A method for predicting electrical parameters of a lightning arrester

Abstract

The present invention belongs to the technical field of electrical parameter prediction, and particularly relates to a method for predicting electrical parameters of a lightning arrester. The electrical parameters of the lightning arrester are collected and stored according to the time stamp; a hierarchical hybrid prediction model is constructed to predict the electrical parameters of the lightning arrester, including: decomposing the electrical parameters into a horizontal component, a trend component and a seasonal component, constructing an electrical parameter prediction equation to perform a preliminary prediction on the electrical parameters to obtain an electrical parameter prediction value; calculating the difference between the true value of the electrical parameters and the electrical parameter prediction value, inputting the difference as a residual sequence into a lightweight time series convolutional network to extract residual non-linear compensation features, calculating the residual variance and dynamically adjusting the convolutional kernel weights, and outputting an electrical parameter correction value after compressing the residual non-linear compensation features through a fully connected layer, and finally correcting the electrical parameter prediction value.

Description

Technical Field

[0001] The present invention belongs to the technical field of electrical parameter prediction, and particularly relates to a method for predicting electrical parameters of lightning arresters. Background Art

[0002] In the power system, lightning arresters, as key protection devices, are crucial for protecting grid equipment from lightning overvoltage and switching overvoltage. Their performance directly affects the safe and stable operation of the power grid. However, since lightning arresters are mostly placed in the open air, during long-term operation, they will inevitably be affected by the combined effects of power frequency voltage and complex and harsh natural environments, resulting in problems such as aging and moisture absorption of the lightning arrester valve discs, thereby reducing the protection performance of the lightning arresters. In severe cases, it may even cause large-scale power outages in the power grid.

[0003] To effectively evaluate the health status of lightning arresters, timely detect potential fault hazards, and provide a scientific basis for subsequent condition-based maintenance, currently, it is mainly achieved by real-time monitoring of key electrical parameters of lightning arresters, such as total current, resistive current, lightning strike times, and lightning strike moments. By collecting and analyzing these electrical parameters, the current health status of the lightning arrester can be evaluated, usually presented in the form of a health status index.

[0004] The electrical parameters of lightning arresters are affected by various environmental factors such as seasonal changes and load cycles, showing complex non-linear trends and seasonal fluctuation characteristics. However, due to the lack of the ability to analyze these key factors, traditional prediction methods are difficult to accurately capture these change laws of electrical parameters and cannot give early warnings of abnormal changes in key electrical parameters. Summary of the Invention

[0005] In order to overcome the problems in the prior art, the present invention proposes a method for predicting electrical parameters of lightning arresters.

[0006] The technical solution of the present invention to solve the above technical problems is as follows:

[0007] The present invention provides a method for predicting electrical parameters of lightning arresters, including the following steps:

[0008] Collect the electrical parameters of the lightning arrester and store the electrical parameters of the lightning arrester according to the time stamp;

[0009] Construct a hierarchical hybrid prediction model to predict the electrical parameters of lightning arresters, including: decomposing the electrical parameters into a horizontal component, a trend component, and a seasonal component, constructing an electrical parameter prediction equation, making a preliminary prediction of the electrical parameters to obtain an electrical parameter prediction value; calculating the difference between the true value of the electrical parameters and the predicted value of the electrical parameters, inputting the difference as a residual sequence into a lightweight temporal convolutional network, extracting residual non-linear compensation features, calculating the residual variance and dynamically adjusting the convolutional kernel weights, outputting an electrical parameter correction value after compressing the residual non-linear compensation features through a fully connected layer, and finally correcting the electrical parameter prediction value.

[0010] Further, the decomposing the electrical parameters into a horizontal component, a trend component, and a seasonal component, and constructing an electrical parameter prediction equation includes:

[0011] Horizontal component: ;

[0012] Trend component: ;

[0013] Seasonal component: ;

[0014] Full current prediction equation: ;

[0015] In the formula, represents the electrical parameter at time t; represents the horizontal component of the electrical parameter at time t; represents the horizontal component of the electrical parameter at time t - 1; represents the trend component of the electrical parameter at time t; represents the trend component of the electrical parameter at time t - 1; represents the seasonal component of the electrical parameter at time t; represents the seasonal component of the electrical parameter at time represents the seasonal cycle length; represents the horizontal parameter, represents the damping factor, represents the trend parameter, represents the seasonal parameter; represents the prediction time step; represents the predicted value of the electrical parameter at time represents the seasonal component of the electrical parameter at time

[0016] Further, iteratively solve the optimal solution combination of the horizontal parameter, damping factor, trend parameter, and seasonal parameter through the gradient descent method.

[0017] Further, inputting the difference as a residual sequence into a lightweight temporal convolutional network and extracting residual non-linear compensation features includes:

[0018] ;

[0019] In the formula, represents the residual non - linear compensation feature of the output of the l -th layer in the lightweight temporal convolutional network at time ; represents the weight of the l -th layer and the -th convolutional kernel in the convolutional network; K represents the size of the convolutional kernel; represents the bias term;

[0020] Further, calculating the residual variance and dynamically adjusting the convolutional kernel weight includes:

[0021] Calculating the residual variance:

[0022] ;

[0023] In the above formula, represents the current residual variance; represents the size of the sliding window; represents the average value of recent residuals; represents t - i the difference between the true value and the predicted value of the electrical parameter at time

[0024] If the residual variance is greater than the preset residual variance threshold, then dynamically adjust the convolutional kernel weight in the convolutional network:

[0025] ;

[0026] In the above formula, represents the convolutional kernel weight in the adjusted lightweight temporal convolutional network; represents the historical residual variance; is the sensitivity coefficient.

[0027] Further, after compressing the residual non - linear compensation feature through the fully - connected layer, the corrected value of the electrical parameter is output:

[0028] ;

[0029] In the above formula, l is the number of convolutional layers; is the weight of the fully - connected layer; is the bias term of the fully - connected layer; represents the corrected value of the electrical parameter at time j represents the feature nodes in the fully - connected layer; Denote the residual non - linear compensation feature of the l -th layer and the j -th feature node in the light - weight temporal convolutional network at time in the convolutional output.

[0030] Furthermore, the final corrected value of the electrical parameter prediction includes: correcting the electrical parameter prediction value based on the electrical parameter correction value:

[0031] ;

[0032] In the above formula, denotes the corrected value of the electrical parameter prediction at ; denotes the electrical parameter prediction value at .

[0033] Furthermore, it also includes: real - time identifying mutation events through sliding - window outlier detection. If a mutation event is detected, reset the trend component, shorten the seasonal cycle length, and reduce the horizontal smoothing parameter to dynamically adjust the parameters of the full - current hierarchical hybrid prediction model.

[0034] Furthermore, the electrical parameters include full - current and resistive current.

[0035] Compared with the prior art, the present invention has the following technical effects:

[0036] Through the collaborative mechanism of adaptive decomposition - deep residual compensation - dynamic parameter linkage, the present invention systematically solves the triangular contradiction of accuracy, computing power, and robustness in the prediction of arrester electrical parameters: proposing a lightweight hierarchical hybrid architecture - constructing an adaptive damping model with the triple - exponential smoothing algorithm as the core, dynamically optimizing parameters by introducing a damping factor and a rolling time window, effectively suppressing trend divergence and enhancing the ability to fit temporal periodicity, while avoiding parameter solidification; on this basis, designing a lightweight temporal convolutional network (L - TCN) as residual compensation. For the non - linear spike residuals caused by transient events such as lightning strikes, low - computing - power and high - accuracy compensation are achieved through local temporal feature extraction in the convolutional layer, and the prediction error is reduced compared with traditional residual correction methods; furthermore, a real - time detection - parameter linkage rule for mutation events driven by statistics is proposed: quickly identifying abnormal conditions such as lightning strikes and short - circuits based on sliding - window outlier detection, eliminating historical trend interference through trend - component reset, dynamically compressing the seasonal cycle to strengthen short - term response, and adaptively attenuating the horizontal smoothing parameter to suppress long - term noise, so that the model improves the local sensitivity to mutation events while maintaining long - term stability, forming a "globally stable - locally agile" balance mechanism. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0038] Figure 1 It is a schematic flow chart of the present invention. Specific embodiments

[0039] To further elaborate on the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following, in combination with the drawings and preferred embodiments, will detail the specific embodiments, structures, features, and effects of the technical solutions proposed according to the present invention. The specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form. Unless otherwise defined, all technical and scientific terms used in the present invention have the same meaning as commonly understood by those skilled in the technical field to which the present invention belongs.

[0040] The present invention proposes a hierarchical hybrid prediction model, which separates the trend, seasonality, and non - linear residual features of time - series data through a hierarchical structure, combines traditional statistical models with lightweight deep learning to improve prediction accuracy and robustness; and realizes the adaptability to mutation events through a dynamic residual feedback mechanism and enhanced feature engineering. Finally, it realizes the high - precision and low - latency prediction of the electrical parameters of lightning arresters, solves the problems of the above - mentioned prior art, and provides a reliable basis for the state assessment and fault warning of lightning arresters.

[0041] Referring to Figure 1 , a method for predicting the electrical parameters of a lightning arrester includes the following steps: collecting the electrical parameters of the lightning arrester and storing the electrical parameters of the lightning arrester according to the time stamp, where the electrical parameters include the total current and the resistive current; constructing a hierarchical hybrid prediction model to predict the electrical parameters of the lightning arrester, including: decomposing the electrical parameters into a horizontal component, a trend component, and a seasonal component, constructing an electrical parameter prediction equation to perform a preliminary prediction on the electrical parameters to obtain an electrical parameter prediction value; calculating the difference between the true value of the electrical parameter and the electrical parameter prediction value, taking the difference as a residual sequence and inputting it into a lightweight temporal convolutional network to extract residual non - linear compensation features, calculating the residual variance and dynamically adjusting the convolutional kernel weights, and outputting an electrical parameter correction value after compressing the residual non - linear compensation features through a fully - connected layer, and finally correcting the electrical parameter prediction value.

[0042] The following will expand on each of the above steps in detail:

[0043] Step 100: Collect the electrical parameters of the lightning arrester and store the electrical parameters of the lightning arrester according to the time stamp.

[0044] The on-line monitoring system of the lightning arrester collects the actual value of the total current of the lightning arrester and stores the historical data set of the total current according to the time stamp.

[0045] Step 200: Construct a hierarchical hybrid prediction model to predict the electrical parameters of the lightning arrester.

[0046] Construct a hierarchical hybrid prediction model, where the hierarchical hybrid prediction model includes three layers. The first layer is the adaptive damping triple exponential smoothing, the second layer is the lightweight temporal convolutional network, and the third layer is the dynamic perception of mutation events.

[0047] Introduce the level parameter, damping factor, trend parameter, and seasonal parameter, decompose the electrical parameter into the level component, trend component, and seasonal component; and construct a prediction equation based on the level component, trend component, and seasonal component to obtain the predicted value of the electrical parameter, including:

[0048] Level component: ;

[0049] Trend component: ;

[0050] Seasonal component: ;

[0051] Total current prediction equation: ;

[0052] In the formula, represents the electrical parameter at time t; represents the level component of the electrical parameter at time t; represents the level component of the electrical parameter at time t-1; represents the trend component of the electrical parameter at time t; represents the trend component of the electrical parameter at time t-1; represents the seasonal component of the electrical parameter at time t; represents the seasonal component of the electrical parameter at time represents the seasonal cycle length; represents the level parameter, represents the damping factor, represents the trend parameter, represents the seasonal parameter; represents the prediction time step; represents the predicted value of the electrical parameter at time represents the seasonal component of the electrical parameter at time. Iteratively solve the optimal solution combination of the level parameter, damping factor, trend parameter, and seasonal parameter by the gradient descent method.

[0053] Take the difference as the residual sequence and input it into the lightweight temporal convolutional network to extract the residual non-linear compensation feature, including:

[0054] ;

[0055] In the formula, represents the residual non - linear compensation feature of the output of the l -th layer in the lightweight temporal convolutional network at time ; represents the weight of the l -th convolutional kernel in the k -th layer of the convolutional network; K represents the convolutional kernel size; represents the bias term; represents the difference between the true value and the predicted value of the electrical parameter at time t.

[0056] The calculation of the residual variance and the dynamic adjustment of the convolutional kernel weight include:

[0057] Calculating the residual variance:

[0058] ;

[0059] In the above formula, represents the current residual variance; represents the size of the sliding window; represents the average value of the recent residuals; represents t - i the difference between the true value and the predicted value of the electrical parameter at time

[0060] If the residual variance is greater than the preset residual variance threshold, then dynamically adjust the convolutional kernel weight in the convolutional network:

[0061] ;

[0062] In the above formula, represents the convolutional kernel weight in the adjusted lightweight temporal convolutional network; represents the historical residual variance; is the sensitivity coefficient.

[0063] Output the corrected value of the electrical parameter after compressing the residual non - linear compensation feature through the fully - connected layer:

[0064] ;

[0065] In the above formula, l is the number of convolutional layers; is the weight of the fully - connected layer; is the bias term of the fully - connected layer; represents the corrected value of the electrical parameter at time j represents the feature node in the fully - connected layer; Indicates the residual non - linear compensation feature of the l layer and the j feature node in the temporal convolutional output.

[0066] Finally, correct the predicted value of the electrical parameter, including: correcting the predicted value of the electrical parameter based on the corrected value of the electrical parameter:

[0067] ;

[0068] In the above formula, represents the corrected predicted value of the electrical parameter at represents the predicted value of the electrical parameter at

[0069] Real - time identify mutation events through sliding - window outlier detection. If a mutation event is detected, reset the trend component, shorten the seasonal cycle length, and reduce the horizontal smoothing parameter to dynamically adjust the parameters of the full - current hierarchical hybrid prediction model.

[0070] Taking the full - current as an example, this step 200 includes:

[0071] Construct a full - current hierarchical hybrid prediction model, which includes three layers. The first layer is adaptive damped triple exponential smoothing, the second layer is a lightweight temporal convolutional network, and the third layer is dynamic mutation - event perception.

[0072] Step 210: Decompose the full - current into a horizontal component, a trend component, and a seasonal component, construct a full - current prediction equation, make a preliminary prediction of the full - current, and obtain the full - current predicted value; calculate the difference between the true value of the full - current and the full - current predicted value.

[0073] The step 210 may include the following sub - steps:

[0074] Step 2101: Introduce a horizontal parameter, a damping factor, a trend parameter, and a seasonal parameter, decompose the full - current into a full - current horizontal component, a full - current trend component, and a full - current seasonal component, and construct a prediction equation based on the full - current horizontal component, the full - current trend component, and the full - current seasonal component to obtain the full - current predicted value.

[0075] Full - current horizontal component: ;

[0076] Full - current trend component: ;

[0077] Full - current seasonal component: ;

[0078] Full - current prediction equation: ;

[0079] In the formula, represents the total current at time t; represents the horizontal component of the total current at time t; represents the horizontal component of the total current at time t - 1; represents the trend component of the total current at time t; represents the trend component of the total current at time t - 1; represents the seasonal component of the total current at time t; represents the seasonal component of the total current at time represents the seasonal cycle length; represents the horizontal parameter, represents the damping factor, represents the trend parameter, represents the seasonal parameter; where, , , take values in the range (0, 1), take values in the range (0.9, 1); represents the prediction step size; represents the predicted value of the total current at time represents the seasonal component at time

[0080] The damping factor restricts the exponential growth of the trend and avoids overfitting. It restricts the exponential growth of the trend component, making the predicted value tend to be reasonable in the long term and solving the "trend explosion" problem of the traditional exponential smoothing model.

[0081] Step 2102: Optimize the horizontal parameter, damping factor, trend parameter, and seasonal parameter to solve the optimal solution parameter combination.

[0082] Adopt rolling time window cross - validation to minimize the mean square error (MSE):

[0083] ;

[0084] where, represents the predicted value of the total current at time t.

[0085] Iteratively solve the optimal solution , , , parameter combination through the gradient descent method.

[0086] Online update parameters through a rolling time window to adapt to model drift caused by equipment aging and environmental changes. The model can adapt to data distribution changes under different working conditions (such as humid environment, high-temperature environment). Compared with traditional fixed-parameter models, the prediction can reduce errors.

[0087] Step 220: Use the difference between the true value of the total current and the predicted value of the total current as the residual sequence and input it into the lightweight temporal convolutional network to extract residual non-linear compensation features, calculate the residual variance, and dynamically adjust the convolutional kernel weights. After compressing the features through the fully connected layer, output the corrected value of the total current, and finally correct the predicted value of the total current.

[0088] The said step 220 includes the following steps:

[0089] Step 2201: Use the difference between the true value of the total current and the predicted value of the total current as the residual sequence and input it into the lightweight temporal convolutional network (L-TCN) to perform residual feature extraction and obtain the convolutional output.

[0090] Capture the residual non-linear features after the first layer of prediction through the convolutional operation of the lightweight temporal convolutional network to correct the baseline prediction value.

[0091] Take the predicted residual sequence output by the first layer as the input: ; where represents the predicted value at time

[0092] Design causal convolution relying on historical data to ensure temporal causality:

[0093] ;

[0094] In the formula, represents the residual non-linear compensation feature of the convolutional output of the l th layer in the convolutional network at time ; is the weight of the l th layer and the k-th convolutional kernel in the convolutional network; K is the size of the convolutional kernel;

[0095] Step 2202: Calculate the residual variance and dynamically adjust the convolutional kernel weights in the lightweight temporal convolutional network according to the statistical characteristics of the residual variance.

[0096] Calculate the residual variance:

[0097] ;

[0098] In the above formula, represents the current residual variance; represents the size of the sliding window; Represents the average value of recent residuals.

[0099] If the residual variance is greater than a preset residual variance threshold, dynamically adjust the weights of the convolutional kernels in the convolutional network:

[0100] ;

[0101] In the above formula, Represents the weights of the convolutional kernels in the adjusted convolutional network; Represents the historical residual variance; Is the sensitivity coefficient.

[0102] At this time,

[0103] ;

[0104] In the formula, Represents the residual non-linear compensation feature of the convolutional output of the l th layer in the adjusted convolutional network at time .

[0105] Step 2203: Compress the residual non-linear compensation feature through a fully connected layer and output the full current correction value:

[0106] ;

[0107] In the above formula, l Is the number of convolutional layers; Is the weight of the fully connected layer; Is the bias term of the fully connected layer; Represents the full current correction value; j Represents the feature nodes in the fully connected layer.

[0108] Step 2204: Correct the full current prediction value based on the full current correction value:

[0109] ;

[0110] In the above formula, Represents The full current prediction correction value at

[0111] Step 230: Detect mutation events in real time through sliding window outlier detection. If a mutation event is detected, reset the trend component, shorten the seasonal cycle length, and reduce the horizontal smoothing parameter, and dynamically adjust the parameters of the full current hierarchical hybrid prediction model to enhance the short-term response ability.

[0112] Detect mutation events such as lightning strikes in real time based on statistical methods, and dynamically adjust the parameters of the full current hierarchical hybrid prediction model.

[0113] Step 2301: Based on the residual sequence, real-time identification of mutation events is carried out through sliding window outlier detection:

[0114] For the predicted residual sequence within the sliding window , ,…, calculate the median ;

[0115] Median absolute deviation ;

[0116] If , then mark as a mutation event, where is the threshold coefficient, usually taken as 3.

[0117] Step 2302: If a mutation event is detected, reset the trend component to avoid historical trend interference:

[0118] ;

[0119] In the above formula, represents the reset total current trend component.

[0120] Shorten the seasonal cycle length , lasting time steps to enhance the short-term response ability:

[0121] ;

[0122] In the above formula, represents the reset seasonal cycle length.

[0123] If the residuals exceed the threshold for n consecutive times, reduce the horizontal smoothing parameter :

[0124] ;

[0125] The updated horizontal smoothing parameter is passed to the first layer for prediction at the next moment.

[0126] Step 300: Electrical parameter warning and health status index.

[0127] Accurate predicted values of each electrical parameter can be obtained through Steps 100 - 200. The program sets the warning value J as needed. When the predicted value of the electrical parameter > J, the device sends a warning message and displays it to the user.

[0128] Calculate the arrester health status index (actual value and predicted value) through the comprehensive evaluation model:

[0129] ;

[0130] Among them, is the dynamic weight, which is calculated in real time by the entropy weight method and reflects the influence degree of each parameter on the health state; is the membership function, which is mapped to a score from 0 to 1 according to the degree of deviation of the parameter from the reference value.

[0131] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements on some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included in the protection scope of the present invention.

Claims

1. A method for predicting the electrical parameters of a lightning arrester, characterized in that, It includes the following steps: Collect the electrical parameters of the arrester and store the electrical parameters of the arrester according to the time stamp; Construct a hierarchical hybrid prediction model to predict the electrical parameters of the arrester, including: decomposing the electrical parameters into a horizontal component, a trend component and a seasonal component, constructing an electrical parameter prediction equation, making a preliminary prediction of the electrical parameters to obtain an electrical parameter prediction value; calculating the difference between the true value of the electrical parameter and the predicted value of the electrical parameter, inputting the difference as a residual sequence into a lightweight temporal convolutional network, extracting residual non-linear compensation features, calculating the residual variance and dynamically adjusting the convolutional kernel weights, and outputting an electrical parameter correction value after compressing the residual non-linear compensation features through a fully connected layer, and finally correcting the electrical parameter prediction value; Input the difference as a residual sequence into a lightweight temporal convolutional network to extract residual non-linear compensation features, including: ； In the formula, represents the residual non - linear compensation feature of the convolutional output at time for the l -th layer in the lightweight temporal convolutional network; represents the weight of the -th convolutional kernel in the l -th layer of the convolutional network; K represents the convolutional kernel size; k represents the bias term; represents the difference between the true value and the predicted value of the electrical parameter at time t;​ The calculating of the residual variance and dynamically adjusting the convolutional kernel weights includes: Calculating the residual variance: ； In the above formula, represents the current residual variance; represents the size of the sliding window; represents the average value of recent residuals; represents t - i the difference between the true value and the predicted value of the electrical parameter when If the residual variance is greater than a preset residual variance threshold, then dynamically adjust the convolutional kernel weights in the convolutional network: ； In the above formula, represents the convolutional kernel weights in the adjusted lightweight temporal convolutional network; represents the historical residual variance; is the sensitivity coefficient; Outputting an electrical parameter correction value after compressing the residual non-linear compensation features through a fully connected layer: ； In the above formula, l is the number of convolutional layers; is the weight of the fully connected layer; is the bias term of the fully connected layer; represents the correction value of electrical parameters at j represents the feature nodes in the fully connected layer; represents the l -th j feature node in the residual non-linear compensation feature of the convolutional output at time 2. The method for predicting the electrical parameters of a lightning arrester according to claim 1, wherein The decomposing of the electrical parameters into a horizontal component, a trend component and a seasonal component, and constructing an electrical parameter prediction equation includes: Horizontal component: ; Trend component: ; Seasonal component: ; Full current prediction equation: ; In the formula, represents the electrical parameter at time t; represents the horizontal component of the electrical parameter at time t; represents the horizontal component of the electrical parameter at time t - 1; represents the trend component of the electrical parameter at time t; represents the trend component of the electrical parameter at time t - 1; represents the seasonal component of the electrical parameter at time t; represents the seasonal component of the electrical parameter at represents the seasonal cycle length; represents the horizontal parameter, represents the damping factor, represents the trend parameter, represents the seasonal parameter; represents the prediction time step; represents the predicted value of the electrical parameter at represents the seasonal component of the electrical parameter at 3. A method for predicting electrical parameters of a lightning arrester according to claim 2, characterized in that, Iteratively solving the optimal solution combination of the horizontal parameter, damping factor, trend parameter and seasonal parameter by the gradient descent method.

4. A method for predicting electrical parameters of a lightning arrester according to claim 1, characterized in that, Finally correcting the electrical parameter prediction value includes: correcting the electrical parameter prediction value based on the electrical parameter correction value: ； In the above formula, represents the predicted correction value of electrical parameters at represents the predicted value of electrical parameters at 5. A method for predicting electrical parameters of a lightning arrester according to claim 2, characterized in that, It also includes: Real-time identify mutation events through sliding window outlier detection. If a mutation event is detected, then reset the trend component, shorten the seasonal cycle length and reduce the horizontal smoothing parameter, and dynamically adjust the parameters of the full current hierarchical hybrid prediction model.

6. The method for predicting electrical parameters of a lightning arrester according to claim 1, wherein The electrical parameters include full current and resistive current.

Images