A method and system for locating a lightning stroke fault of a power transmission line

Abstract

This invention provides a method and system for locating and analyzing lightning strike faults in power transmission lines. The method includes establishing a multi-monitoring station BeiDou time synchronization system to unify the time reference of each monitoring station; collecting lightning pulse signals, capturing lightning events according to a composite triggering strategy, and retaining data using a local storage strategy; transmitting the lightning signals to a backend service platform via an encrypted communication link, where the platform verifies the legality and scores the quality of the signals, eliminates abnormal signals, and extracts characteristic parameters; this application enables automatic retrieval of the plane projection of the lightning strike point and the topology parameters of the transmission line within the adjacent area, accurately distinguishing between three typical lightning strike fault types: backflash, backflash, and direct strike; classifying the lightning strike impact level; and generating a standardized lightning strike fault analysis report, providing complete and accurate quantitative data for transmission line operation and maintenance, and fault review and judgment.

Description

Technical Field

[0001] This invention relates to the field of lightning protection technology for power transmission lines, and in particular to a method and system for locating and analyzing lightning strike faults in power transmission lines. Background Technology

[0002] As a crucial component of the power system, the safe and stable operation of transmission lines directly impacts the reliability of power supply. Transmission lines in high-altitude and complex terrain areas are significantly more susceptible to lightning strikes due to frequent thunderstorms, high soil resistivity, and poor grounding conditions. Existing lightning strike location methods suffer from low time synchronization accuracy, failure to consider terrain effects in location calculations, and insufficient integration with line fault analysis. This results in large errors in lightning strike location and inaccurate fault type determination, failing to provide precise decision-making basis for lightning protection and maintenance of transmission lines.

[0003] Specifically, most mainstream lightning strike location methods are based on the TDOA (Time Difference of Arrival) algorithm. However, conventional methods have low time synchronization accuracy among multiple monitoring stations, making it difficult to meet the microsecond or even nanosecond level time difference calculation requirements. Furthermore, the calculation process does not consider the impact of complex terrain on lightning signal propagation, resulting in a large deviation between the location result and the actual landing point. In addition, it only achieves a simple calculation of the lightning strike point and does not combine the geographical information and signal characteristics of the transmission line to complete the analysis of fault type and impact level.

[0004] To address the aforementioned issues, a method and system for locating and analyzing lightning strike faults in power transmission lines are now designed. Summary of the Invention

[0005] This application provides a method and system for locating and analyzing lightning strike faults in transmission lines, which solves the problems of low time synchronization accuracy, failure to consider terrain influence in location calculation, and insufficient integration with line fault analysis in existing lightning strike location methods. These problems result in large errors in lightning strike location and inaccurate fault type determination, making it impossible to provide accurate decision-making basis for lightning protection and maintenance of transmission lines.

[0006] To achieve the above objectives, the technical solution adopted by this invention is: a method for locating and analyzing lightning strike faults in transmission lines, comprising the following steps: S1: Establish a multi-monitoring station BeiDou time synchronization system to unify the time reference of each monitoring station; S2: Collects lightning pulse signals, captures lightning events according to a composite triggering strategy, and completes data retention in conjunction with a local storage strategy; S3: The lightning strike signal is sent to the back-end service platform via an encrypted communication link. The back-end service platform verifies the legality and scores the quality of the lightning strike signal, eliminates abnormal signals, and extracts feature parameters. S4: Based on the TDOA time difference of arrival algorithm, a set of equations is constructed by combining the coordinates of the monitoring station, the signal detection time and the signal quality weight, and the weighted least squares method is used to solve for the initial position of the lightning strike. S5: Correct the initial position of the lightning strike according to the engineering correction strategy, and output the accurate lightning strike point with error ellipse and confidence level; S6: Project the precise lightning strike point onto the geographic information system of the transmission line corridor, and combine spatial correlation parameters with lightning signal characteristics to determine the type of lightning fault and classify the level of lightning impact; S7: Based on the lightning strike location results, fault analysis results, and lightning signal characteristics, trigger the graded early warning rules and retain all lightning strike data.

[0007] Preferably, the step of building a multi-monitoring station BeiDou time synchronization system in step S1 includes deploying GNSS disciplined clocks, PPS second pulse output modules and acquisition boards at the transmission line monitoring stations, aligning the acquisition boards with the BeiDou time signal through clock phase-locking, calibrating the cable delay of each station to unify the time reference, monitoring the timekeeping stability of a single station and the relative error of multiple stations in real time, triggering maintenance alarms when abnormalities occur, and achieving 200ns-level time reference unification for monitoring stations; In step S1, a unified time source is provided by the BeiDou GNSS disciplined clock. The clock synchronization of the acquisition board is achieved by 1PPS pulse phase-locking. Combined with full-link delay calibration compensation, real-time calibration of relative time difference between multiple stations, monitoring of loss of lock and over-threshold anomalies and timekeeping guarantee, the sampling time of each station is unified to the 200ns level accuracy, and a high-precision timestamp is bound to the sampling data, thereby ensuring the time consistency of lightning strike signal acquisition.

[0008] Preferably, in step S2, the lightning pulse signal is acquired using a multi-channel synchronous ADC in parallel with a sampling rate of 1MS / s and 16bit, ensuring no interference between channels. The front end is equipped with a wideband lightning pulse conditioning circuit with attenuation, impedance matching, and anti-aliasing filtering functions to adapt to wide-amplitude pulse signal input from direct lightning strikes and induced lightning strikes, avoiding signal clipping and saturation distortion. The clock source uses a Beidou disciplined clock and a PPS second pulse, calibrated to a high-precision time reference of 200ns through clock phase-locked loop to ensure strict alignment of sampling times across multiple stations and channels. In step S2, the composite triggering strategy is a multi-index weighted triggering based on the level threshold, short-time energy, rising edge slope, bandwidth energy ratio, and impulse count. The weighting ratios for each index are as follows: level threshold 25%, short-time energy 30%, rising edge slope 20%, bandwidth energy ratio 15%, and impulse count 10%. After normalizing each index, the weighted sum is calculated to obtain a comprehensive triggering score. If the comprehensive score is higher than the set threshold, it is determined to be a valid lightning strike trigger.

[0009] Preferably, in step S3, the lightning strike signal being sent to the back-end service platform via an encrypted communication link means that the locally stored lightning strike event data is encrypted and sent up using an industrial-grade 4G / 5G communication module. The communication link uses an APN / VPN private network + TLS two-way authentication to ensure the reliability and security of data transmission. In step S3, the back-end service platform performs a quality score on the lightning strike signal based on four indicators: timestamp validity, channel integrity, signal amplitude stability, and noise ratio. The quality score results serve as the basis for assigning signal quality weights in the subsequent lightning strike location calculation. Abnormal signals with invalid timestamps, missing channels, signal distortion, excessive noise ratio, or substandard quality scores are entered into a dead-signal queue for manual review by the back-end service platform. Valid signals are marked with characteristic parameters and then proceed to the subsequent location calculation stage.

[0010] Preferably, in step S4, based on the TDOA (Time Difference of Arrival) algorithm, a set of equations is constructed by combining the monitoring station coordinates, signal detection time, and signal quality weights. The weighted least squares method is then used to solve the equations to obtain the initial location of the lightning strike. The calculation steps are as follows: Based on the TDOA algorithm, a set of nonlinear constraint equations for the time difference of arrival of lightning signals from multiple stations and the spatial distance are established. The coordinates of each monitoring station and the arrival time of the signal feature points are used as inputs. The set of equations is linearized by first-order Taylor expansion at the initial estimation point. A diagonal weight matrix W is constructed based on the signal quality score results, with higher weights assigned to high-quality signals and lower weights assigned to abnormal signals. Subsequently, a weighted sum of squared residuals is constructed as the objective function, and the weighted least squares solution formula is obtained by taking the extreme value through differentiation. The estimated value of the lightning strike point is continuously corrected through iterative calculation until the residuals converge. Finally, the initial coordinates of the lightning strike and the calculated residuals are output to complete the location calculation.

[0011] Preferably, step S5, which corrects the initial lightning strike location according to an engineering correction strategy and outputs a precise lightning strike point with an error ellipse and confidence level, includes the following steps: Step S5.1: Based on the terrain elevation difference between the monitoring station and the initial point of lightning strike obtained from the geographic information system, introduce an empirical terrain correction coefficient. and path terrain roughness The equivalent propagation speed of electromagnetic waves is corrected, and the corrected model is as follows: ; in The speed of light; For high-altitude mountainous areas, a height compensation item will be further introduced. The actual propagation path length of electromagnetic waves is compensated to eliminate positioning drift caused by altitude differences. The calculation formula is as follows: ; In the formula, To monitor the elevation difference between the monitoring station and the point of lightning strike, The straight-line distance between two points in a plane. This is the horizontal projection distance; Step S5.2 Establish a reconstruction residual verification mechanism and calculate the solution residuals for each monitoring station. The calculation formula is as follows: ; Among them No. The theoretical distance difference between the station and the reference station, , , These are the initial positioning coordinates. This is an estimated distance between the reference station and the point of lightning strike. Set threshold ,in Let the standard deviation of the residuals be denoted as , if the standard deviation of a certain station is . If the threshold is exceeded, the station is determined to be significantly affected by multipath effects or interference, and it is removed from the station array and WLS calculation is performed again until the residual converges. Step S5.3 Performs spatiotemporal clustering analysis on multiple return stroke signals collected during the same thunderstorm process, using an improved territory clustering algorithm. The clustering criteria are as follows: ; ; Pulses that meet the above time and space conditions are grouped into the same lightning cluster, and the energy weighting center is calculated as the final output point, effectively avoiding the aliasing of positioning results caused by multiple return strikes.

[0012] Preferably, step S6, determining the type of lightning strike fault and classifying the impact level of the lightning strike, includes the following steps: If the lightning strike occurs outside the shielded area of ​​the ground wire protection angle and hits the conductor area, it is considered a backflash. If the lightning strike hits the tower body or ground wire and causes line insulation flashover, it is considered a backflash. If the lightning strike directly hits the exposed area of ​​the conductor without ground wire protection, it is considered a direct strike. After determining the fault type, the impact of the lightning strike on the line is classified into four levels: no impact, low impact, medium impact, and high impact, based on the horizontal distance between the lightning strike point and the tower, the signal amplitude, the lightning energy intensity, and the fault type. Among them, the no-impact level corresponds to lightning strikes far from the line corridor and weak signal energy, which do not cause any impact on the power transmission equipment; the low-impact level corresponds to lightning strikes near the line corridor and small signal amplitude, with no risk of equipment failure; the medium-impact level corresponds to lightning strikes close to the towers and conductors, with a certain amount of lightning energy, and there is a potential risk of line flashover; the high-impact level corresponds to direct strikes, bypass strikes, and backflash strikes, which are effective fault lightning strikes with high signal energy and strike points close to the line equipment, which can easily cause line fault tripping and equipment damage.

[0013] Preferably, the graded early warning rules in step S7 include a four-level quantitative graded early warning mechanism of blue, yellow, orange, and red, with a unified and fixed judgment threshold set: A blue alert corresponds to a long-distance, weak lightning event where the vertical distance from the lightning strike point to the line is greater than 200m, there are fewer than 3 effective calculation stations, and the signal amplitude is 1.5 to 2.5 times the background noise. A yellow alert requires at least 3 valid solution sites, a location confidence level of at least 0.75, a lightning strike distance between the tower and the line corridor between 80m and 200m, a signal amplitude of 2.5 to 3.5 times the background noise, and a rising edge slope of less than 0.8mV / μs. The criteria for issuing an orange alert are: the distance from the lightning strike to the tower is no more than 80m, the distance from the outer boundary of the line corridor is no more than 50m, the rise slope is no less than 0.8mV / μs, and the solution residual is less than 30m. A red alert can be triggered if any two of the following conditions are met: a lightning strike within 30m of the tower in a high-risk backflash zone; a lightning strike within 50m of the conductor corridor outside the protection angle in a high-risk backflash zone; a signal amplitude of not less than 3.5 times the background noise; a lightning strike energy of not less than 8.5 × 10³ quantization units; at least two multi-pulse return strokes within 500ms; and a deterministic fault lightning strike event.

[0014] A power transmission line lightning strike fault location and analysis system, comprising: The front-end monitoring device is deployed along the power transmission line to collect lightning pulse signals, complete BeiDou high-precision time synchronization, buffer and store signals locally, and transmit lightning signals to the outside through the communication transmission network. The communication transmission network is used to enable encrypted data transmission and remote configuration command issuance between the front-end monitoring device and the back-end service platform; The back-end service platform is used to receive lightning strike signals uploaded by the front-end monitoring devices and perform decryption verification, signal quality scoring, and preprocessing.

[0015] Preferably, the backend service platform is a server cluster equipped with positioning and calculation software, a geographic information system, an early warning push system, and a database, used to receive lightning strike signals uploaded by the front-end monitoring device and perform decryption verification, signal quality scoring, and preprocessing.

[0016] This invention provides a method and system for locating and analyzing lightning strike faults in power transmission lines, with the following advantages: By relying on a geographic information system to achieve unified matching and precise spatial registration of lightning strike coordinates, the system can automatically retrieve the planar projection of the lightning strike point and the topological parameters of the transmission lines within the surrounding area. Combining multi-dimensional characteristic parameters such as the shielding range of the ground wire protection angle, tower spatial distance parameters, peak amplitude of the lightning signal, short-time pulse energy, and waveform rising edge slope, it can accurately distinguish three typical lightning strike fault types: backflash, backflash, and direct strike. Furthermore, by combining spatial location relationships and lightning electrical characteristics, it can quantitatively classify lightning strike impact levels into four levels: no impact, low impact, medium impact, and high impact. At the same time, it can generate standardized lightning strike fault analysis reports, providing complete and accurate quantitative data for transmission line operation and maintenance, fault review and judgment.

[0017] By constructing a BeiDou 200ns-level multi-station timing synchronization system, a precise time reference is provided for TDOA time difference calculation. Multiple engineering correction methods, such as terrain elevation difference compensation, hierarchical propagation speed correction, abnormal station elimination, and multi-pulse spatiotemporal clustering, are introduced to overcome the influence of high-altitude complex terrain on lightning signal propagation, significantly reduce the lightning strike point positioning error, and output quantitative positioning results with error ellipse and confidence level, making the positioning results more intuitive and accurate.

[0018] By establishing a four-level quantitative grading early warning mechanism (blue, yellow, orange, and red), each level of early warning is equipped with clearly defined spatial distance, signal amplitude, and fixed quantitative judgment thresholds for location convergence. Simultaneously, combined with multi-pulse time clustering, spatial event merging within the same line segment, early warning cooling cycle configuration, and multiple false alarm suppression strategies including interference signal filtering, invalid false alarms and short-term repetitive alarms can be effectively suppressed. Furthermore, by pushing early warning information through multiple channels, the timeliness, stability, and on-site maintenance efficiency of lightning strike fault early warnings are significantly improved.

[0019] High-speed acquisition via a 1MS / s, 16-bit multi-channel synchronous ADC, coupled with a dedicated lightning pulse broadband conditioning circuit, enables simultaneous acquisition of wide-amplitude pulse signals from direct and induced lightning strikes, avoiding signal clipping and saturation distortion. It features on-chip ring buffer for continuous, drop-free acquisition, SSD rolling over storage, and a critical lightning event locking protection mechanism. It supports local caching during network outages and automatic resume transmission after network recovery. It also supports remote configuration of trigger parameters and pre / post-trigger buffer lengths via a backend service platform, making it suitable for long-term deployment and operation in complex field conditions. The overall equipment exhibits strong operational stability and compatibility with field conditions.

[0020] By establishing a full lifecycle management system for lightning strike events, the system enables unified storage and multi-dimensional retrieval of collected data, location results, fault reports, and early warning records. Through RESTful API interfaces, hierarchical data sharing is achieved. Internally, access is authorized according to scheduling, operation and maintenance, and administrator roles. Externally, a least privilege read-only access mode is adopted. All data access and operation behaviors are logged throughout the process, supporting accurate post-event tracing of lightning strike faults and statistical analysis of regional lightning storm patterns. This provides long-term data support for the optimization and upgrading of the overall lightning protection system for transmission lines. Attached Figure Description

[0021] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0022] Figure 1 A flowchart provided for an embodiment of this application. Detailed Implementation

[0023] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0024] This application provides a method and system for locating and analyzing lightning strike faults in transmission lines. It can solve the problems of low time synchronization accuracy, failure to consider terrain influence in location calculation, and insufficient integration with line fault analysis in existing lightning strike location methods. These problems lead to large errors in lightning strike location and inaccurate fault type determination, making it impossible to provide accurate decision-making basis for lightning protection and maintenance of transmission lines.

[0025] like Figure 1 As shown, a method for locating and analyzing lightning strike faults in transmission lines includes the following steps: S1: Establish a multi-monitoring station BeiDou time synchronization system to unify the time reference of each monitoring station; S2: Collects lightning pulse signals, captures lightning events according to a composite triggering strategy, and completes data retention in conjunction with a local storage strategy; S3: The lightning strike signal is sent to the back-end service platform via an encrypted communication link. The back-end service platform verifies the legality and scores the quality of the lightning strike signal, eliminates abnormal signals, and extracts feature parameters. S4: Based on the TDOA time difference of arrival algorithm, a set of equations is constructed by combining the coordinates of the monitoring station, the signal detection time and the signal quality weight, and the weighted least squares method is used to solve for the initial position of the lightning strike. S5: Correct the initial position of the lightning strike according to the engineering correction strategy, and output the accurate lightning strike point with error ellipse and confidence level; S6: Project the precise lightning strike point onto the geographic information system of the transmission line corridor, and combine spatial correlation parameters with lightning signal characteristics to determine the type of lightning fault and classify the level of lightning impact; S7: Based on the lightning strike location results, fault analysis results, and lightning signal characteristics, trigger the graded early warning rules and retain all lightning strike data.

[0026] In one embodiment, the step of building a multi-monitoring station BeiDou time synchronization system in step S1 includes deploying GNSS disciplined clocks, PPS second pulse output modules and acquisition boards at the transmission line monitoring stations, aligning the acquisition boards with the BeiDou time signal through clock phase-locking, calibrating the cable delay of each station to unify the time reference, monitoring the timekeeping stability of a single station and the relative error of multiple stations in real time, triggering maintenance alarms when abnormalities occur, and achieving 200ns-level time reference unification for monitoring stations. In step S1, a unified time source is provided by the BeiDou GNSS disciplined clock. The clock synchronization of the acquisition board is achieved by 1PPS pulse phase-locking. Combined with full-link delay calibration compensation, real-time calibration of relative time difference between multiple stations, monitoring of loss of lock and over-threshold anomalies and timekeeping guarantee, the sampling time of each station is unified to the 200ns level accuracy, and a high-precision timestamp is bound to the sampling data, thereby ensuring the time consistency of lightning strike signal acquisition.

[0027] It should be noted that when abnormal situations such as loss of lock or time deviation exceeding the threshold occur, an operation and maintenance alarm will be triggered immediately to ensure the time consistency of lightning strike signal collection.

[0028] The conditions for determining loss of lock are: continuous loss of satellite signal for ≥10 seconds, loss of PPS second pulse for ≥1 cycle, GNSS discipline clock changing from locked to unlocked, or single-station timekeeping drift within 1 second >100ns. Meeting any one of these conditions constitutes a loss of lock. The conditions for determining time deviation exceeding the threshold are: single-station timekeeping deviation >200ns or relative time difference between any two monitoring stations >200ns.

[0029] In this implementation plan, the conditions for loss of lock and time deviation exceeding the threshold are jointly determined by the accuracy requirements of the TDOA positioning algorithm, the hardware performance of the BeiDou disciplined clock, the power industry monitoring standards, and the positioning error constraints of lightning signal propagation, ensuring that the time synchronization of multiple stations meets the 200ns level requirement.

[0030] In one embodiment, the lightning pulse signal acquisition in step S2 is performed by multi-channel synchronous ADC parallel acquisition at a sampling rate of 1MS / s and 16bit, with no interference between channels. The front end is configured with a wideband lightning pulse conditioning circuit with attenuation, impedance matching, and anti-aliasing filtering functions to adapt to wide-amplitude pulse signal input from direct lightning strikes and induced lightning strikes, avoiding signal clipping and saturation distortion. The clock source uses a Beidou disciplined clock and a PPS second pulse, and is calibrated to a high-precision time reference of 200ns through clock phase-locked loop to ensure strict alignment of sampling times across multiple stations and channels. In step S2, the composite triggering strategy is a multi-index weighted triggering based on the level threshold, short-time energy, rising edge slope, bandwidth energy ratio, and impulse count. The weighting ratios for each index are as follows: level threshold 25%, short-time energy 30%, rising edge slope 20%, bandwidth energy ratio 15%, and impulse count 10%. After normalizing each index, the weighted sum is calculated to obtain a comprehensive triggering score. If the comprehensive score is higher than the set threshold, it is determined to be a valid lightning strike trigger. This implementation scheme's composite triggering strategy supports adaptive noise thresholds and multiple coordinated triggering markers, dynamically adjusting triggering parameters according to the ambient noise level, effectively avoiding missed and false alarms. The acquired signals are stored in a ring buffer, and the pre-trigger and post-trigger lengths are remotely configured.

[0031] Specifically, the adaptive noise threshold is dynamically calculated by statistically analyzing the mean and variance of the background noise in real time. The threshold value is the sum of the noise mean and the adaptive coefficient multiplied by the noise standard deviation. The threshold parameter is updated in real time according to the on-site noise environment. The multi-station collaborative triggering and marking method involves broadcasting a trigger time window after a single-station trigger, with surrounding monitoring stations completing waveform and time consistency verification within the time window; if the multi-station collaborative conditions are met, it is marked as a reliable lightning strike event, while a single-station trigger is marked as suspected interference. The pre-trigger and post-trigger lengths are configured by the backend service platform through a communication link. The collection unit takes effect in real time and sends back the configuration results. The pre-trigger length ranges from 100 to 2000 points, and the post-trigger length ranges from 500 to 5000 points. The multi-station collaborative triggering criteria are: the triggering time difference between multiple stations is less than 5μs, the waveform characteristics are consistent, the lightning strike point is located in the area surrounded by multiple stations and the calculation is converged, and the signal quality score of the participating stations is higher than the threshold. The anti-interference logic is that single-station triggering is not directly reported, but must be verified by neighboring stations. Clustering is performed on pulses of the same thunderstorm process to automatically remove abnormal sites with residuals exceeding the threshold, and electromagnetic interference segment data is shielded to avoid false triggering and duplicate reporting.

[0032] Step S2, in conjunction with a local storage strategy, includes the following for data retention: Data caching is handled by an on-chip high-speed circular buffer to ensure continuous data acquisition without data loss; combined with local industrial-grade SSD storage, it supports rolling over and locking of critical data.

[0033] Data acquisition and storage uses SSDs. When the usage reaches a set threshold of 80%, the acquisition module automatically starts a rolling mode to delete the earliest ordinary lightning strike event in chronological order, ensuring that the storage space is always available, without overflow or downtime. In case of network failure, the data is first written to the local disk and continuously stored. After recovery, it is automatically resumed and then uploaded via 4G / 5G. After successful upload, it can be retained or deleted according to the policy.

[0034] In one embodiment, the transmission of lightning strike signals to the backend service platform via an encrypted communication link in step S3 refers to the local storage of lightning strike event data being encrypted and transmitted via an industrial-grade 4G / 5G communication module. The communication link employs APN / VPN private network + TLS two-way authentication to ensure the reliability and security of data transmission.

[0035] The monitoring stations use industrial-grade SSDs for local data storage, with 80% storage occupancy set as the rolling over threshold. When the storage space utilization reaches 80%, the earliest ordinary lightning strike event data is automatically overwritten in chronological order. At the same time, a critical lightning strike event locking strategy is set. When a lightning strike event meets any of the following conditions: signal amplitude exceeds the threshold, the horizontal distance between the strike point and the tower is less than 50m, an orange / red warning is triggered, it is determined to be a bypass / counterattack / direct strike fault, or multiple lightning strikes to the same tower within a short period of time, it is automatically marked as a critical event and locked in storage, preventing it from being deleted by rolling over. The locked data is retained for a long time for fault tracing and analysis.

[0036] In step S3, the back-end service platform performs a quality score on the lightning strike signal based on four indicators: timestamp validity, channel integrity, signal amplitude stability, and noise ratio. The quality score results serve as the basis for assigning signal quality weights in the subsequent lightning strike location calculation. Abnormal signals with invalid timestamps, missing channels, signal distortion, excessive noise ratio, or substandard quality scores are entered into a dead-signal queue for manual review by the back-end service platform. Valid signals are marked with characteristic parameters and then proceed to the subsequent location calculation stage.

[0037] Among them, the dead letter queue is a dedicated isolated queue in the platform used to temporarily store abnormal data.

[0038] Specifically, the dead-letter queue is a dedicated queue used in the platform's data preprocessing stage to cache and temporarily store abnormal lightning strike signal data. Data that cannot enter the normal location and calculation process due to invalid timestamps, missing channels, signal distortion, excessive noise, or substandard quality scores is sent to this queue for isolation and temporary storage. This prevents abnormal data from interfering with normal calculations, while retaining the original data for subsequent manual review, troubleshooting, and algorithm optimization.

[0039] Location calculation refers to the mathematical calculation process of establishing and solving a system of equations for the lightning strike point based on the Time Difference of Arrival (TDOA) algorithm and spatiotemporal data from monitoring stations to obtain the initial location of the lightning strike. In step S3, the back-end service platform verifies the legality of the lightning strike signal. This means that after receiving the lightning strike data packet transmitted via the 4G / 5G encrypted link, the back-end service platform first performs compliance, integrity, authenticity, and security checks, filtering out illegal, tampered, incomplete, and forged data packets, and only allowing legal and valid data to enter the subsequent quality scoring and location calculation process.

[0040] It should be noted that the abnormal signal removal in this implementation plan is based on the signal quality scoring results, removing interference signals with abnormal timestamps, missing channels, waveform distortion, excessive noise ratio, and failure of multi-station collaborative verification; the feature parameters are extracted from the valid lightning strike signal to extract amplitude, energy, rise edge slope, pulse count, timestamp, and waveform feature parameters, which are used for subsequent TDOA location calculation, fault type determination, and graded early warning.

[0041] Specifically, the abnormal lightning strike signals that do not meet the requirements are excluded, including signals with invalid timestamps or abnormal transitions; signals with missing channels or incomplete waveforms; signals with severe distortion or falsification; interference waveforms with excessively high background noise and substandard quality scores; and suspected electromagnetic interference signals that are triggered in isolation by a single station or fail multi-station collaborative verification.

[0042] For valid lightning strike waveforms that pass verification and are retained, key fixed characteristic parameters are automatically extracted, including peak amplitude of the lightning signal; short-time pulse energy; waveform rising edge slope; frequency band energy ratio; impulse pulse count; signal arrival timestamp; waveform start and end times; and signal quality score.

[0043] In one embodiment, step S4 involves constructing a set of equations based on the TDOA (Time Difference of Arrival) algorithm, combined with the monitoring station coordinates, signal detection time, and signal quality weights. The weighted least squares method is then used to solve the equations to obtain the initial lightning strike location. The calculation steps are as follows: Based on the TDOA algorithm, a set of nonlinear constraint equations for the time difference of arrival of lightning signals from multiple stations and the spatial distance are established. The coordinates of each monitoring station and the arrival time of the signal feature points are used as inputs. The set of equations is linearized by first-order Taylor expansion at the initial estimation point. A diagonal weight matrix W is constructed based on the signal quality score results, with higher weights assigned to high-quality signals and lower weights assigned to abnormal signals. Subsequently, a weighted sum of squared residuals is constructed as the objective function, and the weighted least squares solution formula is obtained by taking the extreme value through differentiation. The estimated value of the lightning strike point is continuously corrected through iterative calculation until the residuals converge. Finally, the initial coordinates of the lightning strike and the calculated residuals are output to complete the location calculation.

[0044] Specifically, a lightning strike point calculation model is constructed based on the TDOA (Time Difference of Arrival) algorithm. Utilizing the time difference in arrival times of the lightning pulse at different monitoring stations, a nonlinear hyperboloid equation system is built for position inversion. The steps include: Let the location of the lightning strike be... The time of occurrence is , No. The coordinates of each monitoring station are , The signal arrival time is The speed of lightning wave propagation is The distance equation is: ; Take the first If station is the reference station, then the first... Distance difference between station and reference station for: ; To avoid non-convergence in nonlinear iteration, an intermediate variable is introduced. Transform the equation into a linear matrix form: ; Construct matrix equations ;in ; Introducing a signal quality weight matrix Solve for the optimal position vector: ; in The signal quality weight matrix is ​​determined by the covariance matrix of the measurement errors of each station. Signals with high quality scores are given greater weight to ensure that the positioning results are statistically optimal.

[0045] Furthermore, the platform uses the precise coordinates of each monitoring station, the detection time of lightning strike signal characteristic points, and signal quality weights as inputs to establish a general form of the time difference of arrival equation system: Set up monitoring stations The coordinates are The time when the characteristic point of the lightning strike signal was detected was The coordinates of the lightning strike point are The speed of signal propagation is Then the arrival time difference satisfies: ; The default propagation speed of a lightning strike signal is the speed of light in air. For scenarios involving complex local terrain, mountain obstruction, and elevation differences, a layered propagation velocity correction model is adopted, introducing a correction coefficient in the range of 0.98 to 1.0 to compensate for the equivalent propagation velocity and improve the solution accuracy.

[0046] In the solution process, a set of nonlinear constraint equations relating the time difference of arrival of lightning signals from multiple stations to the spatial distance is established based on the TDOA algorithm. Using the coordinates of each monitoring station and the arrival times of signal characteristic points as inputs, the equations are linearized by a first-order Taylor expansion at the initial estimation point. A weighted residual sum of squares minimization objective function is constructed, and the initial coordinates of the lightning strike are solved through weighted least squares iteration. The iterative correction formula is as follows: ; Iterate until the residual converges, and finally output the initial coordinates of the lightning strike and the calculated residual to complete the location calculation.

[0047] In one embodiment, step S5, which corrects the initial lightning strike location according to an engineering correction strategy and outputs the accurate lightning strike point with an error ellipse and confidence level, includes the following steps: To address the impact of complex terrain on positioning accuracy, a three-step engineering correction is performed on the initial positioning results to improve positioning accuracy: Step S5.1: Based on the terrain elevation difference between the monitoring station and the initial point of lightning strike obtained from the geographic information system, introduce an empirical terrain correction coefficient. and path terrain roughness The equivalent propagation speed of electromagnetic waves is corrected, and the corrected model is as follows: ; in The speed of light; For high-altitude mountainous areas, a height compensation item will be further introduced. The actual propagation path length of electromagnetic waves is compensated to eliminate positioning drift caused by altitude differences. The calculation formula is as follows:

[0048] In the formula, To monitor the elevation difference between the monitoring station and the point of lightning strike, The straight-line distance between two points in a plane. This is the horizontal projection distance; Step S5.2 Establish a reconstruction residual verification mechanism and calculate the solution residuals for each monitoring station. The calculation formula is as follows: ; Among them No. The theoretical distance difference between the station and the reference station, , , These are the initial positioning coordinates. This is an estimated distance between the reference station and the point of lightning strike. Set threshold ,in Let the standard deviation of the residuals be denoted as , if the standard deviation of a certain station is . If the threshold is exceeded, the station is determined to be significantly affected by multipath effects or interference, and it is removed from the station array and WLS calculation is performed again until the residual converges. Step S5.3 Performs spatiotemporal clustering analysis on multiple return stroke signals collected during the same thunderstorm process, using an improved territory clustering algorithm, which is a pulse event clustering method based on spatiotemporal neighborhood rules. The clustering criteria are as follows: ; ; Pulses that meet the above time and space conditions are grouped into the same lightning cluster, and the energy weighting center is calculated as the final output point, which effectively avoids the aliasing of positioning results caused by multiple return strikes. After completing the corrections in steps S5.1, S5.2, and S5.3, an error covariance matrix is ​​constructed based on the solved residuals and the site covariance matrix. The error ellipse parameters and location confidence are calculated, and complete and accurate lightning strike point data containing coordinates, residuals, error ellipses, and confidence are output.

[0049] In one embodiment, the spatial correlation parameters in step S6 include tower location, conductor segment, line corridor boundary, tower plane coordinates, tower elevation, tower type structure, transmission line span, line corridor width, spatial coordinates of ground wire suspension point, spatial coordinates of conductor suspension point, ground wire protection angle, horizontal straight-line distance between the lightning strike point and the nearest tower, and the shortest vertical distance from the lightning strike point to the transmission conductor segment; the lightning strike signal characteristics include signal peak amplitude, short-time pulse energy, waveform rising edge slope, and background adaptive noise threshold reference.

[0050] Specifically, after completing the lightning strike location calculation and engineering correction, the precise lightning strike points are uniformly projected onto the transmission line corridor geographic information system. Coordinate system matching and precise spatial registration are achieved, and the elevation component of the lightning strike point is removed while retaining the planar projection coordinates. This enables precise overlay and visual labeling of the lightning strike points on the electronic map vector layer. Using the lightning strike projection point as the center and an 800m neighborhood search radius, the transmission line vector data within the range is retrieved, including topological information such as tower locations, conductor segments, and line corridor boundaries. Euclidean distance calculations are used to filter out the nearest and adjacent towers, accurately extracting basic parameters such as tower planar coordinates, elevation, and tower type. Simultaneously, combining the inherent design parameters of the line with spatial calculations, the transmission line span and line corridor width are automatically obtained, and the spatial coordinates of the ground wire and conductor suspension points are used as the basis for further analysis. By accurately calculating the line ground wire protection angle through arctangent calculation, various geographical and electrical characteristic parameters under lightning strike scenarios are fully obtained. Based on this, the horizontal and vertical distances between the lightning strike point and the nearest tower and transmission line are quantitatively calculated to determine whether the lightning strike point is within the ground wire shielding protection range, and to establish a precise spatial correlation between lightning strike events and transmission line equipment. By retrieving the coordinates of the nearest tower and the line segment data of adjacent towers through a geographic information system, the horizontal straight-line distance between the lightning strike point and the target tower and the shortest vertical distance of the line segment from the lightning strike point to the transmission line are calculated respectively. At the same time, the span parameters between adjacent towers are solved to complete the spatial correlation quantitative analysis between the lightning strike location and the transmission line. Simultaneously, characteristic parameters such as the peak amplitude, short-time pulse energy, and waveform rising edge slope of the lightning signal are extracted, and the characteristic quantification calculation is completed with the background adaptive noise threshold as a reference.

[0051] Step S6, determining the type of lightning strike fault and classifying the impact level of the lightning strike, includes the following steps: If the lightning strike occurs outside the shielded area of ​​the ground wire protection angle and hits the conductor area, it is considered a backflash. If the lightning strike hits the tower body or ground wire and causes line insulation flashover, it is considered a backflash. If the lightning strike directly hits the exposed area of ​​the conductor without ground wire protection, it is considered a direct strike. After determining the fault type, the impact of the lightning strike on the line is classified into four levels: no impact, low impact, medium impact, and high impact, based on the horizontal distance between the lightning strike point and the tower, the signal amplitude, the lightning energy intensity, and the fault type. Among them, the no-impact level corresponds to lightning strikes far from the line corridor and weak signal energy, which do not cause any impact on the transmission equipment; the low-impact level corresponds to lightning strikes near the line corridor and small signal amplitude, with no risk of equipment failure; the medium-impact level corresponds to lightning strikes close to the towers and conductors, with a certain amount of lightning energy, and potential line flashover risk; the high-impact level corresponds to direct strikes, bypass strikes, and backflash strikes, which are effective fault lightning strikes with high signal energy and strike points close to the line equipment, which can easily cause line fault tripping and equipment damage. This provides complete and accurate quantitative data support for subsequent lightning strike classification and early warning, fault review and analysis, and line lightning protection optimization and transformation.

[0052] It should be noted that in the process of determining backflash lightning strikes, the horizontal distance between the lightning strike point and the nearest tower is calculated. When the horizontal distance is less than or equal to 30m, and the lightning pulse energy and waveform rise slope reach the preset thresholds respectively, and the lightning strike area is within the influence range of the ground wire and the tower grounding, it is determined to be a backflash lightning strike. In the process of determining lightning strikes that bypass the transmission line, the theoretical shielding and protection area is determined based on the inherent ground wire protection angle of the line. When the lightning strike point is located outside the shielding area of ​​the protection angle, the shortest distance from the transmission line corridor is no more than 50m, and the lightning strike signal amplitude meets the bypass characteristic threshold and the characteristics of the grounding flashover waveform without a tower, it is determined to be a lightning strike that bypasses the transmission line.

[0053] By combining spatial distance threshold, ground wire protection angle range, lightning signal amplitude, impact energy and rising edge characteristics in a multi-dimensional joint discrimination, the two types of typical lightning faults can be accurately distinguished and automatically identified.

[0054] Furthermore, after completing lightning strike location calculation, terrain correction, abnormal data removal, multi-pulse clustering, GIS spatial projection matching, fault type identification, and impact level rating, the lightning strike fault analysis report generation process is automatically triggered.

[0055] Specifically, the system first collects all dimensions of data related to the lightning strike event, including the original acquired waveforms, timing synchronization status, transmission logs, calculation parameters, error ellipse and confidence results, geospatial correlation parameters, lightning waveform characteristics, fault determination results, and early warning records. Data validity verification and cleaning are then performed. Next, based on a built-in standardized report template, the system automatically fills in basic information about the lightning strike event, equipment operating status, waveform feature analysis, a complete review of the location calculation process, location accuracy assessment, line spatial matching analysis, criteria for determining the lightning fault type, and evaluation of the lightning impact level. Combining preset lightning strike mechanism reasoning rules, the system intelligently generates fault cause analysis, risk assessment conclusions, and operation and maintenance optimization suggestions. Finally, it automatically synthesizes and renders a structured lightning strike fault analysis report, which is then bound to and archived with all event data. This achieves a closed-loop process for lightning strike fault analysis, from signal acquisition and location calculation to intelligent analysis and automatic report output.

[0056] In one embodiment, the graded early warning rule in step S7 includes a four-level quantitative graded early warning mechanism of blue, yellow, orange, and red, with a unified fixed judgment threshold set: A blue alert corresponds to a long-distance, weak lightning event where the vertical distance from the lightning strike point to the line is greater than 200m, there are fewer than 3 effective calculation stations, and the signal amplitude is 1.5 to 2.5 times the background noise. A yellow alert requires at least 3 valid solution sites, a location confidence level of at least 0.75, a lightning strike distance between the tower and the line corridor between 80m and 200m, a signal amplitude of 2.5 to 3.5 times the background noise, and a rising edge slope of less than 0.8mV / μs. The criteria for issuing an orange alert are: the distance from the lightning strike to the tower is no more than 80m, the distance from the outer boundary of the line corridor is no more than 50m, the rise slope is no less than 0.8mV / μs, and the solution residual is less than 30m. A red alert can be triggered if any two of the following conditions are met: a lightning strike within 30m of the tower in a high-risk backflash zone; a lightning strike within 50m of the conductor corridor outside the protection angle in a high-risk backflash zone; a signal amplitude of not less than 3.5 times the background noise; a lightning strike energy of not less than 8.5 × 10³ quantization units; at least two multi-pulse return strokes within 500ms; and a deterministic fault lightning strike event.

[0057] A red alert can be triggered if any two of the following conditions are met: lightning strike within 30m of the tower in a high-risk backflash zone; lightning strike outside the protection angle within 50m of the conductor corridor in a high-risk backflash zone; signal amplitude not less than 3.5 times the background noise; lightning strike energy not less than 8.5×10³ quantization units; and at least two multi-pulse return strokes within 500ms. A deterministic fault lightning strike event can directly trigger a red alert without the need for other conditions.

[0058] Furthermore, the distance of the corridor is determined by the shortest vertical distance from the lightning strike point to the line conductor, and the corridor is divided into the core high-risk zone, the 0-50m extended danger zone, the 50-80m adjacent warning zone, the outer safety zone above 80m, and the far-distance risk-free zone above 200m. Various distance thresholds, waveform characteristic thresholds, and positioning accuracy thresholds are all determined comprehensively based on the electrical geometric model of the transmission line lightning protection, the on-site measured lightning statistics, the influence range of equipment grounding, and the error constraints of the TDOA positioning algorithm, to ensure the quantitative and unified standard of early warning classification.

[0059] Furthermore, based on this, false alarm suppression and anti-repeated early warning strategies are deployed simultaneously. A 500ms time window is set to achieve multi-pulse clustering, and multiple return signals of the same thunderstorm process are merged and combined. Using 80m as a spatial threshold, scattered lightning strike events within the same line section and the same adjacent tower range are integrated into the same tower and cluster, and a unified single comprehensive judgment result is generated. A 15-minute early warning cooling cycle is configured. After a single tower or line section triggers an early warning, repeated pushes of the same level are blocked during the cooling period, and only higher-risk lightning strikes are allowed to trigger early warnings. Isolated pulses, residual abnormal data, and low-frequency electromagnetic interference signals are simultaneously removed from single stations to reduce the probability of false alarms from the source. Through multiple collaborative means such as time clustering, spatial merging, graded cooling, and interference filtering, invalid alarms are effectively suppressed and frequent repeated early warnings in a short period of time are avoided, so as to achieve accurate classification of lightning strike faults, stable early warning, and closed-loop analysis of the entire process.

[0060] Meanwhile, the method constructs a full lifecycle management system for lightning strike fault data, which stores all data such as lightning strike events, location results, fault analysis reports, and early warning records in a structured manner, supports multi-dimensional retrieval and data analysis, and uses RESTful API interfaces for data sharing. Internally, it implements hierarchical authorization according to roles such as scheduling, operation and maintenance, and administrators, while externally, it implements least privilege read-only access. All data access and operation are recorded in full audit logs, enabling data support for accurate post-event tracing of lightning strike faults and optimization of lightning protection strategies.

[0061] In step S7, the rule for triggering graded early warning based on lightning strike location results, fault analysis results, and lightning signal characteristics refers to combining spatial correlation parameters such as the horizontal distance between the lightning strike point and the target tower, the shortest vertical distance from the lightning strike point to the transmission line, and the line span. Simultaneously, multi-dimensional characteristic indicators such as the amplitude of the lightning signal, impact energy, and waveform rising edge slope are extracted. Based on the shielding range of the ground wire protection angle and the boundary range of the line corridor, spatial matching and judgment are completed. The three types of lightning faults—backflash, bypass, and direct strike—are quantitatively distinguished. In combination with multi-source evaluation indicators, four levels of lightning hazard are classified: no impact, low impact, medium impact, and high impact.

[0062] In one embodiment, in a power transmission corridor in a high-altitude mountainous area, where the terrain is undulating, the elevation difference is significant, the soil resistivity is high, and thunderstorms are frequent, four monitoring stations are set up along the power transmission line in the mountainous area. Each station is equipped with a GNSS disciplined clock, a PPS second pulse output module, and a high-speed acquisition board. Multi-station clock alignment is achieved by phase-locking with a 1PPS pulse clock, and the communication cable delay of each station is pre-calibrated to control the time synchronization accuracy of the multi-stations to the level of 200ns.

[0063] Each monitoring station uses a 1MS / s, 16-bit multi-channel synchronous ADC for parallel acquisition, and is equipped with a lightning pulse broadband conditioning circuit at the front end to complete signal impedance matching, attenuation and anti-aliasing filtering.

[0064] During a thunderstorm, multiple lightning return pulses occurred in the monitoring area. The system adopted a five-dimensional weighted composite triggering strategy based on level threshold, short-time energy, rise edge slope, frequency band energy ratio, and impact count, with weighting ratios of 25%, 30%, 20%, 15%, and 10%, respectively. The waveform indicators were normalized and weighted to calculate a comprehensive score, and the lightning pulses were determined to be valid lightning strike events.

[0065] The original waveform was stored using an on-chip ring buffer and an industrial-grade SSD. The rolling coverage threshold was set to 80% storage space occupancy. The waveform of this lightning strike was marked as a critical event and locked for retention.

[0066] The collected lightning strike data was encrypted and uploaded via an industrial-grade 5G module, using an APN private network + TLS two-way authentication encrypted link. After receiving the waveform, the backend service platform first verifies the legality of the data packets, and then scores the signal quality based on four indicators: timestamp validity, channel integrity, amplitude stability, and noise ratio. The waveforms uploaded by the four monitoring stations were all free of distortion and packet loss, and the timestamps were accurate. All of them were determined to be valid signals, with no abnormal data generated and no signals entering the dead-letter queue.

[0067] The background system constructs a TDOA nonlinear equation system based on the coordinates of four monitoring stations, pulse arrival time, and signal quality score weights. The equation system is then linearized by a first-order Taylor expansion at the initial estimated position. A diagonal weight matrix is ​​constructed based on the waveform quality of each station, and the main monitoring station with the best quality is assigned the highest weight.

[0068] The weighted least squares method was used for iterative solution, and the estimated impact point was continuously corrected until the residual converged. The initial coordinates of this lightning strike were finally obtained. The initial residual was 42.6m, and it was preliminarily determined that the center of the thunderstorm was located in the mountainous area outside the line corridor.

[0069] Correction measures were implemented to address the unique terrain characteristics of high-altitude mountainous areas. Based on the elevation difference between the monitoring station and the lightning strike point, the terrain roughness and empirical correction coefficient are input to correct the equivalent propagation speed of electromagnetic waves; at the same time, the altitude compensation term is calculated to correct the positioning offset caused by the inclined propagation path in mountainous areas. The residuals of each station were calculated, and the outlier threshold was set at 2.5 times the standard deviation of the residuals. The residuals of the four monitoring stations did not exceed the limit and were all retained for the final calculation. Three lightning return pulses were collected within a 500ms time window during this thunderstorm process. The spatial distance between the pulses was less than 2km, which met the territorial clustering criteria and were classified into the same lightning cluster. The energy weighted center algorithm was used to correct the impact point, and the final output was accurate lightning strike coordinates. After optimization, the solution residual was reduced to 18.3m, and an error ellipse and a location confidence of 92.7% were generated.

[0070] The corrected lightning strike point was projected onto the GIS geographic information system of the mountain transmission line. The parameters of the towers, conductors, ground wire attachment points, and line corridor boundaries within an 800m radius of the lightning strike point were retrieved. Spatial calculations showed that the lightning strike point was 27.4m away from the nearest tower, outside the ground wire protection angle shielding area, and close to the danger zone extending from the conductor corridor. Combined with waveform characteristics, the lightning strike had high energy and a significant rise slope, and there was no tower grounding flashover waveform. Therefore, this lightning strike event was determined to be a bypass fault.

[0071] Based on a comprehensive assessment of spatial distance, lightning strike energy, and waveform amplitude, this lightning strike was determined to be of a high impact level, posing a risk of line insulation flashover and tripping.

[0072] According to the four-level early warning judgment rules, if any two of the following conditions are met for a red early warning: a high-risk area within 30m of the tower, a dangerous area within 50m of the conductor corridor, high signal amplitude, and multiple echoes within 500ms, the platform will directly trigger a red high-level early warning.

[0073] Warning information is sent to maintenance personnel via SMS and WeChat, and a 15-minute warning cooling mechanism is activated to suppress repeated alarms in the same area.

[0074] The waveforms, location coordinates, error parameters, fault judgment conclusions, and early warning records of the entire lightning strike process are uniformly archived and stored in the database, and a lightning strike fault analysis report is automatically generated for use in reviewing lightning protection of mountain lines, risk assessment, and line modification and optimization.

[0075] A power transmission line lightning strike fault location and analysis system, comprising: The front-end monitoring device is deployed along the power transmission line to collect lightning pulse signals, complete BeiDou high-precision time synchronization, buffer and store signals locally, and transmit lightning signals to the outside through the communication transmission network. The communication transmission network is used to enable encrypted data transmission and remote configuration command issuance between the front-end monitoring device and the back-end service platform; The communication transmission network is a 4G / 5G wireless communication link; The back-end service platform is used to receive lightning strike signals uploaded by the front-end monitoring devices and perform decryption verification, signal quality scoring, and preprocessing.

[0076] The implementation backend service platform is a server cluster equipped with positioning and calculation software, geographic information system, early warning push system and database. It is used to receive lightning strike signals uploaded by front-end monitoring devices and perform decryption verification, signal quality scoring and preprocessing. It performs initial calculation of lightning strike point based on the TDOA time difference of arrival algorithm combined with weighted least squares method. Through engineering correction strategy, it obtains accurate lightning strike point with error ellipse and confidence level. Combined with the geographic information of transmission line corridor, it completes the determination of lightning fault type and impact level classification, executes graded early warning rules and pushes early warning information through multiple channels.

[0077] In one embodiment, the front-end monitoring device includes a multi-channel acquisition unit, a BeiDou time synchronization unit, a high-speed cache and storage unit, a communication transmission unit, and a main control processing unit. The multi-channel acquisition unit includes a 16-bit ADC, a front-end signal conditioning circuit and an anti-aliasing filter, supports continuous high-speed acquisition at 1MS / s, and is suitable for acquisition of wide-amplitude pulse signals from direct lightning strikes and induced lightning strikes. The Beidou time synchronization unit includes a GNSS disciplined clock, a PPS second pulse output module, a clock phase-locked loop module, and a cable delay calibration circuit. It achieves 200ns-level time reference unification for monitoring stations, monitors the timekeeping stability of stations and the relative error of multiple stations, and triggers operation and maintenance alarms when abnormalities occur. The high-speed cache and storage unit includes an on-chip FIFO, a DDR ring buffer, and an industrial-grade SSD, supporting pre-triggered / post-triggered data caching, data rolling over, and storage locking for critical lightning strike events. The communication transmission unit adopts an industrial-grade 4G / 5G communication module, which supports APN / VPN private network and TLS bidirectional encrypted transmission; The main control processing unit is used to realize the judgment of combined lightning strike signals, data packaging, and equipment abnormality alarm.

[0078] In one embodiment, the back-end service platform further includes a data receiving module, a signal preprocessing module, a positioning calculation module, a fault analysis module, a graded early warning module, and a data management module; The data receiving module is used to receive encrypted data sent by the front-end monitoring device and to complete decryption and legality verification. The signal preprocessing module is used to score the quality of lightning strike signals, remove abnormal signals, extract valid lightning strike event feature parameters, and store abnormal signals in a dead-letter queue for manual review. The positioning and calculation module is used to construct a set of equations for TDOA lightning strike point calculation, introduce signal quality weights and use weighted least squares method to solve the initial position of lightning strike, perform engineering correction processing such as terrain propagation path correction, abnormal station elimination and multi-pulse cluster clustering, and output accurate lightning strike point with error ellipse and confidence level. The fault analysis module is used to project the lightning strike point onto the geographic information system of the transmission line corridor, calculate the horizontal distance between the strike point and the tower, and the related parameters of the line span. It combines the signal amplitude and energy characteristics to determine the fault types of backflash, backflash, and direct strike, and classifies them into four levels of impact: no impact, low impact, medium impact, and high impact. The graded early warning module is used to execute four-level graded early warning rules (blue, yellow, orange, and red), push early warning information through multiple channels such as WebSocket, SMS, and WeChat Work, and configure false alarm suppression and cooling merging mechanisms for events in the same tower and cluster. The data management module is used to realize the structured storage and multi-dimensional retrieval of lightning strike events, location results, fault analysis reports, and early warning records throughout their entire lifecycle. It achieves data sharing through RESTful API interfaces, implements role-based hierarchical authorization internally, adopts minimum access read-only access externally, and retains full audit logs.

[0079] In this implementation plan, the industrial-grade SSD is set to 80% storage occupancy as the rolling over threshold. Once the threshold is reached, ordinary lightning strike event data will be automatically deleted in chronological order. Events that meet any of the following conditions are marked as critical events and their storage is locked, preventing rolling over deletion: signal amplitude exceeding the threshold, horizontal distance between the strike point and the tower less than 50m, triggering an orange / red warning, being determined as a bypass / counterattack / direct strike fault, or multiple lightning strikes to the same tower in a short period of time.

[0080] In this implementation scheme, when the positioning and calculation module establishes the equation system for calculating the lightning strike point, it approximates the propagation speed of the lightning signal with a uniform medium, and uses a correction factor of 0.98 to 1.0 to correct the propagation speed in layers for complex local terrain. The TDOA nonlinear equation system is linearized by first-order Taylor expansion, and a weight matrix is ​​constructed and solved by weighted least squares iteration.

[0081] In the description of this application, it should be noted that the terms "upper," "lower," etc., indicating the orientation or positional relationship are based on the orientation or positional relationship shown in the accompanying drawings, and are only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of this application. Unless otherwise expressly specified and limited, the terms "installed," "connected," and "linked" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; they can refer to the internal communication between two elements. For those skilled in the art, the specific meaning of the above terms in this application can be understood according to the specific circumstances.

[0082] It should be noted that in this application, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0083] The above description is merely a specific embodiment of this application, enabling those skilled in the art to understand or implement this application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features claimed herein.

Claims

1. A method for locating and analyzing lightning strike faults in power transmission lines, characterized in that, Includes the following steps: S1: Establish a multi-monitoring station BeiDou time synchronization system to unify the time reference of each monitoring station; S2: Collects lightning pulse signals, captures lightning events according to a composite triggering strategy, and completes data retention in conjunction with a local storage strategy; S3: The lightning strike signal is sent to the back-end service platform via an encrypted communication link. The back-end service platform verifies the legality and scores the quality of the lightning strike signal, eliminates abnormal signals, and extracts feature parameters. S4: Based on the TDOA time difference of arrival algorithm, a set of equations is constructed by combining the coordinates of the monitoring station, the signal detection time and the signal quality weight, and the weighted least squares method is used to solve for the initial position of the lightning strike. S5: Correct the initial position of the lightning strike according to the engineering correction strategy, and output the accurate lightning strike point with error ellipse and confidence level; S6: Project the precise lightning strike point onto the geographic information system of the transmission line corridor, and combine spatial correlation parameters with lightning signal characteristics to determine the type of lightning fault and classify the level of lightning impact; S7: Based on the lightning strike location results, fault analysis results, and lightning strike signal characteristics, trigger the graded early warning rules and retain all lightning strike data.

2. The method for locating and analyzing lightning strike faults in transmission lines according to claim 1, characterized in that: The steps in step S1 to build a multi-monitoring station BeiDou time synchronization system include deploying GNSS disciplined clocks, PPS second pulse output modules and acquisition boards at the transmission line monitoring stations, aligning the acquisition boards with the BeiDou time signal through clock phase-locking, calibrating the cable delay of each station to unify the time reference, monitoring the timekeeping stability of a single station and the relative error of multiple stations in real time, triggering operation and maintenance alarms when abnormalities occur, and achieving 200ns-level time reference unification for monitoring stations. In step S1, a unified time source is provided by the BeiDou GNSS disciplined clock. The clock synchronization of the acquisition board is achieved by 1PPS pulse phase-locking. Combined with full-link delay calibration compensation, real-time calibration of relative time difference between multiple stations, monitoring of loss of lock and over-threshold anomalies and timekeeping guarantee, the sampling time of each station is unified to the 200ns level accuracy, and a high-precision timestamp is bound to the sampling data, thereby ensuring the time consistency of lightning strike signal acquisition.

3. The method for locating and analyzing lightning strike faults in transmission lines according to claim 1, characterized in that: In step S2, the lightning pulse signal is acquired using a multi-channel synchronous ADC in parallel with a sampling rate of 1MS / s and 16bit. The channels do not interfere with each other. The front end is equipped with a wideband lightning pulse conditioning circuit with attenuation, impedance matching, and anti-aliasing filtering functions to adapt to wide-amplitude pulse signal input from direct lightning strikes and induced lightning strikes, avoiding signal clipping and saturation distortion. The clock source uses a Beidou disciplined clock and a PPS second pulse, and is calibrated to a high-precision time reference of 200ns through clock phase-locked loop to ensure strict alignment of sampling times across multiple stations and channels. In step S2, the composite triggering strategy is a multi-index weighted triggering based on the level threshold, short-time energy, rising edge slope, bandwidth energy ratio, and impulse count. The weighting ratios for each index are as follows: level threshold 25%, short-time energy 30%, rising edge slope 20%, bandwidth energy ratio 15%, and impulse count 10%. After normalizing each index, the weighted sum is calculated to obtain a comprehensive triggering score. If the comprehensive score is higher than the set threshold, it is determined to be a valid lightning strike trigger.

4. The method for locating and analyzing lightning strike faults in transmission lines according to claim 1, characterized in that: In step S3, the lightning strike signal is sent to the back-end service platform via an encrypted communication link. This means that the locally stored lightning strike event data is encrypted and sent to the back-end service platform via an industrial-grade 4G / 5G communication module. The communication link adopts APN / VPN private network + TLS two-way authentication to ensure the reliability and security of data transmission. In step S3, the back-end service platform scores the quality of the lightning strike signal based on four indicators: timestamp validity, channel integrity, signal amplitude stability, and noise ratio. The quality score results serve as the basis for assigning signal quality weights in subsequent lightning strike location calculations. Abnormal signals with invalid timestamps, missing channels, signal distortion, excessive noise, or substandard quality scores are entered into a dead-signal queue for manual review by the back-end service platform. Valid signals are marked with characteristic parameters and then enter the subsequent location calculation stage.

5. The method for locating and analyzing lightning strike faults in transmission lines according to claim 1, characterized in that: In step S4, based on the TDOA (Time Difference of Arrival) algorithm, a set of equations is constructed by combining the monitoring station coordinates, signal detection time, and signal quality weights. The weighted least squares method is then used to solve the equations to obtain the initial location of the lightning strike. The calculation steps are as follows: Based on the TDOA algorithm, a set of nonlinear constraint equations for the time difference of arrival of lightning signals from multiple stations and the spatial distance are established. The coordinates of each monitoring station and the arrival time of the signal feature points are used as inputs. The set of equations is linearized by first-order Taylor expansion at the initial estimation point. A diagonal weight matrix W is constructed based on the signal quality score results, with higher weights assigned to high-quality signals and lower weights assigned to abnormal signals. Subsequently, a weighted sum of squared residuals minimization objective function is constructed, and the weighted least squares solution formula is obtained by taking the extreme value through differentiation; The estimated lightning strike point is continuously corrected through iterative calculations until the residual converges. Finally, the initial coordinates of the lightning strike and the calculated residual are output, thus completing the location calculation.

6. The method for locating and analyzing lightning strike faults in transmission lines according to claim 1, characterized in that: Step S5 involves correcting the initial lightning strike location according to an engineering correction strategy, and outputting the precise lightning strike point with an error ellipse and confidence level. This includes the following steps: Step S5.1: Based on the terrain elevation difference between the monitoring station and the initial point of lightning strike obtained from the geographic information system, introduce an empirical terrain correction coefficient. and path terrain roughness The equivalent propagation speed of electromagnetic waves is corrected, and the corrected model is as follows: ； in The speed of light; For high-altitude mountainous areas, a height compensation item will be further introduced. The actual propagation path length of electromagnetic waves is compensated to eliminate positioning drift caused by altitude differences. The calculation formula is as follows: ； In the formula, To monitor the elevation difference between the monitoring station and the point of lightning strike, The straight-line distance between two points in a plane. This is the horizontal projection distance; Step S5.2 Establish a reconstruction residual verification mechanism and calculate the solution residuals for each monitoring station. The calculation formula is as follows: ； Among them No. The theoretical distance difference between the station and the reference station, , , These are the initial positioning coordinates. This is an estimated distance between the reference station and the point of lightning strike. Set threshold ,in Let the standard deviation of the residuals be denoted as , if the standard deviation of a certain station is . If the threshold is exceeded, the station is determined to be significantly affected by multipath effects or interference, and it is removed from the station array and WLS calculation is performed again until the residual converges. Step S5.3 Performs spatiotemporal clustering analysis on multiple return stroke signals collected during the same thunderstorm process, using an improved territory clustering algorithm. The clustering criteria are as follows: ； ； Pulses that meet the above time and space conditions are grouped into the same lightning cluster, and the energy weighting center is calculated as the final output point, effectively avoiding the aliasing of positioning results caused by multiple return strikes.

7. The method for locating and analyzing lightning strike faults in transmission lines according to claim 1, characterized in that: Step S6, determining the type of lightning strike fault and classifying the impact level of the lightning strike, includes the following steps: If the lightning strike occurs outside the shielded area of ​​the ground wire protection angle and hits the conductor area, it is considered a backflash. If the lightning strike hits the tower body or ground wire and causes line insulation flashover, it is considered a backflash. If the lightning strike directly hits the exposed area of ​​the conductor without ground wire protection, it is considered a direct strike. After determining the fault type, the impact of the lightning strike on the line is classified into four levels: no impact, low impact, medium impact, and high impact, based on the horizontal distance between the lightning strike point and the tower, the signal amplitude, the lightning energy intensity, and the fault type. Among them, the no-impact level corresponds to lightning strikes far from the line corridor and weak signal energy, which do not cause any impact on the power transmission equipment; the low-impact level corresponds to lightning strikes near the line corridor and small signal amplitude, with no risk of equipment failure; the medium-impact level corresponds to lightning strikes close to the towers and conductors, with a certain amount of lightning energy, and there is a potential risk of line flashover; the high-impact level corresponds to direct strikes, bypass strikes, and backflash strikes, which are effective fault lightning strikes with high signal energy and strike points close to the line equipment, which can easily cause line fault tripping and equipment damage.

8. The method for locating and analyzing lightning strike faults in transmission lines according to claim 1, characterized in that: Step S7 includes a four-level quantitative grading warning mechanism: blue, yellow, orange, and red, with a unified and fixed judgment threshold. A blue alert corresponds to a long-distance, weak lightning event where the vertical distance from the lightning strike point to the line is greater than 200m, there are fewer than 3 effective calculation stations, and the signal amplitude is 1.5 to 2.5 times the background noise. A yellow alert requires at least 3 valid solution sites, a location confidence level of at least 0.75, a lightning strike distance between the tower and the line corridor between 80m and 200m, a signal amplitude of 2.5 to 3.5 times the background noise, and a rising edge slope of less than 0.8mV / μs. The criteria for issuing an orange alert are: the distance from the lightning strike to the tower is no more than 80m, the distance from the outer boundary of the line corridor is no more than 50m, the rise slope is no less than 0.8mV / μs, and the solution residual is less than 30m. A red alert can be triggered if any two of the following conditions are met: a lightning strike within 30m of the tower in a high-risk backflash zone; a lightning strike within 50m of the conductor corridor outside the protection angle in a high-risk backflash zone; a signal amplitude of not less than 3.5 times the background noise; a lightning strike energy of not less than 8.5 × 10³ quantization units; at least two multi-pulse return strokes within 500ms; and a deterministic fault lightning strike event.

9. A transmission line lightning strike fault location analysis system, used to implement the transmission line lightning strike fault location analysis method according to any one of claims 1-8, characterized in that, include: The front-end monitoring device is deployed along the power transmission line to collect lightning pulse signals, complete BeiDou high-precision time synchronization, buffer and store signals locally, and transmit lightning signals to the outside through the communication transmission network. The communication transmission network is used to enable encrypted data transmission and remote configuration command issuance between the front-end monitoring device and the back-end service platform; The back-end service platform is used to receive lightning strike signals uploaded by the front-end monitoring devices and perform decryption verification, signal quality scoring, and preprocessing.

10. The transmission line lightning strike fault location and analysis system according to claim 9, characterized in that: The back-end service platform is a server cluster equipped with positioning and calculation software, geographic information system, early warning push system and database. It is used to receive lightning strike signals uploaded by front-end monitoring devices and perform decryption verification, signal quality scoring and preprocessing.

Images