602: re-fitting tree species coefficients K lightning attracting coefficient α and risk weight with past 6 months of real discharge data, dynamically adjusting early warning threshold through Bayesian optimization ; S603: expanding tree species parameter library K every half year, including new tree species, new voltage level cases into the standard, revising the tree barrier risk supervision SOP and re-training operation personnel.

[0013] A computer readable storage medium, comprising a stored program, wherein the program controls the device where the computer readable storage medium is located to execute the line surrounding environment monitoring and management method based on hidden discharge as described above when the program is running.

[0014] A processor for running a program, wherein the program executes the line surrounding environment monitoring and management method based on hidden discharge as described above when the program is running.

[0015] Compared with the prior art, the beneficial effects of the present application are: The present application fundamentally solves the core pain points such as the inability of traditional monitoring means to identify the discharge development stage, the static lag of risk assessment, and the short warning window, through systematic innovative design. First, the S2 and S3 steps realize real-time extraction and AI three-stage intelligent classification of discharge fingerprint features through wideband sensing and edge computing, upgrading the single current over-limit judgment to accurate identification of the evolution process of corona-spark-flashover, enabling operation and maintenance personnel to capture early hidden danger signals rather than responding to the end-stage flashover. Second, the S4 step builds a dynamic risk index R(x, y), innovatively integrating tree height, tree line distance, real-time discharge intensity, and meteorological elements into a grid space model, automatically adjusting with seasonal weights, completely changing the one-sidedness of traditional static distance assessment, and realizing the spatio-temporal dynamic evolution and visual accurate positioning of risk levels. Finally, the S5 step forms a closed-loop management system of "warning-decision-execution-verification", generates a pruning list through a cost-optimal algorithm, and introduces a dual acceptance mechanism of sensors and laser radars, ensuring that the operation and maintenance effect is quantifiable and verifiable, avoiding the coexistence of excessive pruning and hidden danger omission, and promoting the tree barrier prevention and control from an experience-driven to a data-driven intelligent lean operation and maintenance mode. BRIEF DESCRIPTION OF DRAWINGS

[0016] In order to more clearly illustrate the technical solutions in the specific embodiments or the prior art, the following will briefly introduce the drawings needed to be used in the specific embodiments or the prior art description. In all the drawings, similar elements or parts are generally identified by similar reference numerals. In the drawings, each element or part is not necessarily drawn according to the actual proportion.

[0017] Figure 1 The method flowchart of the present application. DETAILED DESCRIPTION

[0018] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0019] It should be understood that, when used in the present specification and the appended claims, the terms "comprise" and "include" indicate the presence of described features, integers, steps, operations, elements, and / or components, but do not exclude one or more other features, integers, steps, operations, elements, components, and / or sets thereof.

[0020] It should also be understood that the terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to limit the invention. As used in this specification and the appended claims, the singular forms “a,” “an,” and “the” are intended to include the plural forms unless the context clearly indicates otherwise.

[0021] It should also be further understood that the term "and / or" as used in this specification and the appended claims refers to any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0022] In one embodiment of the present invention, a method for monitoring and managing the surrounding environment of a power line based on potential discharge hazards is provided, such as... Figure 1 As shown, it includes the following steps: Step 1: Basic data collection of power lines and vegetation; power line data should include at least the height of the conductor above the ground. Voltage level Mark the tower coordinates Basic vegetation data should include at least the initial tree height. and location Dielectric correction factor based on tree species K (Broadleaf trees 0.8-1.2, coniferous trees 1.0-1.5), Lightning attraction coefficient .

[0023] In addition, some of the data involved in the above process that can be collected by sensors is collected by setting up various sensors, and during the collection process, the number of sensor deployment points is calculated based on the line length L: ( (Effective monitoring distance for a single sensor).

[0024] Step 2: Continuously collect current waveforms. When the amplitude exceeds the threshold, automatically record the waveform and synchronize meteorological data. Extract discharge fingerprint features from the tower side, compress and upload the data.

[0025] S201: Continuously acquires current waveforms, captures discharge pulses exceeding a threshold, and records time window data Δ before and after the pulse. tpre ), and simultaneously record meteorological parameters: humidity ,temperature Wind speed ; For example, the trigger threshold is set to It automatically records the complete pulse waveform exceeding the threshold and the data before and after 10ms, and simultaneously collects meteorological parameters such as humidity, temperature, and wind speed.

[0026] S202: Install an edge computing unit on the tower side to extract discharge feature vectors in real time. And compress the data, only the characteristic value is uploaded to the master station through the MQTT protocol.

[0027] Feature vector ; Wherein, is the basic electrical feature vector; is the spectral feature vector; is the box dimension; is the maximum singular index; is the envelope entropy; Basic electrical feature vector , wherein, is the discharge pulse peak current, is the pulse rise time, is the pulse fall time, is the discharge charge amount, is the time-varying discharge current; Spectral feature vector , wherein, is the low-frequency band energy, is the total signal energy; is the medium-frequency band energy, is the high-frequency band energy, is the total signal energy.

[0028] Envelope entropy , wherein, is the envelope segment number; is the i-th segment probability (the proportion of the total energy of the amplitude value in the segment), is the i-th segment amplitude (the average amplitude of the envelope in the i-th segment).

[0029] Step three: use the AI model to classify each discharge into three stages (corona / spark / flashover), if the confidence is less than 85%, then manual review, otherwise automatically count the frequency and trigger the corresponding level warning.

[0030] S301: Automatically calculate the peak current , rise time , charge amount Q , box dimension and other parameters of each discharge pulse, and construct a standardized feature matrix ; For example, calculate for each discharge event, construct a feature matrix , and perform standardization 。

[0031] S302: Import the historical marked discharge data, train a three-classification model (corona / spark / flashover), and focus on optimizing the flashover false alarm cost wflashover Make the model more sensitive to high-risk events; Example, train a three-classifier based on historical labeled data Algorithm selection: Random Forest / Support Vector Machine, optimization goal: minimize misclassification cost 。

[0032] S303: After the new discharge event feature is input into the model, if the recognition confidence (e.g. 0.85) is marked as "unknown" and pushed to manual review, otherwise automatically archive and count the discharge frequency.

[0033] Step four: divide the corridor into 10m grid, calculate the risk index of each grid Mark the risk level on the GIS heat map, automatically push the top-10 high-risk tree list and disposal suggestions to the team.

[0034] S401: Divide the transmission corridor (the part under the overhead line) into 10m x 10m grid, calculate the risk index of each grid point Increase tree height weight in spring and summer ω 2, increase distance weight in autumn and winter ω 1; Example, calculate the risk index of each grid point Tree species parameters are updated through annual laser radar scanning, and the weight is dynamically adjusted , where represents the phase shift, which is used to adjust the seasonal change phase of the weight, represents the seasonal cycle (month).

[0035] S402: Build a multi-level early warning triggering system, at least including: Level III warning: when or first detected corona discharge, output SMS notification to operation and maintenance team; Level II warning: when or continuous narc spark discharge, output system work order + unmanned aerial vehicle fixed-point inspection; Level I warning: when or detect the precursor of flashover, output line protection lock preparation + emergency team on standby.

[0036] Example, when R ( x , y )>0.3, send a message to the team, R ( x , y )>0.6 or continuous 3 times spark discharge, automatically generate work order and dispatch unmanned aerial vehicle verification,R ( x , y When the value is greater than 0.8, the emergency logging plan is triggered and line protection is pre-set. S403: Overlay risk cloud maps on the GIS platform and mark high-risk clusters. Generate a report containing a ranking of the top-N risk points.

[0037] For example, a GIS risk heat map is automatically generated daily, and a list of the top 10 high-risk trees is automatically marked and pushed to the mobile APP of the operation and maintenance manager.

[0038] Step 5: [Regarding...] R ( x , y The optimal pruning list for tree generation cost > 0.6 is used, and the discharge is eliminated and the tree line distance is met through dual verification by sensors and lidar within 24 hours after the operation. , This is the flashover critical distance.

[0039] S501: Overlay risk cloud maps on the GIS platform and mark high-risk clusters. Generate a report containing a ranking of the top-N risk points; S502: Establish a logging cost model for trees in Level III areas Solving constrained optimization ; S503: High-frequency monitoring confirmed the disappearance of the discharge signal within 24 hours after pruning, and lidar retesting verified the result. Update the vegetation database and risk field model; Step Six: Review the false negative rate, false positive rate, and early warning time; retrain the model parameters using six months of real data; update the tree species knowledge base and operation and maintenance SOPs; and continuously improve the accuracy of supervision.

[0040] S601: Statistical underreporting rate False alarm rate and average warning lead time ,like or If the system is not optimized within 7 days; S602: Refitting tree species coefficients using real discharge data from the past 6 months K Lightning ignition coefficient α and risk weight Dynamically adjust the early warning threshold through Bayesian optimization. ; S603: Expand the tree species parameter library K every six months, incorporate new tree species and new voltage level cases into the standard, revise the "Tree Obstacle Risk Supervision SOP" and retrain maintenance personnel.

[0041] In summary, the present application fundamentally solves the core pain points such as the inability of traditional monitoring methods to identify the discharge development stage, the static lag of risk assessment, and the short warning window, through systematic innovative design. First, steps S2 and S3 realize real-time extraction and AI three-stage intelligent classification of discharge fingerprint features through wideband sensing and edge computing, upgrading the single current over-limit judgment to accurate recognition of the evolution process of corona-spark-flashover, enabling operation and maintenance personnel to capture early hidden danger signals rather than responding to the end-stage flashover. Second, step S4 builds a dynamic risk index R(x, y), innovatively integrating tree height, tree line distance, real-time discharge intensity, and meteorological elements into a grid-based spatial model, automatically adjusting with seasonal weights, completely changing the one-sidedness of traditional static distance assessment, and realizing the dynamic evolution and visual accurate positioning of risk levels in space and time. Finally, step S5 forms a closed-loop management system of "warning-decision-execution-verification", generates a pruning list through a cost-optimal algorithm, and introduces a dual-acceptance mechanism of sensors and laser radars to ensure that the operation and maintenance effect is quantifiable and verifiable, avoiding the coexistence of excessive pruning and hidden danger omission in extensive management, and promoting the intelligent lean operation and maintenance mode of tree barrier prevention and control from experience-driven to data-driven.

[0042] Those skilled in the art can appreciate that the units of the examples described in combination with the embodiments disclosed herein can be realized in electronic hardware, computer software or a combination of both. In order to clearly illustrate the interchangeability of hardware and software, the components of the examples have been described in general terms in the above description. Whether the functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. A person skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present application.

[0043] In the embodiments provided by the present application, it should be understood that the division of units is only a logical functional division, and actual implementation can have another division manner, for example, multiple units can be combined into one unit, one unit can be split into multiple units, or some features can be ignored, etc.

[0044] In addition, each functional unit in each embodiment of the present application can be integrated in one processing unit, or each unit can exist physically, or two or more units can be integrated in one unit. The integrated unit can be realized in the form of hardware or software functional unit.

[0045] The integrated unit, if implemented in the form of a software function unit and sold or used as an independent product, can be stored in a computer readable storage medium. Based on such understanding, the technical solutions of the present application essentially or say the part that contributes to the prior art or the whole or part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium, including a plurality of instructions to make a computer device (which can be a personal computer, a server or a network device, etc.) execute all or part of the steps of the method described in various embodiments of the present application. The foregoing storage medium includes: a U disk, a read-only memory (ROM, Read-0nly Memory), a random access memory (RAM, Random Access Memory), a mobile hard disk, a magnetic disk or an optical disk, and various media that can store program codes.

[0046] Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present application, and not to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: it can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement for part or all of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present application, and they should be covered in the scope of the claims and the specification of the present application.

Claims

1. A method for monitoring and managing the surrounding environment of power lines based on potential discharge hazards, characterized in that, Includes the following steps: S1: Map the traverse lines and the three-dimensional coordinates of the trees, and create a file for each tree, including at least the tree species, tree height, and dielectric constant; S2: Continuously collect current waveforms. When the amplitude exceeds the threshold, automatically record the waveform and synchronize meteorological data. Extract discharge fingerprint features from the tower side, compress and upload the data. S3: Classify each discharge in three stages. If the confidence level is lower than the threshold, it will be manually reviewed. Otherwise, the frequency will be automatically counted and the corresponding level of warning will be triggered. S4: Divide the power transmission corridor into several grids and calculate the risk index for each grid. R ( x , y Mark the risk level on the GIS heat map; S5: Yes R ( x , y Trees exceeding a set threshold are used to generate a cost-optimized pruning list. After the operation, dual verification using sensors and lidar ensures that the discharge is eliminated and that the tree line distance meets the set standards.

2. The method for monitoring and managing the surrounding environment of a power line based on hidden danger discharge as described in claim 1, characterized in that, The data obtained from surveying traverses should include at least the traverse height above the ground. Voltage level Mark the tower coordinates .

3. The method for monitoring and managing the surrounding environment of a power line based on hidden danger discharge as described in claim 1, characterized in that, Step S2 is as follows: S201: Continuously acquires current waveforms, captures discharge pulses exceeding a threshold, and records time window data before and after the pulse. Simultaneously record meteorological parameters: humidity ,temperature Wind speed ; S202: Install an edge computing unit on the tower side to extract discharge feature vectors in real time. Then compress the data before uploading.

4. The method for monitoring and managing the surrounding environment of a power line based on hidden danger discharge as described in claim 3, characterized in that, The feature vector is represented as follows: ； in, The basic electrical feature vector; It is a spectral feature vector; Let the box dimension be 1. The maximum singularity index; Let be the envelope entropy.

5. A method for monitoring and managing the surrounding environment of a power line based on hazard discharge, as described in claim 4, is characterized in that... Step S3 is as follows: Calculate for each discharge event Construct the feature matrix Implement standardization ; Import historically labeled discharge data and train a three-class classification model; After the new discharge event features are input into the model, if the recognition confidence level is less than the set threshold, it is marked as "unknown" and pushed for manual review; otherwise, it is automatically archived and the discharge frequency is counted.

6. The method for monitoring and managing the surrounding environment of a power line based on hidden danger discharge as described in claim 1, characterized in that, Step S4 is as follows: Calculate the risk index for each grid point. Tree species parameters are updated annually via LiDAR scanning, with weights dynamically adjusted. ,in, This indicates the phase offset. Indicates seasonal cycles; Construct a multi-level early warning triggering system; Overlay risk cloud maps on a GIS platform and mark high-risk clusters. Generate a report containing a ranking of the top-N risk points.

7. A method for monitoring and managing the surrounding environment of a power line based on hazard discharge, as described in claim 6, is characterized in that... Step S5 includes: S501: Overlay a risk cloud map on the GIS platform, mark high-risk clusters, and generate a report with a ranking of TOP-N risk points; S502: Establish a logging cost model for trees in Level III areas Solving constrained optimization ; S503: High-frequency monitoring confirmed the disappearance of the discharge signal within 24 hours after pruning, and lidar retesting verified the result. Update the vegetation database and risk field model.

8. A method for monitoring and managing the surrounding environment of a power line based on hazard discharge, as described in claim 6, is characterized in that... It also includes step S6: calculating the false negative rate, false positive rate, and early warning time; retraining the model parameters using six months of real data; and updating the tree species knowledge base and operation and maintenance SOP, as detailed below: S601: Statistical underreporting rate False alarm rate and average warning lead time ,like or If the system is not optimized within 7 days; S602: Refitting tree species coefficients using real discharge data from the past 6 months K Lightning ignition coefficient α and risk weight The early warning threshold is dynamically adjusted through Bayesian optimization. ; S603: Expand the tree species parameter library K every six months, incorporate new tree species and new voltage level cases into the standard, revise the "Tree Obstacle Risk Supervision SOP" and retrain maintenance personnel.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes a stored program, wherein, when the program is executed, it controls the device containing the computer-readable storage medium to perform the line perimeter environment monitoring and management method based on hidden danger discharge as described in any one of claims 1 to 8.

10. A processor, characterized in that, The processor is used to run a program, wherein the program executes the method for monitoring and managing the surrounding environment of a line based on hazard discharge as described in any one of claims 1 to 8.

Images