Insulation strength evaluation method and device for hot-line work combination gap

Abstract

The invention discloses a method and a device for evaluating the insulation strength of a hot-line work combination gap. The method comprises the following steps: constructing a finite element model for a preset data set; the preset data set comprises a preset number of combined gaps with total gap lengths, the combined gaps meeting a first condition are classified as a first subset, and otherwise, the combined gaps meeting the first condition are classified as a second subset; the first condition comprises that the first gap length in the combined gap is greater than a first gap length critical value corresponding to 50% breakdown voltage; respectively acquiring parameter sets of the first subset and the second subset to obtain a first parameter set and a second parameter set; training an LSSVR model according to the first parameter set and the second parameter set to obtain a first model and a second model; acquiring a first gap length actual value and a parameter set actual value of the target combined gap; and if the actual value of the first gap length meets a first condition, performing insulation strength evaluation according to the first model and the actual value of the parameter set, otherwise, performing insulation strength evaluation according to the second model and the actual value of the parameter set.

Description

Technical Field

[0001] The present invention relates to the technical field of power system operation and maintenance, and particularly to a method and device for evaluating the insulation strength of a live working combined gap. Background Art

[0002] In modern power system operation and maintenance, live working technology has become a key measure to ensure the continuous and reliable operation of the power system. Live working allows staff to perform maintenance and repairs on high-voltage transmission lines without power interruption, which is of great significance for reducing power outage time and improving power supply reliability and economic benefits. With the continuous expansion of the power grid scale and the increasing growth of power demand, the importance of live working has become even more prominent.

[0003] The live working combined gap is a key factor in live working, and the accurate evaluation of its insulation strength is crucial for ensuring the safety of operators and the reliable operation of the power system. The live working combined gap refers to the combined air gap formed by the operator, live conductors, and towers during equipotential live working. When the operator enters and exits the equipotential area, the insulation strength of the combined gap is directly related to the operation safety. The breakdown voltage of the combined gap reflects its insulation strength. In actual operation, only by accurately evaluating the breakdown voltage and thus calculating the risk rate of live working can the sufficient safety distance between the operator and the live body be ensured to avoid the occurrence of discharge accidents. However, the traditional method for evaluating the breakdown voltage of the combined gap mainly calculates through relevant empirical formulas based on the safety distance. This method is difficult to generalize to different geometric structures and electric field distributions, cannot well adapt to various types of combined gaps, and thus is difficult to achieve the accurate evaluation of the breakdown voltage of various combined gaps.

[0004] Therefore, there is an urgent need for a new technical solution to solve the technical problem of how to accurately and efficiently evaluate the insulation strength of the live working combined gap. Summary of the Invention

[0005] The present invention provides a method and device for evaluating the insulation strength of a live working combined gap to solve the technical problem of how to accurately and efficiently evaluate the insulation strength of the live working combined gap.

[0006] To achieve the above object, the present invention provides a method for evaluating the insulation strength of a live working combined gap, including: Constructing a finite element model for a preset data set; the preset data set includes combined gaps with a preset number of total gap lengths, where the combined gaps that meet the first condition are classified into the first subset, and the combined gaps that do not meet the first condition are classified into the second subset; the first condition includes that the length of the first gap in the combined gap is greater than the critical value of the length of the first gap corresponding to 50% of the breakdown voltage.

[0007] The finite element model includes a high-voltage electrode, a ground electrode, and a floating conductor electrode; the floating conductor electrode is suspended between the high-voltage electrode and the ground electrode; the gap between the high-voltage electrode and the ground electrode is defined as the total gap, the gap between the floating conductor electrode and the high-voltage electrode is the first gap, and the gap between the floating conductor electrode and the ground electrode is the second gap.

[0008] The parameter sets of the first subset and the second subset are obtained respectively to get the first parameter set and the second parameter set; the parameter set includes the preset electric field intensity characteristic quantities of the first gap and the second gap within the combined gap; the LSSVR models are trained respectively according to the first parameter set and the second parameter set to obtain the first model and the second model.

[0009] The actual value of the first gap length of the target combined gap and the actual value of the parameter set are obtained; if the actual value of the first gap length meets the first condition, the insulation strength is evaluated according to the first model and the actual value of the parameter set, otherwise the insulation strength is evaluated according to the second model and the actual value of the parameter set.

[0010] Preferably, obtaining the parameter sets of the first subset and the second subset respectively to get the first parameter set and the second parameter set includes: The first subset and the second subset are respectively subjected to the first processing to obtain the first parameter set and the second parameter set.

[0011] The first processing includes: preset numbers of sampling points are selected at equal intervals in each of the first gap and the second gap, and the electric field intensities of the sampling points are extracted through finite element simulation; the preset electric field intensity characteristic quantities of the sampling points of the first gap and the second gap are obtained respectively according to the electric field intensities of the sampling points of the first gap and the second gap; the parameter set is obtained according to the preset electric field intensity characteristic quantities of the sampling points of the first gap and the second gap; the first parameter set and the second parameter set are obtained through normalization processing according to the parameter sets of the first gap and the second gap.

[0012] The preset electric field intensity characteristic quantities include the maximum value of the electric field intensity, the minimum value of the electric field intensity, the average value of the electric field intensity, the median of the electric field intensity, the sum of the squares of the electric field intensity, the average value of the sum of the squares of the electric field intensity, the standard deviation of the electric field intensity, the ratio of the sum of the squares of the electric field intensity on the line segment where the electric field intensity is greater than 90% of the maximum value of the electric field intensity to the sum of the squares of the electric field intensity, the non-uniformity coefficient of the electric field intensity, the distortion rate of the electric field intensity, the coefficient of variation of the electric field intensity, the maximum value of the electric field gradient, the minimum value of the electric field gradient, the average value of the electric field gradient, and the median of the electric field gradient.

[0013] Preferably, obtaining the preset electric field intensity characteristic quantities of the sampling points of the first gap and the second gap respectively according to the electric field intensities of the sampling points of the first gap and the second gap includes: The first gap and the second gap are respectively subjected to the second processing to obtain the preset electric field intensity characteristic quantities of the sampling points of the first gap and the second gap; the second processing includes: Calculate the maximum electric field strength , minimum electric field strength , average electric field strength and the median electric field strength ,include: ; ; ; ; in, Indicates the sampling points; Indicates the The electric field strength at each sampling point; Indicates the total number of sampling points in the first gap or the second gap.

[0014] Calculate the sum of squares of the electric field strength , the average value of the sum of squares of the electric field strength , standard deviation of electric field strength And the sum of the squares of the electric field intensity on the line segment where the electric field intensity is greater than 90% of the maximum electric field intensity and the sum of the squares of the electric field intensity Ratio ,include: ; ; ; ; Calculation of electric field strength non-uniformity coefficient and , electric field intensity distortion rate and the coefficient of variation of electric field strength ,include: ; ; ; ; Calculate the maximum value of the electric field gradient , minimum electric field gradient , average electric field gradient and the median electric field gradient ,include: ; ; ; ; Among them, represents the electric field gradient at the th sampling point.

[0015] Preferably, the normalization process according to the parameter sets of the first gap and the second gap includes: The normalization process includes using the z-score normalization method for data processing to make the data conform to the standard normal distribution with a mean of 0 and a standard deviation of 1, including: ; Among them, represents the original data point; and respectively represent the mean and standard deviation of the feature; represents the data point after normalization.

[0016] Preferably, training the LSSVR model according to the first parameter set and the second parameter set respectively to obtain the first model and the second model includes: Optimizing the kernel function parameters and regularization parameters of the LSSVR model using the six-fold cross-validation method according to the first parameter set and the second parameter set respectively to obtain the first intermediate model and the second intermediate model.

[0017] Training the model according to the first intermediate model combined with the first parameter set and validating it through a preset validation set to obtain the first model; training the model according to the second intermediate model combined with the second parameter set and validating it through a preset validation set to obtain the second model.

[0018] Preferably, optimizing the kernel function parameters and regularization parameters of the LSSVR model using the six-fold cross-validation method according to the first parameter set and the second parameter set respectively to obtain the first intermediate model and the second intermediate model includes: Set the value ranges of the kernel function parameters and regularization parameters to be the preset parameter value set; the preset parameter value set includes a preset number of parameter values; exhaust all value combination cases of the kernel function parameters and regularization parameters according to the preset parameter value set, and obtain the parameter optimization training set according to all value combination cases.

[0019] Randomly divide the first parameter set and the second parameter set into 6 parameter subsets on average; initialize the LSSVR model according to the value combinations of the kernel function parameters and regularization parameters in the parameter optimization training set.

[0020] The LSSVR model for each combination of kernel function parameters and regularization parameter values is iteratively optimized 6 times according to the parameter subsets of the first parameter set. In each iteration, 5 parameter subsets are selected for training, 1 parameter subset is reserved for validation, and the mean absolute error MAE of the validation results is calculated. The mean absolute error MAE in all iterative optimizations of the LSSVR model for all combinations of kernel function parameters and regularization parameter values is statistically analyzed, and the combination of kernel function parameters and regularization parameter values corresponding to the minimum mean absolute error MAE is selected as the model parameters to obtain the first intermediate model.

[0021] The LSSVR model for each combination of kernel function parameters and regularization parameter values is iteratively optimized 6 times according to the parameter subsets of the second parameter set. In each iteration, 5 parameter subsets are selected for training, 1 parameter subset is reserved for validation, and the mean absolute error MAE of the validation results is calculated. The mean absolute error MAE in all iterative optimizations of the LSSVR model for all combinations of kernel function parameters and regularization parameter values is statistically analyzed, and the combination of kernel function parameters and regularization parameter values corresponding to the minimum mean absolute error MAE is selected as the model parameters to obtain the second intermediate model.

[0022] Calculating the mean absolute error MAE of the validation results includes: ; where represents the actual value; represents the predicted value; represents the number of samples.

[0023] Preferably, model training is performed according to the first intermediate model in combination with the first parameter set, and verification is performed through a preset validation set to obtain the first model; model training is performed according to the second intermediate model in combination with the second parameter set, and verification is performed through a preset validation set to obtain the second model, including: The optimization problem of the LSSVR model includes: ; The constraint statistics include: ; where represents the weight vector; represents the bias term; represents the mapping function that maps the input data to a high-dimensional feature space; represents the error term, represents the regularization parameter for regulating the model complexity; represents taking the transpose.

[0024] via the Lagrange multiplier With the introduction of [], the optimization problem of the LSSVR model is transformed into a dual problem: ; where represents the kernel function; represents the sample point to be predicted; represents the th sample point in the training set; Construct a kernel function matrix based on the Gaussian kernel function. The kernel function matrix includes the Gaussian kernel function values between all training samples , is the kernel width parameter; Substitute the kernel function matrix into the LSSVR model to obtain a system of linear equations; Solve the system of linear equations to obtain the values of and b, and then obtain the trained model; Verify the trained model based on a preset validation set, and determine the training effect of the model by calculating the mean absolute error MAE and the coefficient of determination R² on the preset validation set; Complete the model training after achieving the required training effect.

[0025] Preferably, the insulation strength evaluation includes: After inputting the actual values of the parameter set into the first model or the second model, the model outputs the predicted value of the 50% breakdown voltage of the target combined gap; Obtain the maximum switching overvoltage of the system according to the operating state and historical data of the power system.

[0026] Obtain the live working hazard rate of the target combined gap based on the predicted value of the 50% breakdown voltage of the target combined gap and the maximum switching overvoltage of the system, evaluate the safety of live working according to the hazard rate, and complete the insulation strength evaluation.

[0027] Preferably, obtaining the live working hazard rate of the target combined gap based on the predicted value of the 50% breakdown voltage of the target combined gap and the maximum switching overvoltage of the system, and evaluating the safety of live working according to the hazard rate includes: Assume that the probability distributions of system switching overvoltage and air gap breakdown both follow a normal distribution, then according to probability statistics, the hazard rate includes: ; ; ; ; where represents the hazard rate; represents the probability density function of the overvoltage amplitude; represents the probability distribution function of the air gap breakdown under the switching overvoltage with an amplitude of U; represents the standard deviation of the switching overvoltage; Represents the predicted value of the 50% breakdown voltage of the target combined gap; Represents the standard deviation of the discharge voltage of the live working combined gap; Represents the average value of switching overvoltage; Represents the relative standard deviation of overvoltage; Represents the maximum switching overvoltage of the system.

[0028] If the failure rate is less than the preset failure threshold, it is determined that the live working is safe; otherwise, it is determined that the live working is unsafe.

[0029] The present invention also provides an insulation strength evaluation device for a live working combined gap, which is used for the method of the present invention. The device includes a first module, a second module and a third module.

[0030] The first module is used to construct a finite element model for a preset data set; the preset data set includes combined gaps with a preset number of total gap lengths. Among them, the combined gaps that meet the first condition are classified into the first subset, and the combined gaps that do not meet the first condition are classified into the second subset; the first condition includes that the first gap length in the combined gap is greater than the first gap length critical value corresponding to the 50% breakdown voltage.

[0031] The finite element model includes a high-voltage electrode, a ground electrode and a floating conductor electrode; the floating conductor electrode is suspended between the high-voltage electrode and the ground electrode; the gap between the high-voltage electrode and the ground electrode is defined as the total gap, the gap between the floating conductor electrode and the high-voltage electrode is defined as the first gap, and the gap between the floating conductor electrode and the ground electrode is defined as the second gap.

[0032] The second module is used to respectively obtain the parameter sets of the first subset and the second subset to obtain the first parameter set and the second parameter set; the parameter set includes the preset electric field strength characteristic quantities of the first gap and the second gap in the combined gap; the first model and the second model are obtained by training the LSSVR model according to the first parameter set and the second parameter set respectively.

[0033] The third module is used to obtain the actual value of the first gap length and the actual value of the parameter set of the target combined gap; if the actual value of the first gap length meets the first condition, the insulation strength is evaluated according to the first model and the actual value of the parameter set, otherwise the insulation strength is evaluated according to the second model and the actual value of the parameter set.

[0034] The present invention has the following beneficial effects: The insulation strength evaluation method for the live working combined gap of the present invention can perform safety evaluation by obtaining the electric field characteristics through constructing a finite element model for the combined gap, making the method of the present invention have good evaluation efficiency and facilitating practical application. The LSSVR model is trained based on the training data of two cases to obtain the first model and the second model, enabling the method of the present invention to better capture the complex patterns and non-linear relationships in the corresponding data. The first model or the second model is dynamically selected for evaluation according to the actual situation of the target combined gap, enabling the method of the present invention to cover the combined gaps in their respective situations, avoiding the possible performance degradation problem of a single model when dealing with diverse data, and improving the accuracy of prediction evaluation, that is, the method of the present invention has flexibility, pertinence and versatility. The method of the present invention can accurately and efficiently evaluate the insulation strength of the live working combined gap.

[0035] The insulation strength evaluation device for the live working combined gap of the present invention is used for the method of the present invention and has the same beneficial effects as the method of the present invention.

[0036] In addition to the purposes, features and advantages described above, the present invention has other purposes, features and advantages. The present invention will be further described in detail below with reference to the accompanying drawings. Description of the Drawings

[0037] The accompanying drawings constituting a part of the present invention are used to provide a further understanding of the present invention. The schematic embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation to the present invention. In the drawings: Figure 1 It is a schematic flow chart of the method of the preferred embodiment of the present invention.

[0038] Figure 2 It is a schematic diagram of the combined gap of the preferred embodiment of the present invention. Detailed Embodiments

[0039] The following will describe the embodiments of the present invention in detail with reference to the accompanying drawings, but the present invention can be implemented in many different ways defined and covered by the claims.

[0040] In the preferred embodiment of the present invention, the high-voltage wire in live working is simplified and simulated by a high-voltage electrode, and the live working personnel at floating potential in the live working operation are simplified and simulated by a floating conductor electrode. The gap between the high-voltage electrode and the floating conductor electrode is used to simulate the potential transfer gap between the high-voltage wire and the live working personnel during the live working process. Refer to Figure 1 , in the preferred embodiment of the present invention, an insulation strength evaluation method for the live working combined gap is provided, including: S1. Construct a finite element model for a preset data set. The preset data set includes combined gaps of a preset number of total gap lengths, where the combined gaps that meet the first condition are classified into the first subset, and the combined gaps that do not meet the first condition are classified into the second subset. The first condition includes that the length of the first gap in the combined gap is greater than the critical value of the first gap length corresponding to 50% of the breakdown voltage. The critical value of the first gap length corresponding to 50% of the breakdown voltage is obtained based on historical data.

[0041] The finite element model includes a high-voltage electrode, a ground electrode, and a floating conductor electrode. The floating conductor electrode is suspended between the high-voltage electrode and the ground electrode. The gap between the high-voltage electrode and the ground electrode is defined as the total gap, the gap between the floating conductor electrode and the high-voltage electrode is defined as the first gap, and the gap between the floating conductor electrode and the ground electrode is defined as the second gap.

[0042] S2. Obtain the parameter sets of the first subset and the second subset respectively to get the first parameter set and the second parameter set. The parameter set includes preset electric field strength characteristic quantities of the first gap and the second gap in the combined gap. Train the LSSVR model according to the first parameter set and the second parameter set respectively to obtain the first model and the second model. The LSSVR model represents the Least Squares Support Vector Regression model.

[0043] In the preferred embodiment of the present invention, obtaining the parameter sets of the first subset and the second subset respectively to get the first parameter set and the second parameter set includes: Perform a first process on the first subset and the second subset respectively to obtain the first parameter set and the second parameter set.

[0044] The first process includes: equally spacedly select a preset number of sampling points in each of the first gap and the second gap, and extract the electric field strength of the sampling points through finite element simulation; obtain the preset electric field strength characteristic quantities of the sampling points of the first gap and the second gap respectively according to the electric field strength of the sampling points of the first gap and the second gap; obtain the parameter set according to the preset electric field strength characteristic quantities of the sampling points of the first gap and the second gap; perform normalization processing on the parameter sets of the first gap and the second gap to obtain the first parameter set and the second parameter set.

[0045] The preset electric field strength characteristic quantities include the maximum value of the electric field strength, the minimum value of the electric field strength, the average value of the electric field strength, the median of the electric field strength, the sum of the squares of the electric field strength, the average value of the sum of the squares of the electric field strength, the standard deviation of the electric field strength, the ratio of the sum of the squares of the electric field strength on the line segment where the electric field strength is greater than 90% of the maximum value of the electric field strength to the sum of the squares of the electric field strength, the non-uniformity coefficient of the electric field strength, the distortion rate of the electric field strength, the coefficient of variation of the electric field strength, the maximum value of the electric field gradient, the minimum value of the electric field gradient, the average value of the electric field gradient, and the median of the electric field gradient.

[0046] In a preferred embodiment of the present invention, obtaining the preset electric field intensity characteristic quantities of the sampling points of the first gap and the second gap according to the electric field intensity of the sampling points of the first gap and the second gap respectively includes: Performing a second process on the first gap and the second gap respectively to obtain a preset electric field intensity characteristic value of the sampling points of the first gap and the second gap; the second process includes: Calculate the maximum electric field strength , minimum electric field strength , average electric field strength and the median electric field strength ,include: ; ; ; ; in, Indicates the sampling points; Indicates the The electric field strength at each sampling point; Indicates the total number of sampling points in the first gap or the second gap.

[0047] Calculate the sum of squares of the electric field strength , the average value of the sum of squares of the electric field strength , standard deviation of electric field strength And the sum of the squares of the electric field intensity on the line segment where the electric field intensity is greater than 90% of the maximum electric field intensity and the sum of the squares of the electric field intensity Ratio ,include: ; ; ; ; Calculation of electric field strength non-uniformity coefficient and , electric field intensity distortion rate and the coefficient of variation of electric field strength ,include: ; ; ; ; Calculate the maximum value of the electric field gradient , minimum electric field gradient , average electric field gradient and the median of the electric field gradient , including: ; ; ; ; wherein, represents the electric field gradient at the th sampling point.

[0048] In a preferred embodiment of the present invention, the normalization process according to the parameter sets of the first gap and the second gap includes: The normalization process includes using the z-score normalization method to process the data so that the data conforms to the standard normal distribution with a mean of 0 and a standard deviation of 1, including: ; wherein, represents the original data point; and respectively represent the mean and standard deviation of the feature; represents the data point after normalization.

[0049] In a preferred embodiment of the present invention, training the LSSVR model according to the first parameter set and the second parameter set respectively to obtain the first model and the second model includes: Respectively optimize the kernel function parameters and regularization parameters of the LSSVR model using the six-fold cross-validation method according to the first parameter set and the second parameter set to obtain the first intermediate model and the second intermediate model.

[0050] The kernel function parameters control the width of the Gaussian kernel function and affect the model's ability to capture local features of the data; the regularization parameters are used to balance the fitting degree and complexity of the model. In a preferred embodiment of the present invention, in order to comprehensively explore the parameter space, the value ranges of the kernel function parameters and the regularization parameters are both set to [0.001, 0.01, 0.1, 1, 10, 100, 1000]. This range covers multiple orders of magnitude from extremely small to extremely large, aiming to find the parameter combination that makes the model perform best on the current data set.

[0051] Train the model according to the first intermediate model combined with the first parameter set and verify it through a preset validation set to obtain the first model; train the model according to the second intermediate model combined with the second parameter set and verify it through a preset validation set to obtain the second model.

[0052] In a preferred embodiment of the present invention, the kernel function parameters and regularization parameters of the LSSVR model are optimized by using the six-fold cross-validation method according to the first parameter set and the second parameter set respectively, and obtaining the first intermediate model and the second intermediate model includes: Set the value ranges of the kernel function parameters and the regularization parameters to be the preset parameter value set; the preset parameter value set includes a preset number of parameter values; exhaust all value combination cases of the kernel function parameters and the regularization parameters according to the preset parameter value set, and obtain the parameter optimization training set according to all value combination cases.

[0053] Randomly divide the first parameter set and the second parameter set into 6 parameter subsets on average; initialize the LSSVR model according to the value combinations of the kernel function parameters and the regularization parameters in the parameter optimization training set.

[0054] Perform 6 iterations of optimization on the LSSVR model with each combination of kernel function parameters and regularization parameters according to the parameter subsets of the first parameter set, where in each iteration, 5 parameter subsets are selected for training, 1 parameter subset is reserved for verification, and the mean absolute error MAE of the verification result is calculated; count the mean absolute error MAE in all iterations of optimization of the LSSVR model with all combinations of kernel function parameters and regularization parameters, and select the combination of kernel function parameters and regularization parameters corresponding to the minimum mean absolute error MAE as the model parameters to obtain the first intermediate model.

[0055] Perform 6 iterations of optimization on the LSSVR model with each combination of kernel function parameters and regularization parameters according to the parameter subsets of the second parameter set, where in each iteration, 5 parameter subsets are selected for training, 1 parameter subset is reserved for verification, and the mean absolute error MAE of the verification result is calculated; count the mean absolute error MAE in all iterations of optimization of the LSSVR model with all combinations of kernel function parameters and regularization parameters, and select the combination of kernel function parameters and regularization parameters corresponding to the minimum mean absolute error MAE as the model parameters to obtain the second intermediate model.

[0056] Calculating the mean absolute error MAE of the verification result includes: ; wherein, represents the actual value; represents the predicted value; represents the number of samples.

[0057] In a preferred embodiment of the present invention, the kernel function parameters and the regularization parameters of the LSSVR model are optimized by using the six-fold cross-validation method, which can effectively prevent the model from overfitting, and then significantly improve the generalization performance of the model when facing new data, making it have stronger applicability and reliability.

[0058] In a preferred embodiment of the present invention, model training is performed according to the first intermediate model in combination with the first parameter set, and verified through a preset validation set to obtain the first model; model training is performed according to the second intermediate model in combination with the second parameter set, and verified through a preset validation set to obtain the second model, including: The optimization problem of the LSSVR model includes: ; Constraint statistics include: ; wherein, represents the weight vector; represents the bias term; represents the mapping function that maps the input data to a high-dimensional feature space; represents the error term, represents the regularization parameter used to regulate the model complexity; represents taking the transpose.

[0059] Via the introduction of the Lagrange multiplier the optimization problem of the LSSVR model is transformed into a dual problem: ; wherein, represents the kernel function; represents the sample point to be predicted; represents the th sample point in the training set.

[0060] Construct a kernel function matrix based on the Gaussian kernel function, and the kernel function matrix includes the Gaussian kernel function values between all training samples , is the kernel width parameter; substitute the kernel function matrix into the LSSVR model to obtain a system of linear equations; solve the system of linear equations to obtain the and the value of b, and then obtain the trained model; verify the trained model based on the preset validation set, and determine the training effect of the model by calculating the mean absolute error MAE and the coefficient of determination R² on the preset validation set; complete the model training after reaching the required training effect. The smaller the MAE and the closer R² is to 1, the higher the prediction accuracy and the better the interpretability of the model.

[0061] S3. Obtain the actual value of the first gap length and the actual value of the parameter set of the target combined gap; if the actual value of the first gap length meets the first condition, evaluate the insulation strength according to the first model and the actual value of the parameter set, otherwise evaluate the insulation strength according to the second model and the actual value of the parameter set.

[0062] A finite element model is established according to the target combined gap, and the actual value of the first gap length and the actual value of the parameter set of the target combined gap are obtained. In a preferred embodiment of the present invention, the insulation strength evaluation includes: After the actual value of the parameter set is input into the first model or the second model, the model outputs the predicted value of the 50% breakdown voltage of the target combined gap; the maximum switching overvoltage of the system is obtained according to the operating state and historical data of the power system; The live working hazard rate of the target combined gap is obtained based on the predicted value of the 50% breakdown voltage of the target combined gap and the maximum switching overvoltage of the system, and the safety of live working is evaluated according to the hazard rate, thus completing the insulation strength evaluation.

[0063] In a preferred embodiment of the present invention, obtaining the live working hazard rate of the target combined gap based on the predicted value of the 50% breakdown voltage of the target combined gap and the maximum switching overvoltage of the system, and evaluating the safety of live working according to the hazard rate includes: Assuming that the probability distributions of the system switching overvoltage and the breakdown of the air gap are both assumed to follow a normal distribution, then the hazard rate according to probability statistics includes: ; ; ; ; Among them, represents the hazard rate; represents the probability density function of the overvoltage amplitude; represents the probability distribution function of the breakdown of the air gap under the switching overvoltage with an amplitude of U; represents the standard deviation of the switching overvoltage (kV), generally taking 12% - 15% of the switching overvoltage; represents the predicted value of the 50% breakdown voltage of the target combined gap (kV); represents the standard deviation of the discharge voltage of the live working combined gap (kV), generally taking 6% of represents the average value of the switching overvoltage (kV); represents the relative standard deviation of the overvoltage, generally taking 12% - 15%; represents the maximum switching overvoltage of the system.

[0064] If the hazard rate is less than the preset hazard threshold, it is determined that the live working is safe, otherwise it is determined that the live working is unsafe. In a preferred embodiment of the present invention, the preset hazard threshold includes .

[0065] The insulation strength evaluation method for live working combined gaps of the present invention can perform safety evaluation by obtaining the electric field characteristics through constructing a finite element model for the combined gaps, so that the method of the present invention has good evaluation efficiency and is convenient for practical application. The LSSVR model is trained based on the training data of two cases to obtain the first model and the second model, so that the method of the present invention can better capture the complex patterns and non-linear relationships in the corresponding data. The first model or the second model is dynamically selected for evaluation according to the actual situation of the target combined gap, so that the method of the present invention can cover the combined gaps of their respective situations, avoiding the problem of performance degradation that may occur when a single model processes diverse data, and improving the accuracy of prediction evaluation, that is, the method of the present invention has flexibility, pertinence and universality. The method of the present invention can accurately and efficiently evaluate the insulation strength of live working combined gaps.

[0066] In a preferred embodiment of the present invention, there is also provided an insulation strength evaluation device for live working combined gaps, which is used for the method of the present invention. The device includes a first module, a second module and a third module.

[0067] The first module is used to construct a finite element model for a preset data set; the preset data set includes combined gaps of a preset number of total gap lengths, wherein the combined gaps that meet the first condition are classified into a first subset, and the combined gaps that do not meet the first condition are classified into a second subset; the first condition includes that the first gap length in the combined gap is greater than the first gap length critical value corresponding to 50% of the breakdown voltage.

[0068] The finite element model includes a high-voltage electrode, a ground electrode and a floating conductor electrode; the floating conductor electrode is suspended between the high-voltage electrode and the ground electrode; the gap between the high-voltage electrode and the ground electrode is defined as the total gap, the gap between the floating conductor electrode and the high-voltage electrode is defined as the first gap, and the gap between the floating conductor electrode and the ground electrode is defined as the second gap.

[0069] The second module is used to respectively obtain the parameter sets of the first subset and the second subset to obtain a first parameter set and a second parameter set; the parameter set includes preset electric field intensity characteristic quantities of the first gap and the second gap in the combined gap; the LSSVR model is respectively trained according to the first parameter set and the second parameter set to obtain the first model and the second model.

[0070] The third module is used to obtain the actual value of the first gap length and the actual value of the parameter set of the target combined gap; if the actual value of the first gap length meets the first condition, the insulation strength is evaluated according to the first model and the actual value of the parameter set, otherwise the insulation strength is evaluated according to the second model and the actual value of the parameter set.

[0071] The insulation strength evaluation device for live working combined gaps of the present invention, which is used for the method of the present invention, has the same beneficial effects as the method of the present invention.

[0072] Verification part: Refer to Figure 2 , according to the on-site working conditions of live working on a certain 220 kV transmission line, a live working combined gap model is constructed in COMSOL Multiphysics, where the total gap length is 2.0 m, the length of sub-gap 1 (i.e., the first gap) is 0.7 m, the length of sub-gap 2 (i.e., the second gap) is 1.1 m, and the length of the suspended conductor electrode is 0.5 m.

[0073] 11 sampling points are set in each of the first gap and the second gap. The electric field strength values at each sampling point are obtained through finite element simulation, and the electric field characteristic quantities at each sampling point are calculated accordingly. According to historical data statistics, for a total gap length of 2.0 m, the critical value of the first gap length corresponding to the 50% breakdown voltage is 0.6 m. Since the length of the first gap in the current configuration is 0.7 m, the first model is selected. The input characteristic quantities are subjected to z-score standardization, and then the standardized feature vector is input into the first model to obtain the prediction result: the 50% breakdown voltage is 578.6 kV. Therefore, the standard deviation of the discharge voltage of the live working combined gap parameters = 578.6×0.06 = 34.7 kV. According to the historical overvoltage level information of this transmission line obtained, the maximum switching overvoltage of the system is 420 kV. The relative standard deviation of the overvoltage is taken as 13%, then the average value of the switching overvoltage: =420 / (1 + 2.05×0.13) = 331.6 kV, and the standard deviation of the switching overvoltage = 331.6×0.13 = 43.1 kV. The live working hazard rate is calculated according to the formula, and is obtained through numerical integration . Since the calculated live working hazard rate , it meets the safety standard requirements. Evaluation conclusion: The live working on this 220 kV transmission line is safe under the current configuration and the operation can be carried out.

[0074] The above are only the preferred embodiments of the present invention and are not used to limit the present invention. For those skilled in the art, the present invention can have various changes and modifications. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. An insulation strength evaluation method for a live working combined gap, characterized in that Including: Constructing a finite element model for a preset data set; The preset data set includes combined gaps with a preset number of total gap lengths, where the combined gaps that meet the first condition are classified into a first subset, and the combined gaps that do not meet the first condition are classified into a second subset; the first condition includes that the first gap length in the combined gap is greater than the first gap length critical value corresponding to 50% of the breakdown voltage; The finite element model includes a high-voltage electrode, a ground electrode, and a floating conductor electrode; the floating conductor electrode floats between the high-voltage electrode and the ground electrode; the gap between the high-voltage electrode and the ground electrode is defined as the total gap, the gap between the floating conductor electrode and the high-voltage electrode is defined as the first gap, and the gap between the floating conductor electrode and the ground electrode is defined as the second gap; Respectively obtaining the parameter sets of the first subset and the second subset to obtain a first parameter set and a second parameter set; the parameter set includes preset electric field intensity characteristic quantities of the first gap and the second gap in the combined gap; Respectively training LSSVR models according to the first parameter set and the second parameter set to obtain a first model and a second model; Obtaining the actual value of the first gap length and the actual value of the parameter set of the target combined gap; if the actual value of the first gap length meets the first condition, then perform insulation strength evaluation according to the first model and the actual value of the parameter set, otherwise perform insulation strength evaluation according to the second model and the actual value of the parameter set.

2. The method for evaluating the insulation strength of the live working combined gap according to claim 1, characterized in that Respectively obtaining the parameter sets of the first subset and the second subset to obtain a first parameter set and a second parameter set includes: Performing a first process on the first subset and the second subset respectively to obtain the first parameter set and the second parameter set; The first process includes: equally spacing and selecting a preset number of sampling points in each of the first gap and the second gap, and extracting the electric field intensity of the sampling points through finite element simulation; respectively obtaining the preset electric field intensity characteristic quantities of the sampling points of the first gap and the second gap according to the electric field intensity of the sampling points of the first gap and the second gap; obtaining the parameter set according to the preset electric field intensity characteristic quantities of the sampling points of the first gap and the second gap; performing normalization processing on the parameter sets of the first gap and the second gap to obtain the first parameter set and the second parameter set; The preset electric field intensity characteristic quantities include the maximum value of the electric field intensity, the minimum value of the electric field intensity, the average value of the electric field intensity, the median of the electric field intensity, the sum of the squares of the electric field intensity, the average value of the sum of the squares of the electric field intensity, the standard deviation of the electric field intensity, the ratio of the sum of the squares of the electric field intensity on the line segment where the electric field intensity is greater than 90% of the maximum value of the electric field intensity to the sum of the squares of the electric field intensity, the electric field intensity non-uniformity coefficient, the electric field intensity distortion rate, the coefficient of variation of the electric field intensity, the maximum value of the electric field gradient, the minimum value of the electric field gradient, the average value of the electric field gradient, and the median of the electric field gradient.

3. The method for evaluating the insulation strength of the live working combined gap according to claim 2, wherein The respectively obtaining the preset electric field intensity characteristic quantities of the sampling points of the first gap and the second gap according to the electric field intensity of the sampling points of the first gap and the second gap includes: Performing a second process on the first gap and the second gap respectively to obtain the preset electric field intensity characteristic quantities of the sampling points of the first gap and the second gap; the second process includes: Calculate the maximum value of the electric field strength , the minimum value of the electric field strength , the average value of the electric field strength and the median value of the electric field strength , including: ； ； ； ； Among them, represents the th sampling point; represents the electric field strength of the th sampling point; represents the total number of sampling points in the first gap or the second gap; Calculating the sum of the squares of the electric field strengths , the average value of the sum of the squares of the electric field strengths , the standard deviation of the electric field strength and the ratio of the sum of the squares of the electric field strengths on the line segment where the electric field strength is greater than 90% of the maximum value of the electric field strength to the sum of the squares of the electric field strengths , including: ​ ； ； ； ； Calculating the non-uniformity coefficient of electric field strength and , the distortion rate of electric field strength and the coefficient of variation of electric field strength , including: ； ； ； ； Calculate the maximum value of the electric field gradient , the minimum value of the electric field gradient , the average value of the electric field gradient and the median value of the electric field gradient , including: ； ； ； ； Among them, represents the electric field gradient at the th sampling point.

4. The method for evaluating the insulation strength of the live working combined gap according to claim 3, characterized in that The normalization process according to the parameter sets of the first gap and the second gap includes: The normalization process includes performing data processing using the z-score normalization method, including: ； Among them, represents the original data point; and respectively represent the mean and standard deviation of the feature; represents the data point after standardization.

5. The method for evaluating the insulation strength of the live working combined gap according to claim 4, characterized in that, Training the LSSVR model according to the first parameter set and the second parameter set respectively to obtain the first model and the second model includes: Optimizing the kernel function parameters and regularization parameters of the LSSVR model using the six-fold cross-validation method according to the first parameter set and the second parameter set respectively to obtain the first intermediate model and the second intermediate model; Training the model according to the first intermediate model combined with the first parameter set and validating it through a preset validation set to obtain the first model; training the model according to the second intermediate model combined with the second parameter set and validating it through a preset validation set to obtain the second model.

6. The method for evaluating the insulation strength of the live working combined gap according to claim 5, characterized in that, Optimizing the kernel function parameters and regularization parameters of the LSSVR model using the six-fold cross-validation method according to the first parameter set and the second parameter set respectively to obtain the first intermediate model and the second intermediate model includes: Setting the value ranges of the kernel function parameters and regularization parameters to a preset parameter value set; the preset parameter value set includes a preset number of parameter values; exhausting all value combination cases of the kernel function parameters and regularization parameters according to the preset parameter value set, and obtaining a parameter optimization training set according to all the value combination cases; Randomly dividing the first parameter set and the second parameter set into 6 parameter subsets on average; initializing the LSSVR model according to the value combinations of the kernel function parameters and regularization parameters in the parameter optimization training set; Performing 6 iterations of optimization on the LSSVR model with each combination of kernel function parameters and regularization parameters according to the parameter subsets of the first parameter set, where 5 parameter subsets are selected for training and 1 parameter subset is reserved for validation in each iteration, and calculating the mean absolute error MAE of the validation result; statistically calculating the mean absolute error MAE in all iterations of optimization of the LSSVR model with all combinations of kernel function parameters and regularization parameters, and selecting the combination of kernel function parameters and regularization parameters corresponding to the minimum mean absolute error MAE as the model parameters to obtain the first intermediate model; Performing 6 iterations of optimization on the LSSVR model with each combination of kernel function parameters and regularization parameters according to the parameter subsets of the second parameter set, where 5 parameter subsets are selected for training and 1 parameter subset is reserved for validation in each iteration, and calculating the mean absolute error MAE of the validation result; statistically calculating the mean absolute error MAE in all iterations of optimization of the LSSVR model with all combinations of kernel function parameters and regularization parameters, and selecting the combination of kernel function parameters and regularization parameters corresponding to the minimum mean absolute error MAE as the model parameters to obtain the second intermediate model; The calculation of the mean absolute error MAE of the validation result includes: ； Among them, represents the actual value; represents the predicted value; represents the number of samples.

7. The method for evaluating the insulation strength of the live working combined gap according to claim 6, characterized in that, Training the model according to the first intermediate model combined with the first parameter set and validating it through a preset validation set to obtain the first model; Performing model training based on the second intermediate model in combination with the second parameter set and validating through a preset validation set to obtain the second model includes: The optimization problems of the LSSVR model include: ； Constraint statistics include: ； Among them, represents the weight vector; represents the bias term; represents the mapping function that maps the input data to a high-dimensional feature space; represents the error term, represents the regularization parameter used to control the model complexity; represents taking the transpose; Via the introduction of Lagrange multipliers the optimization problem of the LSSVR model is transformed into a dual problem: ； Among them, represents a kernel function; represents a sample point to be predicted; represents the th sample point in the training set; Construct a kernel function matrix based on the Gaussian kernel function, where the kernel function matrix includes the Gaussian kernel function values between all training samples , is the kernel width parameter; substitute the kernel function matrix into the LSSVR model to obtain a system of linear equations; solve the system of linear equations to obtain and the value of b, and then obtain the trained model; verify the trained model based on a preset validation set, and determine the training effect of the model by calculating the mean absolute error MAE and the coefficient of determination R² on the preset validation set; complete the model training after achieving the required training effect.

8. The method for evaluating the insulation strength of the live working combined gap according to claim 7, characterized in that, Performing the insulation strength assessment includes: After inputting the actual values of the parameter set into the first model or the second model, the model outputs the predicted value of the 50% breakdown voltage of the target combined gap; obtaining the maximum switching overvoltage of the power system according to the operating state and historical data of the power system; Obtaining the live working hazard rate of the target combined gap based on the predicted value of the 50% breakdown voltage of the target combined gap and the maximum switching overvoltage of the system, and evaluating the safety of live working according to the hazard rate to complete the insulation strength assessment.

9. The method for evaluating the insulation strength of a live working gap according to claim 8, wherein: Obtaining the live working hazard rate of the target combined gap based on the predicted value of the 50% breakdown voltage of the target combined gap and the maximum switching overvoltage of the system, and evaluating the safety of live working according to the hazard rate includes: Assuming that the probability distributions of the system switching overvoltage and the air gap breakdown are both assumed to follow a normal distribution, then according to probability statistics, the hazard rate includes: ； ； ； ； Among them, represents the hazard rate; represents the probability density function of the overvoltage amplitude; represents the probability distribution function of the breakdown of the air gap under the switching overvoltage with an amplitude of U; represents the standard deviation of the switching overvoltage; represents the predicted value of the 50% breakdown voltage of the target combined gap; represents the standard deviation of the discharge voltage of the live working combined gap; represents the average value of the switching overvoltage; represents the relative standard deviation of the overvoltage; represents the maximum switching overvoltage of the system; If the hazard rate is less than the preset hazard threshold, it is determined that the live working is safe; otherwise, it is determined that the live working is unsafe.

10. An insulating strength evaluation device for a live working combined gap, which is used for the method according to any one of claims 1 to 9, and is characterized in that The device includes a first module, a second module, and a third module; The first module is used to construct a finite element model for a preset data set; the preset data set includes combined gaps with a preset number of total gap lengths, where the combined gaps that meet the first condition are classified into the first subset, and the combined gaps that do not meet the first condition are classified into the second subset; the first condition includes that the first gap length in the combined gap is greater than the critical value of the first gap length corresponding to the 50% breakdown voltage; The finite element model includes a high-voltage electrode, a grounding electrode, and a floating conductor electrode; the floating conductor electrode floats between the high-voltage electrode and the grounding electrode; the gap between the high-voltage electrode and the grounding electrode is defined as the total gap, the gap between the floating conductor electrode and the high-voltage electrode is defined as the first gap, and the gap between the floating conductor electrode and the grounding electrode is defined as the second gap; The second module is used to respectively obtain the parameter sets of the first subset and the second subset to obtain a first parameter set and a second parameter set; the parameter set includes preset electric field strength characteristic quantities of the first gap and the second gap in the combined gap; Training the LSSVR model according to the first parameter set and the second parameter set respectively to obtain a first model and a second model; The third module is used to obtain the actual value of the first gap length of the target combined gap and the actual value of the parameter set; if the actual value of the first gap length meets the first condition, perform insulation strength assessment according to the first model and the actual value of the parameter set, otherwise perform insulation strength assessment according to the second model and the actual value of the parameter set.

Images