A method and system for digital monitoring of a surge arrester

Abstract

The application discloses a lightning arrester digital monitoring method and system, comprising: acquiring each column resistance sheet data in real time through a sensor array arranged on the lightning arrester; constructing a current prediction model based on a long short-term memory network, and outputting predicted current data in a future time window; establishing a thermal-electric coupling simulation model of the multi-column lightning arrester, and calculating and acquiring the energy absorption density and the predicted temperature distribution field of each column resistance sheet in the future time window; according to the energy absorption density and the predicted temperature distribution field, performing regional risk assessment on each column resistance sheet, and identifying regions of different risk levels; generating a differentiated early warning signal, and triggering an active regulation strategy to regulate the lightning arrester. The application has the advantages that: through real-time data acquisition by the sensor, combined with LSTM current prediction and thermal-electric coupling simulation, the risk can be accurately assessed and the early warning signal can be generated, active regulation is realized, and the safety and reliability of the lightning arrester are significantly improved.

Description

TECHNICAL FIELD

[0001] The present application relates to digital monitoring, in particular to a lightning arrester digital monitoring method and system. BACKGROUND

[0002] Lightning arresters are important components in power systems that protect equipment from lightning strikes or surge impacts, widely used in power, communication, transportation and other fields. Its main function is to ensure the safe operation of power equipment by leading lightning current or overvoltage, avoiding damage or failure of equipment due to overvoltage. Lightning arresters are diverse, including gas lightning arresters, solid lightning arresters and zinc oxide lightning arresters, etc., different types of lightning arresters differ according to their working principle and application environment.

[0003] Current lightning arrester monitoring methods and systems on the market rely on simple current detection or temperature monitoring, lack comprehensive analysis of current waveform, temperature distribution and environmental factors, resulting in a lag in predicting potential risks. In addition, many existing systems do not fully utilize advanced machine learning and simulation technology, making it difficult to accurately predict future current changes and temperature distribution, and capture complex thermal-electric coupling effects. Temperature monitoring in existing methods is usually based on local data only, lacking dynamic tracking of global temperature field, and unable to provide detailed risk assessment. Risk assessment is often based on static standards, unable to adapt to changing working environments and operating conditions in a timely manner, making it difficult to respond effectively in emergency situations. Finally, most existing methods lack intelligent control capabilities, unable to actively adjust device operating parameters based on real-time monitoring data, reducing the risk of lightning arresters. SUMMARY

[0004] To improve existing methods and systems, a lightning arrester digital monitoring method and system is provided, which collects data in real time through sensors, combines LSTM current prediction and thermal-electric coupling simulation, accurately assesses risks and generates warning signals, and actively controls to significantly improve the safety and reliability of lightning arresters.

[0005] To achieve the above purpose, the technical solution adopted by the present application is:

[0006] A lightning arrester digital monitoring method, comprising:

[0007] Through the sensor array deployed on the lightning arrester, real-time acquisition of each column resistance sheet current full waveform signal data, core axial temperature distribution data and environmental temperature and humidity data is performed;

[0008] A current prediction model based on long short-term memory network is constructed, using historical current full waveform data, temperature distribution data and environmental temperature and humidity data as the training set, and outputting predicted current data in the future time window;

[0009] Based on the axial temperature distribution of each column resistance sheet at the current moment and the predicted current data, a thermal-electric coupling simulation model of the multi-column lightning arrester is established, and the energy absorption density and the predicted temperature distribution field of each column resistance sheet in the future time window are calculated and obtained;

[0010] According to the energy absorption density and the predicted temperature distribution field, the regional risk assessment of each column resistance sheet is carried out, and the regions of different risk levels are identified;

[0011] Based on the identification result, a differentiated early warning signal is generated, and an active regulation strategy is triggered to regulate the lightning arrester.

[0012] Preferably, the current prediction model based on the long short-term memory network is constructed, the historical current full waveform data, the temperature distribution data and the environmental temperature and humidity data are used as the training set, and the predicted current data in the future time window is output, which specifically includes:

[0013] Based on the historical current full waveform data, the temperature distribution data and the environmental temperature and humidity data, the noise is removed and the missing values are filled;

[0014] Based on the preprocessed data, the time scale is unified, the time series data is generated, and the characteristic values are extracted to obtain the peak value, the mean value, the variance and the amplitude data of the current;

[0015] The data is divided into a training set, a validation set and a test set, and the training set is input into the model for training;

[0016] The long-term dependence relationship in the time series data is captured based on the LSTM layer of the model;

[0017] Based on the trained model, real-time acquisition data is input into the model, and the predicted current data in the future time window is output.

[0018] Preferably, the thermal-electric coupling simulation model of the multi-column lightning arrester is established based on the axial temperature distribution of each column resistance sheet at the current moment and the predicted current data, and the energy absorption density and the predicted temperature distribution field of each column resistance sheet in the future time window are calculated and obtained, which specifically includes:

[0019] Based on the current heat equation and the heat conduction equation, a thermal-electric coupling simulation model is constructed, and real-time acquisition data is used as the initial condition of the simulation model;

[0020] The heat conduction equation is solved by the finite difference method, and the temperature change of the resistance sheet at each position of the same column at each moment is calculated combined with the current data;

[0021] The energy absorption density of each resistance sheet is calculated by calculating the heat source intensity of each resistance sheet at each time step;

[0022] The temperature of each resistance piece is finally obtained by the same thermal-electric coupling simulation of all the resistance pieces, and the temperature distribution field of the multi-column lightning arrester is formed.

[0023] The temperature of each resistance piece is updated based on each time step, and the predicted temperature distribution field of the dynamic multi-column lightning arrester is obtained in combination with the temperature of adjacent resistance pieces.

[0024] Preferably, the temperature change of each resistance piece at each time is obtained by solving the heat conduction equation by the finite difference method in combination with current data, and specifically includes:

[0025] The resistance piece is discretized into a plurality of small units along the axial space, and each position is defined, and the time is discretized into a plurality of small steps;

[0026] Based on the initial condition, the initial temperature of each resistance piece position is set;

[0027] At each time step, the Joule heat of each position is calculated according to the predicted current data, and substituted into the discretized heat conduction equation;

[0028] Based on the discretized heat conduction equation, the temperature of each position is calculated, and the update of the temperature field at both ends is processed according to the boundary condition;

[0029] The above process is repeated step by step with the time advancing until a predetermined prediction period is reached, and the temperature of each resistance piece at each position of the same column at each time is output.

[0030] Preferably, the regional risk assessment of each column resistance piece according to the energy absorption density and the predicted temperature distribution field, and the identification of regions of different risk levels specifically includes:

[0031] Based on the material properties and design standards of the resistance piece, the risk level at different temperatures is determined, and the threshold of the energy absorption density is set according to the heat capacity and heat dissipation capacity of the resistance piece;

[0032] Based on the spatial position of the resistance piece, each resistance piece is divided into different regions;

[0033] By weighting and merging the predicted temperature and the energy absorption density, the comprehensive risk index of each region is calculated and obtained;

[0034] Based on the obtained comprehensive risk index, the risk of each region is classified, and the regions of each risk level are identified and obtained.

[0035] Preferably, based on the identification result, a differentiated early warning signal is generated, and an active regulation strategy is triggered to regulate the lightning arrester.

[0036] Based on the identification result, a differentiated early warning signal is generated, including three-level early warning signals of low risk, medium risk and high risk.

[0037] Triggering the active regulation strategy, by adjusting the current sharing impedance of the valve plate between the parallel columns, the current distribution ratio tends to 1, if the single column risk level does not decrease for 3 prediction periods, the column is forced to exit operation;

[0038] Based on the feedback of the early warning signal and the regulation strategy, the existing regulation algorithm is optimized, and the risk assessment standard is calibrated regularly.

[0039] Further, a digital monitoring system for surge arresters is proposed, comprising:

[0040] The data acquisition module acquires real-time current full waveform signal data of each column resistance sheet, axial temperature distribution data of the core body and environmental temperature and humidity data through the deployed sensor array;

[0041] The current prediction module constructs a current prediction model based on historical data using LSTM, and outputs predicted current data in the future time window;

[0042] The thermal-electric coupling simulation module constructs a multi-column surge arrester thermal-electric coupling simulation model based on the axial temperature distribution and predicted current data at the current time, and calculates the energy absorption density and temperature distribution field of each column resistance sheet;

[0043] The temperature change module solves the heat conduction equation by finite difference method, calculates the temperature change of the resistance sheet at each time combined with the current data, and updates the dynamic temperature distribution of each column resistance sheet;

[0044] The risk assessment module performs regional risk assessment on each resistance sheet according to the energy absorption density and predicted temperature distribution field, and identifies regions of different risk levels;

[0045] The early warning and regulation module generates differentiated early warning signals based on the risk assessment results, and triggers the active regulation strategy to optimize the operation state of the surge arrester to reduce the risk;

[0046] The processor is used to process the calculation process of each formula and the construction and calculation process of each model.

[0047] Compared with the prior art, the advantages of the present application are:

[0048] By integrating multiple advanced technologies, a comprehensive and accurate real-time monitoring and risk assessment solution is provided. First, by deploying a sensor array, real-time acquisition of current waveform, temperature distribution and environmental data of the lightning arrester is realized, providing high-quality data support for prediction and simulation models. The current prediction model constructed by using long short-term memory network (LSTM) can accurately capture long-term dependencies and provide reliable prediction for future current trends. In addition, combined with the thermal-electric coupling simulation model, the internal temperature change and energy absorption of the lightning arrester are accurately calculated, further improving the prediction accuracy. Through regionalized risk assessment, different risk level areas can be identified, timely warning and triggering differentiated regulation strategies can effectively reduce the risk of failure. BRIEF DESCRIPTION OF DRAWINGS

[0049] Figure 1 The method proposed by the present application is shown in the schematic diagram;

[0050] Figure 2 The predicted current data proposed by the present application is shown in the schematic diagram;

[0051] Figure 3 The thermal-electric coupling simulation model proposed by the present application is shown in the schematic diagram;

[0052] Figure 4 The solution of heat conduction equation proposed by the present application is shown in the schematic diagram;

[0053] Figure 5 The risk assessment proposed by the present application is shown in the schematic diagram;

[0054] Figure 6 The early warning and regulation proposed by the present application is shown in the schematic diagram. DETAILED DESCRIPTION

[0055] The following description is used to disclose the present application so that those skilled in the art can implement the present application. The preferred embodiments in the following description are only as examples, and other obvious modifications can be made by those skilled in the art.

[0056] A lightning arrester digital monitoring system comprises:

[0057] Data acquisition module: the data acquisition module acquires real-time current full waveform signal data of each column resistance sheet, axial temperature distribution data of the core body and environmental temperature and humidity data through the deployed sensor array;

[0058] Current prediction module: the current prediction module constructs a current prediction model based on historical data using LSTM, and outputs predicted current data in the future time window;

[0059] A thermal-electric coupling simulation module: based on the axial temperature distribution of the current moment and the predicted current data, the thermal-electric coupling simulation module constructs a multi-column arrester thermal-electric coupling simulation model, and calculates the energy absorption density and temperature distribution field of each column of resistance sheet;

[0060] A temperature change module: the temperature change module solves the heat conduction equation by finite difference method, calculates the temperature change of the resistance sheet at each moment combined with the current data, and updates the dynamic temperature distribution of each column of resistance sheet;

[0061] A risk assessment module: the risk assessment module performs regionalized risk assessment on each resistance sheet according to the energy absorption density and the predicted temperature distribution field, and identifies regions of different risk levels;

[0062] A pre-warning control module: the pre-warning control module generates differentiated pre-warning signals based on the risk assessment results, and triggers active control strategies to optimize the operation state of the arrester to reduce risks;

[0063] A processor: the processor is used to process the calculation process of each formula and the construction and calculation process of each model.

[0064] Referring to Figure 1 As shown in the figure, a digital monitoring method of arrester includes:

[0065] Step one: through the sensor array deployed on the arrester, real-time acquisition of each column of resistance sheet current full waveform signal data, core axial temperature distribution data and environmental temperature and humidity data is realized;

[0066] Step two: a current prediction model based on long short-term memory network is constructed, and historical current full waveform data, temperature distribution data and environmental temperature and humidity data are used as a training set to output predicted current data in a future time window;

[0067] Step three: based on the axial temperature distribution of each column of resistance sheet at the current moment and the predicted current data, a multi-column arrester thermal-electric coupling simulation model is established, and the energy absorption density and predicted temperature distribution field of each column of resistance sheet in the future time window are calculated and obtained;

[0068] Step four: according to the energy absorption density and the predicted temperature distribution field, regionalized risk assessment is performed on each column of resistance sheet, and regions of different risk levels are identified;

[0069] Step five: based on the identification results, differentiated pre-warning signals are generated, and active control strategies are triggered to control the arrester.

[0070] Referring to Figure 2As shown, a current prediction model based on a long short-term memory network is constructed, historical current full waveform data, temperature distribution data and environmental temperature and humidity data are used as a training set, and predicted current data in a future time window is output. Specifically, it includes:

[0071] Based on historical current full waveform data, temperature distribution data and environmental temperature and humidity data, noise is removed and missing values are filled in;

[0072] Based on the preprocessed data, the time scale is unified, the time series data is generated, and the characteristic values are extracted to obtain the peak value, mean value, variance, and amplitude data of the current;

[0073] The data is divided into a training set, a validation set and a test set, and the training set is input into the model for training;

[0074] The LSTM layer of the model captures the long-term dependence in the time series data;

[0075] Based on the trained model, real-time collected data is input into the model, and predicted current data in a future time window is output.

[0076] Specifically, all data are aligned to the same time point, and for historical current data, temperature data, and environmental temperature and humidity data, resampling is performed to unify to the same time step;

[0077] Significant features are extracted from the time series data, including the peak value, mean value, variance and amplitude of the current, the peak value of the current sequence is the maximum value in the sequence, the mean value of the current sequence is the average of all data points in the sequence, the variance of the current sequence reflects the dispersion degree of the data, and the amplitude is the difference between the maximum and minimum values of the current sequence;

[0078] The data is divided into a training set, a validation set and a test set, wherein 80% of the data is used as a training set, 10% of the data is used as a validation set, and 10% of the data is used as a test set.

[0079] Referring to Figure 3 As shown, based on the axial temperature distribution of each column resistance sheet at the current moment and the predicted current data, a multi-column lightning arrester thermal-electric coupling simulation model is established, and the energy absorption density and predicted temperature distribution field of each column resistance sheet in a future time window are calculated. Specifically, it includes:

[0080] Based on the current heat equation and the heat conduction equation, a thermal-electric coupling simulation model is constructed, and real-time collected data is used as the initial condition of the simulation model;

[0081] The heat conduction equation is solved by finite difference method, and the temperature change of the resistance sheet at each position of the same column at each moment is calculated combined with the current data;

[0082] The energy absorption density of each resistance disc is calculated by calculating the heat source intensity of each resistance disc at each time step;

[0083] The temperature distribution field of the multi-column lightning arrester is formed by performing the same thermal-electric coupling simulation on all resistance discs to obtain the temperature of each resistance disc;

[0084] The predicted temperature distribution field of the dynamic multi-column lightning arrester is obtained by updating the temperature of each resistance disc at each time step and combining the temperature of adjacent resistance discs.

[0085] Specifically, when the current passes through the resistance disc, the Joule heat generated by the resistance can be calculated by the following formula:

[0086]

[0087] wherein, is the heat generated by the resistance disc, is the predicted current, is the resistance of the resistance disc;

[0088] The heat conduction equation describes the process of heat propagation in materials. For a three-dimensional problem, the heat conduction equation can be written as:

[0089]

[0090] wherein, is the temperature of the resistance disc at position x and time t, is the thermal diffusion coefficient, is the heat source, is the specific heat capacity of the resistance disc;

[0091] The finite difference method is used to numerically solve the heat conduction equation, the heat source intensity is determined by the power generated by the current, and the energy absorption density represents the energy absorbed by each resistance disc at each time step, and the calculation formula is:

[0092]

[0093] wherein, is the energy absorption density of the resistance disc at position x and time t, is the heat source, is the volume of the resistance disc;

[0094] The temperature distribution field of the entire lightning arrester is formed by performing the same thermal-electric coupling simulation on all resistance discs of the multi-column lightning arrester to obtain the temperature of each resistance disc;

[0095] At each time step, the temperature of the resistance sheet is updated, considering the temperature conduction effect of adjacent resistance sheets, the temperature exchange between the i-th resistance sheet and its adjacent resistance sheets is carried out through heat conduction, the heat flow between adjacent resistance sheets is calculated through the heat conduction formula, and the formula is:

[0096]

[0097] wherein, is the heat flow between the resistance sheets i and i+1, is the thermal conductivity, is the distance between adjacent resistance sheets, , is the temperature of the resistance sheet i and i+1;

[0098] Through the temperature update of all resistance sheets, the dynamic temperature distribution field of the entire multi-column lightning arrester is finally obtained.

[0099] Referring to Figure 4 , the temperature change of each position of the resistance sheet at each moment is calculated by solving the heat conduction equation by the finite difference method combined with current data, which specifically includes:

[0100] The resistance sheet is discretized into a plurality of small units along the axial space, and each position is defined, and the time is discretized into a plurality of small steps;

[0101] Based on the initial condition, the initial temperature of each resistance sheet position is set;

[0102] At each time step, the Joule heat of each position is calculated according to the predicted current data, and is substituted into the discretized heat conduction equation;

[0103] Based on the discretized heat conduction equation, the temperature of each position is calculated, and the temperature field at both ends is updated according to the boundary condition;

[0104] The above process is repeated step by step until the predetermined prediction period is reached, and the temperature of each position of the resistance sheet at each moment is output.

[0105] Specifically, the heat conduction equation is discretized in space and time to obtain the discrete formula of the finite difference method, and the formula is:

[0106]

[0107] wherein, is the temperature of position x at the n-th time step, is the heat source intensity of the i-th position of the resistance sheet, is the thermal diffusivity, is the specific heat capacity of the resistance sheet;

[0108] The temperature update formula is obtained by arranging the above discretization formula:

[0109]

[0110] At each time step n, according to the temperature distribution and heat source distribution at the current time n, the temperature distribution at the next time step n+1 is calculated;

[0111] Through continuous iteration calculation, the temperature distribution at each time step is obtained, and a complete temperature field is formed.

[0112] Referring to Figure 5 As shown, according to the energy absorption density and the predicted temperature distribution field, the risk assessment of each column resistance sheet is regionalized, and the regions of different risk levels are identified, which specifically includes:

[0113] Based on the material properties and design standards of the resistance sheet, the risk level at different temperatures is determined, and according to the heat capacity and heat dissipation capacity of the resistance sheet, the threshold value of the energy absorption density is set;

[0114] Based on the spatial position of the resistance sheet, each resistance sheet is divided into different regions;

[0115] By weighting and merging the predicted temperature and the energy absorption density, the comprehensive risk index of each region is calculated and obtained;

[0116] Based on the obtained comprehensive risk index, the risk classification of each region is carried out, and the regions of each risk level are identified.

[0117] Specifically, according to the material properties and design standards of the resistance sheet, the risk level at different temperatures is defined. The risk level here is mainly set according to the maximum working temperature of the resistance sheet and the thermal stability of the material;

[0118] The temperature and energy absorption density of each region are weighted and merged to obtain the comprehensive risk index of each region. Through the comprehensive risk index, the risk classification of each region is carried out, which is divided into low risk area, medium risk area and high risk area. According to the classification result of the comprehensive risk index, the risk level of each region is identified.

[0119] Referring to Figure 6 Based on the identification result, a differentiated early warning signal is generated, and an active regulation strategy is triggered to regulate the lightning arrester, which specifically includes:

[0120] Based on the identification result, a differentiated early warning signal is generated, including low risk, medium risk and high risk three-level early warning signals;

[0121] Trigger the active regulation strategy to adjust the current sharing impedance of the parallel column valve sheet, so that the current distribution ratio tends to 1. If the single column risk level does not decrease for 3 consecutive prediction periods, the column is forced to exit operation.

[0122] Based on the feedback of the early warning signal and the regulation strategy, the existing regulation algorithm is optimized, and the risk assessment standard is calibrated regularly.

[0123] Specifically, according to the comprehensive risk assessment index of the high-risk area, the early warning signal is divided into different levels, low risk: the temperature and energy absorption density of the resistance sheet are within the normal range, the risk can be ignored, medium risk: the temperature of the resistance sheet is close to the safety threshold, or the energy absorption density increases, there is a potential risk of overheating, high risk: the temperature or energy absorption density of the resistance sheet exceeds the set threshold, there is a possibility of immediate failure, emergency measures need to be taken;

[0124] When the comprehensive risk index of a certain resistance sheet or area exceeds the preset threshold, the corresponding early warning signal is triggered, and according to the identification of the high-risk area and the early warning signal, the active regulation strategy will take different regulation measures based on different risk levels to ensure the safe operation of the lightning arrester. The regulation of low-risk areas includes no immediate intervention, continues normal operation, but needs regular inspection and monitoring. Regular inspection is carried out on low-risk areas within the planned maintenance period to ensure their continued stability.

[0125] The regulation of medium-risk areas includes reducing the load of the resistance sheet or reducing the working current to reduce the heat generated, avoiding further temperature rise, reducing the current through the resistance sheet to reduce the generation of Joule heat, redistributing the load from the medium-risk area to other low-risk areas to reduce heat accumulation, and enhancing cooling measures if the resistance sheet temperature is close to the safety threshold, such as improving air circulation, increasing ventilation, and starting the standby cooling system.

[0126] The regulation of high-risk areas includes immediately disabling the high-risk area: if the risk level of a certain area is high, the disabling program is triggered immediately to cut off the current to prevent failure, quickly cut off the current to stop the current through the high-risk resistance sheet to prevent overheating from further causing equipment damage, and immediately start the safety mechanisms such as automatic power-off protection and overheating protection to ensure safety. The cooling system in the high-risk area needs to be started as soon as possible to quickly reduce the temperature of the resistance sheet. When a high-risk area appears, maintenance personnel are immediately arranged to conduct detailed inspection on site to determine the cause of the failure and take temporary repair measures.

[0127] It should be noted that the above-mentioned order of the embodiments of the present application is only for description, and does not represent the advantages and disadvantages of the embodiments. The above describes specific embodiments of the present application. In addition, the processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multi-task processing and parallel processing are possible or may be advantageous.

[0128] The various embodiments in the specification are described in progressive manner, and the same or similar parts between the various embodiments can be mutually referred to, and each embodiment focuses on the difference from other embodiments.

[0129] The above description is merely preferred embodiments of the present application, and is not intended to limit the present application. Any modification, equivalent replacement, improvement, etc. made within the principles of the present application shall be included in the protection scope of the present application.

Claims

1. A method for digital monitoring of a surge arrester, characterized in that, The application relates to a multi-column lightning arrester risk assessment method and device. Real-time acquisition of full-wave signal data of each column resistance sheet current, axial temperature distribution data of the core body and environmental temperature and humidity data through a sensor array arranged on the lightning arrester; A current prediction model based on a long short-term memory network is constructed, historical full-wave current data, temperature distribution data and environmental temperature and humidity data are used as a training set, and predicted current data in a future time window is output; Based on the axial temperature distribution of each column resistance sheet at the current moment and the predicted current data, a thermal-electric coupling simulation model of the multi-column lightning arrester is established, the energy absorption density and the predicted temperature distribution field of each column resistance sheet in the future time window are calculated and acquired; According to the energy absorption density and the predicted temperature distribution field, the regional risk of each column resistance sheet is assessed, and regions with different risk levels are identified; Based on the identification result, a differential early warning signal is generated, and an active regulation strategy is triggered to regulate the lightning arrester. The method specifically comprises the following steps: Based on the material properties and design standards of the resistance sheet, the risk level at different temperatures is determined, and the threshold value of the energy absorption density is set according to the heat capacity and heat dissipation capacity of the resistance sheet; Based on the spatial position of the resistance sheet, each resistance sheet is divided into different regions; By weighting and merging the predicted temperature and the energy absorption density, the comprehensive risk index of each region is calculated and acquired; Based on the acquired comprehensive risk index, the risk of each region is classified, and the regions with different risk levels are identified.

2. The method of claim 1, wherein the method comprises: The method specifically comprises the following steps: Based on the historical full-wave current data, temperature distribution data and environmental temperature and humidity data, noise is removed and missing values are filled; Based on the preprocessed data, time scale unification is performed to generate time series data, and characteristic values are extracted to obtain peak value, mean value, variance and amplitude data of the current; The data is divided into a training set, a validation set and a test set, and the training set is input into the model for training; The LSTM layer of the model captures the long-term dependence relationship in the time series data; Based on the trained model, real-time acquisition data is input into the model, and predicted current data in a future time window is output.

3. The method of claim 1, wherein the method comprises: The method specifically comprises the following steps: Based on the current time, the axial temperature distribution of each column resistance sheet and the predicted current data, a thermal-electric coupling simulation model of the multi-column lightning arrester is established, the energy absorption density and the predicted temperature distribution field of each column resistance sheet in the future time window are calculated and acquired. Based on the current time, the axial temperature distribution of each column resistance sheet and the predicted current data, a thermal-electric coupling simulation model of the multi-column lightning arrester is established, the energy absorption density and the predicted temperature distribution field of each column resistance sheet in the future time window are calculated and acquired. The thermal-electric coupling simulation model is constructed based on a current heat equation and a heat conduction equation, and the real-time acquisition data is used as an initial condition of the simulation model; The heat conduction equation is solved by the finite difference method, and the temperature change of each resistance sheet at each position of the same column is calculated and acquired in combination with the current data; The heat source intensity of each resistance sheet at each time step is calculated, and the energy absorption density of each resistance sheet is calculated and acquired. The temperature of each resistance piece is finally obtained by simulating the same thermal-electric coupling of all column resistance pieces, and the temperature distribution field of the multi-column lightning arrester is formed. The temperature of each resistance piece is updated based on each time step, and the predicted temperature distribution field of the dynamic multi-column lightning arrester is obtained by combining the temperatures of adjacent resistance pieces.

4. The method of claim 3, wherein the method comprises: The temperature change of each resistance piece at each time is calculated by solving the heat conduction equation by the finite difference method and combining the current data, and the dynamic temperature distribution of each resistance piece is updated. The resistance piece is discretized into a plurality of small units along the axial space, and each position is defined, and the time is discretized into a plurality of small steps; Based on the initial condition, the initial temperature of each resistance piece position is set; At each time step, the Joule heat of each position is calculated according to the predicted current data, and substituted into the discretized heat conduction equation; Based on the discretized heat conduction equation, the temperature of each position is calculated, and the temperature field at both ends is updated according to the boundary condition; The time is gradually advanced, and the above process is repeated until the predetermined prediction period is reached, and the temperature of each resistance piece at each position of the same column at each time is output.

5. The method of claim 1, wherein the method comprises: The difference warning signal is generated based on the identification result, and the active regulation strategy is triggered to regulate the lightning arrester, which specifically includes: Based on the identification result, a difference warning signal is generated, including low-risk, medium-risk and high-risk three-level warning signals; Triggering the active regulation strategy, adjusting the current sharing impedance of the valve plate between the parallel columns, making the current distribution ratio tend to 1, and if the single-column risk level does not decrease for 3 prediction periods, the column is forced to exit operation; Based on the feedback of the warning signal and the regulation strategy, the existing regulation algorithm is optimized, and the risk assessment standard is calibrated regularly.

6. A digital monitoring system for a surge arrester for implementing a method according to any one of claims 1 to 5, characterized in that It includes: The data acquisition module acquires real-time current full waveform signal data, core axial temperature distribution data and environmental temperature and humidity data of each column resistance piece through the deployed sensor array; The current prediction module constructs a current prediction model based on historical data using LSTM to output predicted current data in a future time window; The thermal-electric coupling simulation module constructs a multi-column lightning arrester thermal-electric coupling simulation model based on the current axial temperature distribution and predicted current data, calculates the energy absorption density and temperature distribution field of each column resistance piece; The temperature change module solves the heat conduction equation by the finite difference method, calculates the temperature change of the resistance piece at each time by combining the current data, and updates the dynamic temperature distribution of each column resistance piece; The risk assessment module regionalizes the risk assessment of each resistance piece according to the energy absorption density and the predicted temperature distribution field, and identifies regions of different risk levels; The warning and regulation module generates a difference warning signal based on the risk assessment result, and triggers the active regulation strategy to optimize the operation state of the lightning arrester to reduce the risk; The processor is used to process the calculation process of each formula and the construction and calculation process of each model.

Images