Vacuum arc extinguish chamber contact temperature inversion method, device and equipment in environment-friendly switch and storage medium

Abstract

The invention discloses a vacuum arc extinguish chamber contact temperature inversion method, device and equipment in an environment-friendly switch and a storage medium, and relates to the technical field of environment-friendly switch temperature distribution.The method comprises the steps that a vacuum circuit breaker three-dimensional geometric model built based on the environment-friendly switch is used for simulating a vacuum arc extinguish chamber; obtaining a temperature gradient distribution cloud picture and a gas streamline distribution diagram in the gas chamber, and determining a temperature point location based on each distribution diagram; determining an association relationship between the initial contact temperature and the temperature point location, and constructing an initial contact temperature inversion calculation model based on the point location temperature of the temperature point location, the association relationship and the initial contact temperature; hyper-parameters of the initial contact temperature inversion calculation model are adjusted, the model is adjusted by means of the target hyper-parameters, and contact temperature inversion operation is carried out on the basis of the load current of the vacuum arc-extinguishing chamber, the air chamber pressure intensity, the environment temperature, the incidence relation and the initial contact temperature by means of the target model to obtain the target contact temperature. Therefore, the contact temperature inversion efficiency can be improved.

Description

Technical Field

[0001] This invention relates to the field of temperature distribution technology for environmental protection switches, and in particular to a method, apparatus, equipment, and storage medium for inverting the temperature of the vacuum interrupter contacts in an environmental protection switch. Background Technology

[0002] Currently, environmentally friendly GIS (Gas-Insulated Metal Enclosed Switchgear) switches possess excellent insulation performance and environmentally friendly characteristics. As the core component of environmentally friendly GIS, the operating state of the internal contacts of the vacuum interrupter directly affects the performance of the entire switchgear. Because the contacts are completely sealed within a high-vacuum, closed cavity, and entirely isolated from the external environment, their convective heat dissipation performance is extremely poor. Traditional contact-based temperature measurement methods cannot be directly applied to online monitoring of contact temperature. During long-term operation, the contacts heat up due to carrying the load current, and the accumulated heat will significantly impact contact performance and even the mechanical and electrical performance of the entire switchgear, potentially leading to serious malfunctions. Therefore, knowing the contact temperature and mastering a precise, real-time, non-contact inversion method for measuring the internal contact temperature of the vacuum interrupter is crucial for ensuring its safe and stable operation.

[0003] Existing research on the temperature characteristics of vacuum interrupter contacts in environmentally friendly switches mainly focuses on methods for calculating the contact temperature. For example, one method for inverting and training the contact temperature of a vacuum interrupter involves obtaining temperature data from measurement points and contacts through finite element simulation of the vacuum interrupter, selecting the optimal measurement point, and then using an RBF (Radial Basis Function) neural network. This model uses the optimal measurement point temperature data as training data and the contact temperature as the target variable to train and fit a model for contact temperature inversion. Another method for monitoring the temperature rise of GIS contacts based on deep learning networks involves establishing a sample dataset and selecting a labeled dataset, inputting it into a ConvGRU (Convolutional Gated Recurrent Unit) model to obtain a spatiotemporal relationship function, and then mapping it to construct a GIS contact temperature rise model. However, methods for the actual measurement and inversion calculation of vacuum interrupter contact temperature in environmentally friendly GIS equipment are relatively lacking.

[0004] As can be seen from the above, improving the efficiency of temperature inversion of the contacts in the vacuum interrupter in environmental protection switches is an urgent problem to be solved. Summary of the Invention

[0005] In view of this, the purpose of this invention is to provide a method, apparatus, device, and storage medium for inverting the temperature of the contacts in a vacuum interrupter in an environmentally friendly switch, which can improve the efficiency of temperature inversion of the contacts in the vacuum interrupter during the temperature inversion process. The specific solution is as follows:

[0006] In a first aspect, this application provides a method for inverting the contact temperature of a vacuum interrupter in an environmentally friendly switch, including:

[0007] Based on the aforementioned environmentally friendly switch, a three-dimensional geometric model of a vacuum circuit breaker is constructed. This model is then used to perform multi-physics simulation calculations on the vacuum interrupter chamber, resulting in a temperature gradient distribution cloud map on the outer surface of the ceramic shell of the interrupter chamber and a gas streamline distribution map within the chamber. Based on the temperature gradient distribution cloud map, the gas streamline distribution map, and a preset heat and mass transfer theory, the temperature points are determined.

[0008] The correlation between the initial contact temperature and the temperature point is determined. Then, the XGBoost algorithm is used to construct an initial contact temperature inversion calculation model based on the point temperature corresponding to the temperature point, the correlation, and the initial contact temperature.

[0009] The initial hyperparameters corresponding to the initial contact temperature inversion calculation model are adjusted using the grid search method and cross-validation method to obtain the target hyperparameters. The target hyperparameters are then used to adjust the initial contact temperature inversion calculation model to obtain the target contact temperature inversion calculation model.

[0010] The target contact temperature is obtained by using the target contact temperature inversion calculation model and performing a contact temperature inversion operation based on the load current of the vacuum interrupter, the gas chamber pressure, the ambient temperature, the correlation relationship, and the initial contact temperature.

[0011] Optionally, the step of constructing a three-dimensional geometric model of the vacuum circuit breaker based on the environmentally friendly switch, and using the three-dimensional geometric model of the vacuum circuit breaker to perform multi-physics simulation calculations on the vacuum interrupter, to obtain the temperature gradient distribution cloud map on the outer surface of the ceramic shell of the interrupter and the gas streamline distribution map inside the gas chamber, includes:

[0012] A three-dimensional geometric model of the vacuum circuit breaker is constructed based on the GIS equipment corresponding to the environmental protection switch. The ambient temperature of the vacuum interrupter is determined using the three-dimensional geometric model of the vacuum circuit breaker, and then the spatial derivative of the ambient temperature in each direction is determined.

[0013] The temperature change is determined based on the spatial derivatives of each of the above, the corresponding modulus is determined based on the temperature change, and the temperature change rate is determined based on the modulus. Then, the rate of change with the largest value among the temperature change rates is set as the target rate of change.

[0014] The temperature gradient distribution cloud map on the outer surface of the ceramic shell of the arc-extinguishing chamber is determined using the three-dimensional geometric model of the vacuum circuit breaker and based on the target rate of change.

[0015] The motion trajectory of fluid particles in the vacuum interrupter at the same moment is simulated using the three-dimensional geometric model of the vacuum circuit breaker to obtain a gas streamline distribution diagram; the gas streamline distribution diagram is used to reflect the influence of gas flow on heat and mass transfer in the vacuum interrupter.

[0016] Optionally, determining the temperature point based on the temperature gradient distribution cloud map, the gas streamline distribution map, and the preset heat and mass transfer theory includes:

[0017] Extract each temperature gradient from the temperature gradient distribution cloud map and determine whether the absolute value corresponding to each temperature gradient is greater than a preset temperature gradient threshold. If the absolute value corresponding to each temperature gradient is greater than the preset temperature gradient threshold, then set each temperature measurement point corresponding to the temperature gradient whose absolute value is greater than the preset temperature gradient threshold as the first temperature measurement point to be screened.

[0018] Extract the streamlines to be judged corresponding to each of the first temperature measurement points to be screened from the gas streamline distribution map, and determine whether the streamline distribution corresponding to the streamline to be judged meets the preset streamline distribution conditions. If the streamline distribution corresponding to the streamline to be judged meets the preset streamline distribution conditions, then set the first temperature measurement point to be screened corresponding to the streamline distribution that meets the preset streamline distribution conditions as the second temperature measurement point to be screened.

[0019] The second temperature measurement points to be screened are selected using a preset heat and mass transfer theory to obtain temperature points; wherein, the selected temperature points include the area of ​​the outer surface of the ceramic shell of the vacuum interrupter near the end cover plate of the stationary guide rod, the area of ​​the outer surface of the main shield that is in the same plane as the lower surface of the moving contact plate, and the area near the end cover plate of the moving guide rod.

[0020] Optionally, the step of determining the correlation between the initial contact temperature and the temperature point, and then using the XGBoost algorithm to construct an initial contact temperature inversion calculation model based on the point temperature corresponding to the temperature point, the correlation, and the initial contact temperature, includes:

[0021] The correlation between the initial contact temperature and the temperature point is determined by using a preset correlation relationship determination rule. Then, a new decision tree is constructed using the XGBoost algorithm and the prediction residual of the previous iteration of the initial contact temperature inversion calculation model. The Hessian matrix corresponding to the correlation relationship is used to perform Taylor expansion to obtain the optimized tree structure.

[0022] The weights corresponding to the leaf nodes in the optimization tree structure are determined, and the initial contact temperature inversion calculation model is evaluated based on the weights and the split gain of the optimization tree structure to obtain the evaluation results. The loss function is then determined based on the evaluation results.

[0023] The regularization term is determined based on the number of leaf nodes in the optimized tree structure, the weight of each leaf node, and the preset regularization parameter. The objective function is then determined based on the loss function and the regularization term. The parameters of the initial contact temperature inversion calculation model are then adjusted using the objective function and based on the point temperature corresponding to the temperature point, the correlation relationship, and the initial contact temperature to obtain a new initial contact temperature inversion calculation model.

[0024] Optionally, the step of adjusting the initial hyperparameters corresponding to the initial contact temperature inversion calculation model using the grid search method and cross-validation method to obtain target hyperparameters, and then using the target hyperparameters to adjust the initial contact temperature inversion calculation model to obtain the target contact temperature inversion calculation model, includes:

[0025] The first hyperparameter affecting the model's learning ability is determined from the initial hyperparameters corresponding to the initial contact temperature inversion calculation model. The first hyperparameter is then combined using grid search and cross-validation methods to obtain the first target hyperparameter. The first hyperparameter includes the learning rate and the number of decision trees.

[0026] The second hyperparameter in the initial hyperparameters is adjusted using a preset tree complexity condition to obtain a second target hyperparameter. Then, the third hyperparameter in the initial hyperparameters is adjusted using a preset overfitting condition to obtain a third target hyperparameter. The target hyperparameter is determined based on the first target hyperparameter, the second target hyperparameter, and the third target hyperparameter. The second hyperparameter includes the maximum tree depth and the minimum sum of leaf weights. The third hyperparameter includes a split threshold, L2 regularization, and L1 regularization.

[0027] Optionally, adjusting the initial contact temperature inversion calculation model using the target hyperparameters to obtain the target contact temperature inversion calculation model includes:

[0028] The model parameters corresponding to the initial contact temperature inversion calculation model are adjusted using the target hyperparameters to obtain the target contact temperature inversion calculation model; wherein, the learning rate is used to control the contribution of a single tree; the number of decision trees is used to determine the model capacity; the maximum tree depth is used to limit the tree complexity; the minimum leaf weight is used to control the number of samples for node splitting; the splitting threshold is used to suppress low-gain splitting; the L2 regularization is used to constrain the weight amplitude; and the L1 regularization is used to promote feature sparsity.

[0029] Optionally, the step of using the target contact temperature inversion calculation model and performing a contact temperature inversion operation based on the load current, chamber pressure, ambient temperature, correlation, and initial contact temperature of the vacuum interrupter to obtain the target contact temperature includes:

[0030] The load current, chamber pressure, and ambient temperature corresponding to the vacuum interrupter are determined, and the target contact temperature is obtained by using the target contact temperature inversion calculation model and performing contact temperature inversion operation based on the load current, chamber pressure, ambient temperature, correlation, and initial contact temperature.

[0031] The target contact temperature output by the target contact temperature inversion calculation model is evaluated using an error evaluation index to obtain an evaluation result. The model parameters of the target contact temperature inversion calculation model are then adjusted using the evaluation result to obtain a new target contact temperature inversion calculation model.

[0032] The error evaluation indicators include root mean square error, mean absolute error, mean absolute percentage error, and maximum absolute error; the root mean square error is used to quantify the standard deviation between the predicted value and the true value; the mean absolute error is used to calculate the absolute average of the prediction error; the mean absolute percentage error is used to evaluate the mean of the relative error percentage between the predicted value and the true value; the maximum absolute error is used to identify the most severe deviation in the prediction result; the predicted value is the target contact temperature; and the true value is the initial contact temperature.

[0033] Secondly, this application provides a device for retrieving the temperature of the vacuum interrupter contact in an environmentally friendly switch, comprising:

[0034] The temperature point determination module is used to construct a three-dimensional geometric model of the vacuum circuit breaker based on the environmental protection switch, and to use the three-dimensional geometric model of the vacuum circuit breaker to perform multi-physics field simulation calculations on the vacuum interrupter to obtain the temperature gradient distribution cloud map on the outer surface of the ceramic shell of the interrupter and the gas streamline distribution map in the gas chamber, and to determine the temperature point based on the temperature gradient distribution cloud map, the gas streamline distribution map and the preset heat and mass transfer theory.

[0035] The correlation determination module is used to determine the correlation between the initial contact temperature of the contact and the temperature point, and then use the XGBoost algorithm to construct an initial contact temperature inversion calculation model based on the point temperature corresponding to the temperature point, the correlation, and the initial contact temperature.

[0036] The hyperparameter adjustment module is used to adjust the initial hyperparameters corresponding to the initial contact temperature inversion calculation model using the grid search method and cross-validation method to obtain the target hyperparameters, and then use the target hyperparameters to adjust the initial contact temperature inversion calculation model to obtain the target contact temperature inversion calculation model.

[0037] The contact temperature determination module is used to perform contact temperature inversion operation based on the target contact temperature inversion calculation model and the load current, chamber pressure, ambient temperature, correlation relationship and initial contact temperature of the vacuum interrupter to obtain the target contact temperature.

[0038] Thirdly, this application provides an electronic device, comprising:

[0039] Memory, used to store computer programs;

[0040] A processor is used to execute the computer program to implement the aforementioned method for inverting the temperature of the vacuum interrupter contacts in the environmentally friendly switch.

[0041] Fourthly, this application provides a computer-readable storage medium for storing a computer program, wherein the computer program, when executed by a processor, implements the aforementioned method for inverting the temperature of the vacuum interrupter contacts in an environmentally friendly switch.

[0042] As can be seen from the above, before performing the temperature inversion of the vacuum interrupter contact in the environmental protection switch, this application needs to construct a three-dimensional geometric model of the vacuum circuit breaker based on the environmental protection switch. This model is then used to perform multiphysics simulation calculations on the vacuum interrupter, obtaining the temperature gradient distribution cloud map on the outer surface of the ceramic shell and the gas streamline distribution map within the gas chamber. Based on the temperature gradient distribution cloud map, the gas streamline distribution map, and a preset heat and mass transfer theory, the temperature points are determined. The correlation between the initial contact temperature and the temperature points is then determined. Finally, the XGBoost algorithm is used, and the temperature points are used to... An initial contact temperature inversion calculation model is constructed using the corresponding point temperature, the aforementioned correlation, and the initial contact temperature. The initial hyperparameters corresponding to the initial contact temperature inversion calculation model are adjusted using a grid search method and cross-validation to obtain target hyperparameters. These target hyperparameters are then used to adjust the initial contact temperature inversion calculation model to obtain a target contact temperature inversion calculation model. Finally, using the target contact temperature inversion calculation model and based on the load current of the vacuum interrupter, the chamber pressure, the ambient temperature, the aforementioned correlation, and the initial contact temperature, a contact temperature inversion operation is performed to obtain the target contact temperature.

[0043] Therefore, this application first needs to construct a three-dimensional geometric model of the vacuum circuit breaker based on the aforementioned environmentally friendly switch, and then use this three-dimensional geometric model to perform multiphysics simulation calculations on the vacuum interrupter chamber to obtain the temperature gradient distribution cloud map on the outer surface of the ceramic shell of the interrupter chamber and the gas streamline distribution map inside the chamber. Based on the temperature gradient distribution cloud map, the gas streamline distribution map, and a preset heat and mass transfer theory, the temperature points are determined. Then, the correlation between the initial contact temperature of the contacts and the temperature points is determined. Finally, the XGBoost algorithm is used, based on the point temperature corresponding to the temperature point and the correlation... An initial contact temperature inversion calculation model is constructed based on the initial contact temperature. Then, the initial hyperparameters corresponding to the initial contact temperature inversion calculation model are adjusted using a grid search method and cross-validation to obtain target hyperparameters. These target hyperparameters are then used to adjust the initial contact temperature inversion calculation model to obtain a target contact temperature inversion calculation model. Finally, using the target contact temperature inversion calculation model and based on the load current, chamber pressure, ambient temperature, and the correlation between the initial contact temperature and the vacuum interrupter, a contact temperature inversion operation is performed to obtain the target contact temperature. This improves the efficiency of temperature inversion for the vacuum interrupter contacts in the environmental protection switch, thereby enhancing the user experience. Attached Figure Description

[0044] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0045] Figure 1 This is a flowchart of a method for inverting the temperature of the vacuum interrupter contact in an environmentally friendly switch disclosed in this application;

[0046] Figure 2 This application discloses a flowchart for temperature inversion of the vacuum interrupter contacts in an environmental protection switch.

[0047] Figure 3 This is a schematic diagram illustrating the principle of a specific XGBoost algorithm disclosed in this application;

[0048] Figure 4 This is a schematic diagram of a three-dimensional geometric model of a specific vacuum circuit breaker chamber disclosed in this application;

[0049] Figure 5 This application discloses a specific temperature gradient distribution cloud map along the gravity direction on the outer surface of the ceramic shell of an arc-extinguishing chamber; wherein, Figure 5 (a) Temperature gradient distribution cloud map corresponding to clean air at 0.7 MPa. Figure 5 (b) Temperature gradient distribution cloud map corresponding to CO2 gas at 0.5 MPa;

[0050] Figure 6 This is a distribution diagram of clean air inside the gas chamber of a specific vacuum circuit breaker disclosed in this application; wherein, Figure 6 (a) is a diagram showing the gas streamline vector distribution when the gas inside the vacuum circuit breaker chamber is clean air. Figure 6 (b) is a gas streamline distribution diagram of clean air inside the vacuum circuit breaker chamber;

[0051] Figure 7 This application discloses a specific gas streamline distribution diagram of a vacuum circuit breaker chamber containing CO2 gas; wherein, Figure 7 (a) is a diagram showing the gas streamline distribution corresponding to the front of the vacuum circuit breaker's gas chamber. Figure 7 (b) is a diagram showing the gas streamline distribution on the side of the vacuum circuit breaker chamber. Figure 7 (c) is a diagram showing the gas streamline distribution behind the vacuum circuit breaker chamber;

[0052] Figure 8 This is a schematic diagram showing the distribution of key external temperature measuring points for three specific vacuum interrupters disclosed in this application;

[0053] Figure 9 This is a schematic diagram of the loss change RMSE curve during the training process of a specific contact temperature inversion calculation model disclosed in this application.

[0054] Figure 10 This is a schematic diagram illustrating the feature importance distribution of a specific contact temperature inversion calculation model disclosed in this application.

[0055] Figure 11 This is a schematic diagram illustrating the effect verification of a specific method disclosed in this application, where test set data is input into a contact temperature inversion calculation model; wherein, Figure 11 (a) is a scatter plot of the true values ​​and the inverted values. Figure 11 (b) is a schematic diagram comparing the true value and the inverted value. Figure 11 (c) is a schematic diagram showing the absolute error between the inverted value and the true value of the contact temperature;

[0056] Figure 12 This is a schematic diagram of the structure of a vacuum interrupter contact temperature inversion device in an environmentally friendly switch disclosed in this application;

[0057] Figure 13 This is a structural diagram of an electronic device disclosed in this application. Detailed Implementation

[0058] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0059] Currently, environmentally friendly GIS switches possess excellent insulation performance and environmentally friendly characteristics. As a core component of environmentally friendly GIS, the operating state of the internal contacts of the vacuum interrupter directly affects the performance of the entire switchgear. Traditional contact-based temperature measurement methods cannot be directly applied to online monitoring of contact temperature. During long-term operation, the contacts heat up due to carrying the load current, and the accumulated heat will significantly impact contact performance and even the mechanical and electrical performance of the entire switchgear, potentially leading to serious malfunctions. Therefore, this application provides a method for inverting the contact temperature of the vacuum interrupter in environmentally friendly switches, which can improve the efficiency of temperature inversion of the contacts within the vacuum interrupter during the process.

[0060] See Figure 1 As shown, this invention discloses a method for inverting the temperature of the vacuum interrupter contact in an environmentally friendly switch, comprising:

[0061] Step S11: Construct a three-dimensional geometric model of the vacuum circuit breaker based on the environmental protection switch, and use the three-dimensional geometric model of the vacuum circuit breaker to perform multi-physics field simulation calculations on the vacuum interrupter to obtain the temperature gradient distribution cloud map on the outer surface of the ceramic shell of the interrupter and the gas streamline distribution map in the gas chamber. Based on the temperature gradient distribution cloud map, the gas streamline distribution map and the preset heat and mass transfer theory, determine the temperature points.

[0062] In this embodiment, the flowchart for temperature inversion of the vacuum interrupter contacts in the environmental protection switch is as follows: Figure 2 As shown: First, key temperature points on the outer surface of the vacuum interrupter that are strongly correlated with the contact temperature are selected. Then, a three-dimensional geometric model of the vacuum circuit breaker in the environmentally friendly GIS equipment is established. The temperature gradient distribution cloud map of the outer surface of the ceramic shell and the gas streamline distribution map of the gas chamber are obtained through multi-physics field simulation calculation. Then, the correlation between the selected points and the contact temperature is analyzed based on the temperature gradient and streamline heat and mass transfer theory, and N1-N2 strongly correlated temperature measurement points are selected.

[0063] It is worth mentioning that, since the heat from the contacts inside the vacuum interrupter is mainly transferred to the external components via thermal conduction along the conductive rod, the contact temperature affects the external temperature distribution of the vacuum interrupter. Under different operating conditions (load current, chamber pressure, and ambient temperature), the steady-state temperature of the contacts corresponds to the temperature of the ceramic shell and main shield of the vacuum interrupter. That is, under different operating conditions, the internal contact temperature can be inverted by measuring the temperatures of key temperature measurement points on the ceramic shell and main shield of the vacuum interrupter. Specifically, key temperature measurement points can be selected based on the temperature gradient on the outer surface of the ceramic shell and the streamline distribution of the chamber. The temperature gradient is a physical quantity used to describe the direction and rate of the fastest temperature change around a specific location. It not only determines the direction of heat transfer but also directly affects the rate of heat transfer. In the process of heat conduction, the temperature gradient is the main driving force for heat transfer within a material; in the process of heat convection, the temperature gradient affects the flow velocity and direction of the fluid, thus affecting the heat transfer efficiency, as shown in the following expression:

[0064] ;

[0065] Among them, the right side of the equation , and Temperature exist , , Spatial derivative in the direction; The sign of α indicates the direction of the fastest temperature change, and its magnitude indicates the maximum rate of temperature change. In other words, in one-dimensional space, its absolute value is the magnitude of the rate of temperature change.

[0066] Furthermore, streamlines are curves formed by different fluid particles at the same moment, where the velocity direction of the fluid particles on the curve is consistent with the tangent direction of the curve. That is, streamlines represent the velocity direction of particles in the flow field at that moment. Conversely, traces represent the velocity direction of the same fluid particle at different moments. When the temperature reaches a steady state, the traces coincide with the streamlines. Therefore, temperature gradient and streamline distribution in the gas chamber are important indicators for selecting temperature measurement points.

[0067] Specifically, a three-dimensional geometric model of a vacuum circuit breaker is constructed based on the environmental protection switch. This model is then used to perform multiphysics simulation calculations on the vacuum interrupter, obtaining a temperature gradient distribution cloud map on the outer surface of the ceramic shell and a gas streamline distribution map within the chamber. This process includes: constructing a three-dimensional geometric model of the vacuum circuit breaker based on the GIS equipment corresponding to the environmental protection switch; determining the ambient temperature of the vacuum interrupter using this model; determining the spatial derivatives of the ambient temperature in each direction; determining the temperature change based on these spatial derivatives; determining the corresponding modulus based on the temperature change; determining the rate of temperature change based on the modulus; setting the rate of temperature change with the largest value among these rates as the target rate of change; determining the temperature gradient distribution cloud map on the outer surface of the ceramic shell of the interrupter using the three-dimensional geometric model of the vacuum circuit breaker and based on the target rate of change; simulating the motion trajectory of fluid particles in the vacuum interrupter at the same moment using the three-dimensional geometric model of the vacuum circuit breaker to obtain a gas streamline distribution map; and using this gas streamline distribution map to reflect the influence of gas flow on heat and mass transfer within the vacuum interrupter.

[0068] In this embodiment, the multiphysics simulation method is used to calculate the temperature gradient distribution cloud map on the outer surface of the ceramic shell of the vacuum interrupter of the environmental protection switch and the gas streamline distribution map inside the gas chamber. Locations with large absolute values ​​of temperature gradients on the outer surface of the ceramic shell and locations with denser streamline distribution inside the gas chamber are selected as key temperature measurement points. Specifically, determining temperature points based on the temperature gradient distribution cloud map, the gas streamline distribution map, and a preset heat and mass transfer theory may include: extracting each temperature gradient from the temperature gradient distribution cloud map and determining whether the absolute value corresponding to each temperature gradient is greater than a preset temperature gradient threshold. If the absolute value corresponding to each temperature gradient is greater than the preset temperature gradient threshold, then the temperature measurement points corresponding to the temperature gradients with absolute values ​​greater than the preset temperature gradient threshold are set as the first temperature measurement points to be screened; extracting the streamlines to be judged corresponding to each of the first temperature measurement points to be screened from the gas streamline distribution map, and determining the streamlines to be judged... Whether the streamline distribution meets the preset streamline distribution conditions. If the streamline distribution corresponding to the streamline to be judged meets the preset streamline distribution conditions, then the first temperature measurement point to be screened corresponding to the streamline distribution that meets the preset streamline distribution conditions is set as the second temperature measurement point to be screened. The preset heat and mass transfer theory is used to screen each second temperature measurement point to be screened to obtain the temperature point. The selected location of the temperature point includes the area of ​​the outer surface of the ceramic shell of the vacuum interrupter near the end cover plate of the stationary guide rod, the outer surface area of ​​the main shield that is in the same plane as the lower surface of the moving contact plate, and the area near the end cover plate of the moving guide rod.

[0069] Step S12: Determine the correlation between the initial contact temperature and the temperature point, and then use the XGBoost algorithm to construct an initial contact temperature inversion calculation model based on the point temperature corresponding to the temperature point, the correlation, and the initial contact temperature.

[0070] In this embodiment, the XGBoost algorithm is used to establish a quantitative calculation model relating the temperature of the vacuum interrupter contact to the temperature of key points. Then, the key temperature measurement points are used as input data, and the contact temperature is used as output data, both forming a training set for model training to obtain the contact temperature inversion calculation model. It is worth noting that the Extreme Gradient Boosting (XGBoost) algorithm has good computational efficiency, memory management, and generalization ability, and supports distributed training and parallel processing, making it suitable for large-scale structured data modeling. Its core principle is to iteratively combine weak learners (decision trees) to gradually correct prediction errors and introduce a regularization mechanism to suppress overfitting. This embodiment trains the model by minimizing the objective function using the XGBoost algorithm. The objective function is derived from the loss function. and regularization term It consists of two parts. This is the actual value, i.e., the simulated value of the contact temperature; The predicted value, i.e., the inverse calculation value of the contact temperature, is expressed as follows:

[0071] ;

[0072] XGBoost then minimizes the loss function by progressively adding trees, with each step improving the prediction based on the previous one. Assuming there are already t-1 trees, the goal of the t-th tree is to minimize the incremental loss function shown below:

[0073] ;

[0074] Subsequently, this embodiment utilizes second-order gradient information (i.e., the Hessian matrix) during the training process of XGBoost, enabling it to more accurately capture the curvature of the loss function, improve optimization efficiency, and accelerate model convergence. The loss function employs a Taylor expansion (second-order approximation) in each iteration to simplify the optimization process, as shown below:

[0075] ;

[0076] in, It is the first derivative; It is the second derivative; This is a constant term and has no effect on minimization.

[0077] Subsequently, the loss function obtained after removing the constant term and simplifying is as follows:

[0078] ;

[0079] The model for each tree in XGBoost can be represented by the following formula:

[0080] ;

[0081] in, The index function for the leaf nodes will take the input... Mapped to leaf nodes ; leaf node The weight.

[0082] Furthermore, the definition expression for the regularization term is as follows:

[0083] ;

[0084] in, This represents the number of leaf nodes; and These are regularization parameters; the former controls the tree complexity, and the latter controls the weight size. For the first The weight of each leaf:

[0085] Furthermore, in this embodiment, the model and regularization term definition expression of each tree are substituted into the simplified loss function after removing the constant term, and the same leaf node (the first...) is... Summing the samples from (each leaf node), For the first The sum of the gradients of the leaf nodes, For the first The sum of the Hessian (second-order gradient) values ​​of the leaf nodes yields the simplified final objective function as shown below:

[0086] ;

[0087] Subsequently, the objective function mentioned above... By taking the derivative and setting it to zero, we obtain the optimal leaf node weight expression:

[0088] ;

[0089] It is worth mentioning that, in the tree construction embodiment of this application, XGBoost selects the optimal split by optimizing the gain of the split point, that is, it selects the split point with the largest gain for splitting, and the expression for the tree split gain Gain is defined as follows:

[0090] ;

[0091] in, and These are the sum of the gradient and Hessian of the left child node, respectively; and These are the gradients and Hessian sums of the right child node, respectively.

[0092] It is worth mentioning that the schematic diagram of the XGBoost algorithm is as follows: Figure 3 As shown, the residuals obtained in each iteration are used to correct the previous predictor to optimize the specified loss function. This involves iteratively building new trees, each improving upon the previous one, ultimately resulting in a powerful model. Specifically, the correlation between the initial contact temperature and the temperature point is determined. Then, the XGBoost algorithm is used to construct an initial contact temperature inversion calculation model based on the corresponding temperature point, the correlation, and the initial contact temperature. This can include: determining the correlation between the initial contact temperature and the temperature point using preset correlation rules; then using the XGBoost algorithm and the prediction residuals from the previous iteration of the initial contact temperature inversion calculation model to construct a new decision tree; and finally, using the Hessian matrix corresponding to the correlation to perform a Taylor expansion to obtain the optimized tree structure. The weights corresponding to the leaf nodes in the optimization tree structure are determined, and the initial contact temperature inversion calculation model is evaluated based on the weights and the split gain of the optimization tree structure. The evaluation results are then used to determine the loss function. The regularization term is determined based on the number of leaf nodes in the optimization tree structure, the weight of each leaf node, and the preset regularization parameters. The objective function is then determined based on the loss function and the regularization term. Finally, the parameters of the initial contact temperature inversion calculation model are adjusted using the objective function and based on the point temperature corresponding to the temperature point, the correlation relationship, and the initial contact temperature to obtain a new initial contact temperature inversion calculation model.

[0093] Step S13: Adjust the initial hyperparameters corresponding to the initial contact temperature inversion calculation model using the grid search method and cross-validation method to obtain the target hyperparameters, and then use the target hyperparameters to adjust the initial contact temperature inversion calculation model to obtain the target contact temperature inversion calculation model.

[0094] In this embodiment, hyperparameter optimization of XGBoost is one of the key steps to improve model performance. Proper hyperparameter tuning can significantly improve the training effect. In one specific implementation, commonly used XGBoost hyperparameters are shown in Table 1:

[0095] Table 1. XGBoost Core Hyperparameter Optimization Framework

[0096]

[0097] Specifically, adjusting the initial contact temperature inversion calculation model using target hyperparameters to obtain the target contact temperature inversion calculation model can include: adjusting the model parameters corresponding to the initial contact temperature inversion calculation model using target hyperparameters to obtain the target contact temperature inversion calculation model; wherein, the learning rate is used to control the contribution of a single tree; the number of decision trees is used to determine the model capacity; the maximum tree depth is used to limit the tree complexity; the minimum sum of leaf weights is used to control the number of samples for node splits; the split threshold is used to suppress low-gain splits; L2 regularization is used to constrain the weight amplitude; and L1 regularization is used to promote feature sparsity.

[0098] Subsequently, this embodiment employs a combination of Grid Search (GS) and Cross-Validation (CV) for parameter optimization. The specific hyperparameter optimization strategy is as follows: First, hyperparameters affecting the model's learning ability, such as learning_rate and n_estimators, are adjusted. It is understood that a smaller learning rate typically requires more trees to achieve sufficient learning. Next, GS search is used to find parameter combinations, and CV validation is used to evaluate model performance, ultimately determining the optimal parameters. Then, max_depth and min_child_weight are adjusted using the same parameter optimization method to control tree complexity. Finally, gamma, lambda, and alpha parameters are adjusted to avoid overfitting.

[0099] Specifically, the initial hyperparameters of the initial contact temperature inversion calculation model are adjusted using grid search and cross-validation to obtain target hyperparameters. These target hyperparameters are then used to adjust the initial contact temperature inversion calculation model to obtain a target contact temperature inversion calculation model. This process includes: determining the first hyperparameter affecting the model's learning ability among the initial hyperparameters of the initial contact temperature inversion calculation model; combining the first hyperparameter using grid search and cross-validation to obtain the first target hyperparameter; the first hyperparameter includes the learning rate and the number of decision trees; adjusting the second hyperparameter among the initial hyperparameters using a preset tree complexity condition to obtain the second target hyperparameter; then adjusting the third hyperparameter among the initial hyperparameters using a preset overfitting condition to obtain the third target hyperparameter; and finally determining the target hyperparameter based on the first, second, and third target hyperparameters; where the second hyperparameter includes the maximum tree depth and the minimum sum of leaf weights; and the third hyperparameter includes a splitting threshold, L2 regularization, and L1 regularization.

[0100] Step S14: Using the target contact temperature inversion calculation model and based on the load current, chamber pressure, ambient temperature, correlation relationship, and initial contact temperature of the vacuum interrupter, perform contact temperature inversion operation to obtain the target contact temperature.

[0101] In this embodiment, an optimized XGBoost correlation quantization calculation model is used to calculate and invert the contact temperature. Subsequently, this embodiment can scientifically evaluate the merits of the optimized contact temperature inversion calculation model and the accuracy of the inverted contact temperature through error analysis. In one specific implementation, the error evaluation indicators selected in this embodiment are shown in Table 2:

[0102] Table 2. Error Evaluation Indicators for XGBoost Model

[0103]

[0104] in, The number of samples; The actual value; These are predicted values.

[0105] Specifically, the target contact temperature is obtained by using a target contact temperature inversion calculation model and performing a contact temperature inversion operation based on the load current, chamber pressure, ambient temperature, correlation, and initial contact temperature of the vacuum interrupter. This process may include: determining the load current, chamber pressure, and ambient temperature corresponding to the vacuum interrupter; performing a contact temperature inversion operation based on the load current, chamber pressure, ambient temperature, correlation, and initial contact temperature using a target contact temperature inversion calculation model to obtain the target contact temperature; and evaluating the target contact temperature output by the target contact temperature inversion calculation model using an error evaluation index to obtain the evaluation result. As a result, the evaluation results were used to adjust the model parameters of the target contact temperature inversion calculation model, resulting in a new target contact temperature inversion calculation model. The error evaluation indicators included root mean square error (RMSE), mean absolute error (MAE), mean absolute percentage error (MAS%), and maximum absolute error (MAE). RMSE was used to quantify the standard deviation between the predicted and actual values. MAE was used to calculate the absolute average of the prediction error. MAS% was used to evaluate the mean percentage of the relative error between the predicted and actual values. MAE was used to identify the most severe deviation in the prediction results. The predicted value was the target contact temperature, and the actual value was the initial contact temperature.

[0106] In one specific embodiment, this application takes the vacuum interrupter chamber of a 252kV environmentally friendly switch as the research object, and models it at a 1:1 scale. The schematic diagram of the three-dimensional geometric model of the vacuum circuit breaker chamber is shown below. Figure 4 As shown, where, Figure 4 101 in the text refers to an insulating support platform. Figure 4 102 in the figure is the stationary end support. Figure 4 103 in the figure is the conductive rod at the moving end of the air chamber. Figure 4 104 in the text refers to a flange. Figure 4The 105 in the text refers to a 252kV single-break vacuum interrupter. Figure 4 106 in the text refers to the spring-loaded finger. Figure 4 107 in the figure is the conductive rod at the stationary end of the air chamber. Figure 4 108 in the figure is an insulating tie rod. Figure 4 109 in the text refers to the control transmission mechanism. Figure 4 110 in the middle is the moving end support. Figure 4 111 in the text refers to the insulating support platform. Figure 4 112 in the figure represents the cover plate. Then, electromagnetic-thermal flow field calculations are performed based on the geometric model. The insulating gases in the chamber are clean air with an absolute pressure of 0.7 MPa and CO2 gas with an absolute pressure of 0.5 MPa, respectively. The temperature gradient distribution cloud map along the gravity direction on the outer surface of the ceramic shell of the arc-extinguishing chamber is shown in the figure. Figure 5 As shown: Among them, Figure 5 (a) Temperature gradient distribution cloud map corresponding to clean air at 0.7 MPa. Figure 5 (b) Temperature gradient distribution cloud map corresponding to CO2 gas at 0.5 MPa.

[0107] In this embodiment, under the condition that the gas chamber is filled with different insulating gases, the absolute values ​​of the temperature gradients on the outer surfaces of the ceramic shells at the moving and stationary ends of the arc-extinguishing chamber both increase as they move away from the main shield and closer to the end cover plate of the arc-extinguishing chamber. Therefore, it can be considered that the temperature change in the region of the ceramic shell near the end cover plate of the arc-extinguishing chamber is more sensitive. The gas streamline vector distribution diagram and gas streamline distribution diagram of the clean air inside the vacuum circuit breaker gas chamber are shown in the figure. Figure 6 (a) and Figure 6 As shown in (b), Figure 6 The left image in (a) is a vector diagram showing the gas streamlines of clean air corresponding to the side of the vacuum circuit breaker chamber. Figure 6 The right image in (a) is a vector diagram showing the gas streamlines of clean air behind the vacuum circuit breaker chamber; Figure 6 The left image in (b) is a diagram showing the gas streamline distribution of clean air corresponding to the side of the vacuum circuit breaker chamber. Figure 6 The right image in (b) shows the gas streamline distribution of clean air corresponding to the rear of the vacuum circuit breaker chamber; in addition, the gas streamline distribution of CO2 gas inside the vacuum circuit breaker chamber is shown in the image below. Figure 7 As shown, where, Figure 7 (a) is a diagram showing the gas streamline distribution corresponding to the front of the vacuum circuit breaker's gas chamber. Figure 7 (b) is a diagram showing the gas streamline distribution on the side of the vacuum circuit breaker chamber. Figure 7(c) is a diagram of the gas streamline distribution behind the gas chamber of the vacuum circuit breaker. As can be seen from the figure, the streamline distribution is relatively dense on the outer surface of the ceramic shell of the arc-extinguishing chamber near the end cover plate. According to the streamline heat and mass transfer theory, this area can be considered to be closely related to heat transfer and flow.

[0108] Therefore, in this embodiment, key temperature measurement points should be selected on the outer surface of the ceramic shell of the arc-extinguishing chamber near the end cover plate, such as... Figure 8 As shown in T1 and T3. Simultaneously, a key temperature measurement point should be selected on the outer surface of the main shield, which is on the same plane as the lower surface of the moving contact piece. This location is closest to the hot spot temperature of the contact inside the arc-extinguishing chamber. For example... Figure 8 As shown in T2, this is the location most affected by the thermal radiation from the contacts. That is, the three key external temperature measurement points selected in this embodiment of the vacuum interrupter are as follows: Figure 8 As shown, these three key temperature measurement points, T1, T2, and T3, have a strong correlation with the internal contact temperature of the vacuum interrupter. Their corresponding locations are as follows: T1: The area on the outer surface of the ceramic shell of the interrupter near the end cover of the stationary guide rod. T2: The area on the outer surface of the main shield, which is on the same plane as the lower surface of the moving contact piece. T3: The area on the outer surface of the ceramic shell of the interrupter near the end cover of the moving guide rod.

[0109] Subsequently, this embodiment of the application requires the use of the XGBoost algorithm to establish a quantitative calculation model for the correlation between the contact temperature of the vacuum interrupter and the temperature of key points. That is, the load current, gas chamber pressure, ambient temperature, and key temperature measurement points T1, T2, and T3 outside the vacuum interrupter are used as input data, and the corresponding steady-state temperature of the vacuum interrupter contact is used as output data to establish a training set and a test set for the contact temperature inversion calculation model. Then, the XGBoost algorithm is used to train the model, thereby outputting the number of iterations and defining the loss function. Then, the gradient and Hessian are calculated. Then, the loss function is updated while building the tree. If the iteration requirements are met, the process ends; otherwise, it is recalculated. In this way, a 252kV environmentally friendly GIS vacuum interrupter contact temperature inversion calculation model is constructed.

[0110] In one specific implementation, a set of data on load current (0~4400 A), gas chamber pressure (0~0.8 MPa), ambient temperature (-8~32 ℃), and corresponding key temperature measurement points T1, T2, and T3 outside the arc-extinguishing chamber are used as inputs, and the corresponding steady-state temperature data of the vacuum arc-extinguishing chamber contacts are used as outputs, forming a sample dataset of 144 sets of data. Then, the sample dataset is divided into a training set and a test set using a random number extraction method, with 116 sets of data used as the training set and 28 sets of data used as the test set.

[0111] Subsequently, the embodiments of this application can optimize the hyperparameters of the XGBoost algorithm model based on the parameter optimization strategy described above, and finally obtain the optimal combination of model hyperparameters, as shown in Table 3.

[0112] Table 3. Schematic diagram of hyperparameter combinations for the XGBoost model

[0113]

[0114] In this embodiment, the loss change RMSE curve during the training process of the contact temperature inversion calculation model is as follows: Figure 9 As shown, the RMSE curves of the training set and the test set steadily decrease, indicating that the model training is good and the performance is gradually improving.

[0115] It is worth mentioning that the feature importance distribution diagram of the contact temperature inversion calculation model is as follows: Figure 10 As shown in the figure, the key temperature measurement point T1 has the greatest impact on the inversion results of the inversion calculation model; the load current and the air chamber pressure have a significant impact; while the ambient temperature has a relatively small impact on the inversion results of the inversion calculation model.

[0116] Subsequently, the test set data is input into the contact temperature inversion calculation model to verify the effect, as shown in the embodiment of this application. Figure 11 As shown, in the 28 test sets, the scatter plots and comparison plots of the true values ​​and inverted values ​​are as follows: Figure 11 (a) and Figure 11 As shown in (b), the inverted values ​​are all close to the true values ​​on the diagonal; the absolute error between the inverted value and the true value of the contact temperature is as follows: Figure 11 As shown in (c), the maximum absolute error between the inverted value and the true value is 3.3 ℃, the average absolute error is 1.9 ℃, and the average absolute percentage error is 7.7%, which is less than 10%. This indicates that the model fits the nonlinear relationship between the input and output variables very well and has good temperature inversion accuracy, thus verifying the effectiveness of the temperature inversion calculation model for the vacuum interrupter contact proposed in this application embodiment.

[0117] As can be seen from the above, the embodiments of this application first need to construct a three-dimensional geometric model of the vacuum circuit breaker based on the environmental protection switch, so as to use the three-dimensional geometric model of the vacuum circuit breaker to perform multi-physics field simulation calculations on the vacuum interrupter, obtain the temperature gradient distribution cloud map on the outer surface of the ceramic shell of the interrupter and the gas streamline distribution map in the gas chamber, and determine the temperature points based on the temperature gradient distribution cloud map, the gas streamline distribution map and the preset heat and mass transfer theory; then, determine the correlation between the initial contact temperature of the contacts and the temperature points, and then use the XGBoost algorithm and the point temperature corresponding to the temperature point, A correlation relationship and initial contact temperature are used to construct an initial contact temperature inversion calculation model. Then, the initial hyperparameters corresponding to the initial contact temperature inversion calculation model are adjusted using grid search and cross-validation methods to obtain target hyperparameters. These target hyperparameters are then used to adjust the initial contact temperature inversion calculation model to obtain a target contact temperature inversion calculation model. Finally, using the target contact temperature inversion calculation model and based on the load current, chamber pressure, ambient temperature, correlation relationship, and initial contact temperature of the vacuum interrupter, a contact temperature inversion operation is performed to obtain the target contact temperature. This improves the efficiency of temperature inversion for the vacuum interrupter contacts in environmental protection switches, thereby enhancing the user experience.

[0118] Accordingly, see Figure 12 As shown, this application also provides a device for retrieving the temperature of the vacuum interrupter contact in an environmentally friendly switch, comprising:

[0119] The temperature point determination module is used to construct a three-dimensional geometric model of the vacuum circuit breaker based on the environmental protection switch, and to use the three-dimensional geometric model of the vacuum circuit breaker to perform multi-physics field simulation calculations on the vacuum interrupter to obtain the temperature gradient distribution cloud map on the outer surface of the ceramic shell of the interrupter and the gas streamline distribution map in the gas chamber, and to determine the temperature point based on the temperature gradient distribution cloud map, the gas streamline distribution map and the preset heat and mass transfer theory.

[0120] The correlation determination module is used to determine the correlation between the initial contact temperature of the contact and the temperature point, and then use the XGBoost algorithm to construct an initial contact temperature inversion calculation model based on the point temperature corresponding to the temperature point, the correlation, and the initial contact temperature.

[0121] The hyperparameter adjustment module is used to adjust the initial hyperparameters corresponding to the initial contact temperature inversion calculation model using the grid search method and cross-validation method to obtain the target hyperparameters, and then use the target hyperparameters to adjust the initial contact temperature inversion calculation model to obtain the target contact temperature inversion calculation model.

[0122] The contact temperature determination module is used to perform contact temperature inversion operation based on the target contact temperature inversion calculation model and the load current, chamber pressure, ambient temperature, correlation relationship and initial contact temperature of the vacuum interrupter to obtain the target contact temperature.

[0123] In some specific embodiments, the temperature point determination module 11 may specifically include:

[0124] The spatial derivative determination unit is used to construct a three-dimensional geometric model of the vacuum circuit breaker based on the GIS equipment corresponding to the environmental protection switch, and to determine the ambient temperature of the vacuum interrupter using the three-dimensional geometric model of the vacuum circuit breaker, and then determine the spatial derivative of the ambient temperature in each direction.

[0125] The temperature change determination unit is used to determine the temperature change based on each of the spatial derivatives, determine the corresponding modulus based on the temperature change, determine the temperature change rate based on the modulus, and then set the temperature change rate with the largest value among the temperature change rates as the target change rate.

[0126] The temperature gradient distribution cloud map determination unit is used to determine the temperature gradient distribution cloud map of the outer surface of the ceramic shell of the arc-extinguishing chamber based on the three-dimensional geometric model of the vacuum circuit breaker and the target rate of change.

[0127] The gas streamline distribution diagram determination unit is used to simulate the motion trajectory of fluid particles in the vacuum interrupter at the same moment using the three-dimensional geometric model of the vacuum circuit breaker, and obtain the gas streamline distribution diagram; the gas streamline distribution diagram is used to reflect the influence of gas flow on heat and mass transfer in the vacuum interrupter.

[0128] In some specific embodiments, the temperature point determination module 11 may specifically include:

[0129] The temperature gradient extraction unit is used to extract each temperature gradient from the temperature gradient distribution cloud map and determine whether the absolute value corresponding to each temperature gradient is greater than a preset temperature gradient threshold. If the absolute value corresponding to each temperature gradient is greater than the preset temperature gradient threshold, then the temperature measurement points corresponding to the temperature gradients whose absolute values ​​are greater than the preset temperature gradient threshold are set as the first temperature measurement points to be screened.

[0130] The streamline extraction unit is used to extract the streamlines to be judged corresponding to each of the first temperature measurement points to be screened from the gas streamline distribution map, and to determine whether the streamline distribution corresponding to the streamline to be judged meets the preset streamline distribution conditions. If the streamline distribution corresponding to the streamline to be judged meets the preset streamline distribution conditions, the first temperature measurement point to be screened corresponding to the streamline distribution that meets the preset streamline distribution conditions is set as the second temperature measurement point to be screened.

[0131] The temperature point determination subunit is used to screen each of the second temperature measurement points to be screened using a preset heat and mass transfer theory to obtain the temperature points; wherein, the selected locations of the temperature points include the area of ​​the outer surface of the ceramic shell of the vacuum interrupter near the end cover plate of the stationary guide rod, the outer surface area of ​​the main shield that is in the same plane as the lower surface of the moving contact plate, and the area near the end cover plate of the moving guide rod.

[0132] In some specific embodiments, the association determination module 12 may specifically include:

[0133] An optimized tree structure generation unit is used to determine the correlation between the initial contact temperature of the contact and the temperature point using a preset correlation relationship determination rule. Then, a new decision tree is constructed using the XGBoost algorithm and the prediction residual of the previous iteration of the initial contact temperature inversion calculation model. Finally, a Taylor expansion is performed using the Hessian matrix corresponding to the correlation relationship to obtain the optimized tree structure.

[0134] The loss function determination unit is used to determine the weights corresponding to the leaf nodes in the optimization tree structure, and to evaluate the initial contact temperature inversion calculation model based on the weights and the split gain of the optimization tree structure to obtain the evaluation result, and to determine the loss function based on the evaluation result.

[0135] The regularization term determination unit is used to determine the regularization term based on the number of leaf nodes in the optimization tree structure, the weight of each leaf node, and the preset regularization parameters. The objective function is determined based on the loss function and the regularization term. Then, the parameters of the initial contact temperature inversion calculation model are adjusted using the objective function and based on the point temperature corresponding to the temperature point, the correlation relationship, and the initial contact temperature to obtain a new initial contact temperature inversion calculation model.

[0136] In some specific embodiments, the hyperparameter adjustment module 13 may specifically include:

[0137] The parameter combination unit is used to determine the first hyperparameter affecting the model's learning ability among the initial hyperparameters corresponding to the initial contact temperature inversion calculation model, and to combine the first hyperparameter using grid search and cross-validation methods to obtain the first target hyperparameter; the first hyperparameter includes the learning rate and the number of decision trees;

[0138] The hyperparameter tuning subunit is used to adjust the second hyperparameter in the initial hyperparameters using a preset tree complexity condition to obtain a second target hyperparameter, and then adjust the third hyperparameter in the initial hyperparameters using a preset overfitting condition to obtain a third target hyperparameter, and determine the target hyperparameter based on the first target hyperparameter, the second target hyperparameter, and the third target hyperparameter; wherein, the second hyperparameter includes the maximum tree depth and the minimum sum of leaf weights; the third hyperparameter includes a split threshold, L2 regularization, and L1 regularization.

[0139] In some specific embodiments, the hyperparameter adjustment module 13 may specifically include:

[0140] The model parameter adjustment unit is used to adjust the model parameters corresponding to the initial contact temperature inversion calculation model using the target hyperparameters to obtain the target contact temperature inversion calculation model; wherein, the learning rate is used to control the contribution of a single tree; the number of decision trees is used to determine the model capacity; the maximum tree depth is used to limit the tree complexity; the minimum leaf weight is used to control the number of samples for node splitting; the splitting threshold is used to suppress low-gain splitting; the L2 regularization is used to constrain the weight amplitude; and the L1 regularization is used to promote feature sparsity.

[0141] In some specific embodiments, the contact temperature determination module 14 may specifically include:

[0142] The contact temperature inversion unit is used to determine the load current, gas chamber pressure and ambient temperature corresponding to the vacuum interrupter, and to perform contact temperature inversion operation based on the target contact temperature inversion calculation model and the load current, the gas chamber pressure, the ambient temperature, the correlation and the initial contact temperature to obtain the target contact temperature.

[0143] The evaluation result generation unit is used to evaluate the target contact temperature output by the target contact temperature inversion calculation model using error evaluation indicators, obtain evaluation results, and adjust the model parameters of the target contact temperature inversion calculation model using the evaluation results to obtain a new target contact temperature inversion calculation model. The error evaluation indicators include root mean square error, mean absolute error, mean absolute percentage error, and maximum absolute error. The root mean square error is used to quantify the standard deviation between the predicted value and the true value. The mean absolute error is used to calculate the absolute average of the prediction error. The mean absolute percentage error is used to evaluate the mean percentage of the relative error between the predicted value and the true value. The maximum absolute error is used to identify the most severe deviation in the prediction results. The predicted value is the target contact temperature; the true value is the initial contact temperature.

[0144] Furthermore, embodiments of this application also disclose an electronic device, Figure 13 This is a structural diagram of an electronic device 20 according to an exemplary embodiment. The content of the diagram should not be construed as limiting the scope of this application. The electronic device 20 may specifically include: at least one processor 21, at least one memory 22, a power supply 23, a communication interface 24, an input / output interface 25, and a communication bus 26. The memory 22 stores a computer program, which is loaded and executed by the processor 21 to implement the relevant steps in the vacuum interrupter contact temperature inversion method of the environmentally friendly switch disclosed in any of the foregoing embodiments. Alternatively, the electronic device 20 in this embodiment may specifically be an electronic computer.

[0145] In this embodiment, the power supply 23 is used to provide operating voltage for each hardware device on the electronic device 20; the communication interface 24 can create a data transmission channel between the electronic device 20 and external devices, and the communication protocol it follows can be any communication protocol applicable to the technical solution of this application, and is not specifically limited here; the input / output interface 25 is used to acquire external input data or output data to the outside world, and its specific interface type can be selected according to specific application needs, and is not specifically limited here.

[0146] In addition, the memory 22, as a carrier for resource storage, can be a read-only memory, random access memory, disk or optical disk, etc. The resources stored thereon can include operating system 221, computer program 222, etc., and the storage method can be temporary storage or permanent storage.

[0147] The operating system 221 is used to manage and control the various hardware devices on the electronic device 20 and the computer program 222, which may be Windows Server, Netware, Unix, Linux, etc. In addition to including a computer program capable of performing the vacuum interrupter contact temperature inversion method in the environmentally friendly switch executed by the electronic device 20 as disclosed in any of the foregoing embodiments, the computer program 222 may further include computer programs capable of performing other specific tasks.

[0148] Furthermore, this application also discloses a computer-readable storage medium for storing a computer program; wherein, when the computer program is executed by a processor, it implements the aforementioned method for inverting the temperature of the vacuum interrupter contacts in an environmentally friendly switch. Specific steps of this method can be found in the corresponding content disclosed in the foregoing embodiments, and will not be repeated here.

[0149] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the apparatus disclosed in the embodiments, since it corresponds to the method disclosed in the embodiments, the description is relatively simple; relevant parts can be referred to in the method section.

[0150] Those skilled in the art will further recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0151] The steps of the methods or algorithms described in conjunction with the embodiments disclosed herein can be implemented directly by hardware, a software module executed by a processor, or a combination of both. The software module can be located in random access memory (RAM), main memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, removable disk, CD-ROM, or any other form of storage medium known in the art.

[0152] Finally, it should be noted that in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0153] The technical solutions provided in this application have been described in detail above. Specific examples have been used to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the methods and core ideas of this application. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of this application. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A method for inverting the temperature of the vacuum interrupter contacts in an environmentally friendly switch, characterized in that, include: Based on the aforementioned environmentally friendly switch, a three-dimensional geometric model of a vacuum circuit breaker is constructed. This model is then used to perform multi-physics simulation calculations on the vacuum interrupter chamber, resulting in a temperature gradient distribution cloud map on the outer surface of the ceramic shell of the interrupter chamber and a gas streamline distribution map within the chamber. Based on the temperature gradient distribution cloud map, the gas streamline distribution map, and a preset heat and mass transfer theory, the temperature points are determined. The correlation between the initial contact temperature and the temperature point is determined. Then, the XGBoost algorithm is used to construct an initial contact temperature inversion calculation model based on the point temperature corresponding to the temperature point, the correlation, and the initial contact temperature. The initial hyperparameters corresponding to the initial contact temperature inversion calculation model are adjusted using the grid search method and cross-validation method to obtain the target hyperparameters. The target hyperparameters are then used to adjust the initial contact temperature inversion calculation model to obtain the target contact temperature inversion calculation model. The target contact temperature is obtained by using the target contact temperature inversion calculation model and performing a contact temperature inversion operation based on the load current of the vacuum interrupter, the gas chamber pressure, the ambient temperature, the correlation relationship, and the initial contact temperature.

2. The method for inverting the temperature of the vacuum interrupter contact in the environmentally friendly switch according to claim 1, characterized in that, The method involves constructing a three-dimensional geometric model of the vacuum circuit breaker based on the environmentally friendly switch, and then using this model to perform multiphysics simulation calculations on the vacuum interrupter chamber. This yields a temperature gradient distribution cloud map on the outer surface of the ceramic shell of the interrupter chamber and a gas streamline distribution map within the chamber, including: A three-dimensional geometric model of the vacuum circuit breaker is constructed based on the GIS equipment corresponding to the environmental protection switch. The ambient temperature of the vacuum interrupter is determined using the three-dimensional geometric model of the vacuum circuit breaker, and then the spatial derivative of the ambient temperature in each direction is determined. The temperature change is determined based on the spatial derivatives of each of the above, the corresponding modulus is determined based on the temperature change, and the temperature change rate is determined based on the modulus. Then, the rate of change with the largest value among the temperature change rates is set as the target rate of change. The temperature gradient distribution cloud map on the outer surface of the ceramic shell of the arc-extinguishing chamber is determined using the three-dimensional geometric model of the vacuum circuit breaker and based on the target rate of change. The motion trajectory of fluid particles in the vacuum interrupter at the same moment is simulated using the three-dimensional geometric model of the vacuum circuit breaker to obtain a gas streamline distribution diagram; the gas streamline distribution diagram is used to reflect the influence of gas flow on heat and mass transfer in the vacuum interrupter.

3. The method for inverting the temperature of the vacuum interrupter contact in the environmentally friendly switch according to claim 1, characterized in that, The determination of temperature points based on the temperature gradient distribution cloud map, the gas streamline distribution map, and the preset heat and mass transfer theory includes: Extract each temperature gradient from the temperature gradient distribution cloud map and determine whether the absolute value corresponding to each temperature gradient is greater than a preset temperature gradient threshold. If the absolute value corresponding to each temperature gradient is greater than the preset temperature gradient threshold, then set each temperature measurement point corresponding to the temperature gradient whose absolute value is greater than the preset temperature gradient threshold as the first temperature measurement point to be screened. Extract the streamlines to be judged corresponding to each of the first temperature measurement points to be screened from the gas streamline distribution map, and determine whether the streamline distribution corresponding to the streamline to be judged meets the preset streamline distribution conditions. If the streamline distribution corresponding to the streamline to be judged meets the preset streamline distribution conditions, then set the first temperature measurement point to be screened corresponding to the streamline distribution that meets the preset streamline distribution conditions as the second temperature measurement point to be screened. The second temperature measurement points to be screened are selected using a preset heat and mass transfer theory to obtain temperature points; wherein, the selected temperature points include the area of ​​the outer surface of the ceramic shell of the vacuum interrupter near the end cover plate of the stationary guide rod, the area of ​​the outer surface of the main shield that is in the same plane as the lower surface of the moving contact plate, and the area near the end cover plate of the moving guide rod.

4. The method for inverting the temperature of the vacuum interrupter contact in the environmentally friendly switch according to claim 1, characterized in that, The process involves determining the correlation between the initial contact temperature and the temperature point, then using the XGBoost algorithm and constructing an initial contact temperature inversion calculation model based on the point temperature corresponding to the temperature point, the correlation, and the initial contact temperature. This includes: The correlation between the initial contact temperature and the temperature point is determined by using a preset correlation relationship determination rule. Then, a new decision tree is constructed using the XGBoost algorithm and the prediction residual of the previous iteration of the initial contact temperature inversion calculation model. The Hessian matrix corresponding to the correlation relationship is used to perform Taylor expansion to obtain the optimized tree structure. The weights corresponding to the leaf nodes in the optimization tree structure are determined, and the initial contact temperature inversion calculation model is evaluated based on the weights and the split gain of the optimization tree structure to obtain the evaluation results. The loss function is then determined based on the evaluation results. The regularization term is determined based on the number of leaf nodes in the optimized tree structure, the weight of each leaf node, and the preset regularization parameter. The objective function is then determined based on the loss function and the regularization term. The parameters of the initial contact temperature inversion calculation model are then adjusted using the objective function and based on the point temperature corresponding to the temperature point, the correlation relationship, and the initial contact temperature to obtain a new initial contact temperature inversion calculation model.

5. The method for inverting the temperature of the vacuum interrupter contact in the environmentally friendly switch according to claim 1, characterized in that, The process of adjusting the initial hyperparameters of the initial contact temperature inversion calculation model using grid search and cross-validation to obtain target hyperparameters, and then using these target hyperparameters to adjust the initial contact temperature inversion calculation model to obtain a target contact temperature inversion calculation model, includes: The first hyperparameter affecting the model's learning ability is determined from the initial hyperparameters corresponding to the initial contact temperature inversion calculation model. The first hyperparameter is then combined using grid search and cross-validation methods to obtain the first target hyperparameter. The first hyperparameter includes the learning rate and the number of decision trees. The second hyperparameter in the initial hyperparameters is adjusted using a preset tree complexity condition to obtain a second target hyperparameter. Then, the third hyperparameter in the initial hyperparameters is adjusted using a preset overfitting condition to obtain a third target hyperparameter. The target hyperparameter is determined based on the first target hyperparameter, the second target hyperparameter, and the third target hyperparameter. The second hyperparameter includes the maximum tree depth and the minimum sum of leaf weights. The third hyperparameter includes a split threshold, L2 regularization, and L1 regularization.

6. The method for inverting the temperature of the vacuum interrupter contact in the environmentally friendly switch according to claim 5, characterized in that, The step of adjusting the initial contact temperature inversion calculation model using the target hyperparameters to obtain the target contact temperature inversion calculation model includes: The model parameters corresponding to the initial contact temperature inversion calculation model are adjusted using the target hyperparameters to obtain the target contact temperature inversion calculation model; wherein, the learning rate is used to control the contribution of a single tree; the number of decision trees is used to determine the model capacity; the maximum tree depth is used to limit the tree complexity; the minimum leaf weight is used to control the number of samples for node splitting; the splitting threshold is used to suppress low-gain splitting; the L2 regularization is used to constrain the weight amplitude; and the L1 regularization is used to promote feature sparsity.

7. The method for inverting the temperature of the vacuum interrupter contact in the environmentally friendly switch according to claim 1, characterized in that, The step of using the target contact temperature inversion calculation model and performing a contact temperature inversion operation based on the load current, chamber pressure, ambient temperature, correlation, and initial contact temperature of the vacuum interrupter to obtain the target contact temperature includes: The load current, chamber pressure, and ambient temperature corresponding to the vacuum interrupter are determined, and the target contact temperature is obtained by using the target contact temperature inversion calculation model and performing contact temperature inversion operation based on the load current, chamber pressure, ambient temperature, correlation, and initial contact temperature. The target contact temperature output by the target contact temperature inversion calculation model is evaluated using an error evaluation index to obtain an evaluation result. The model parameters of the target contact temperature inversion calculation model are then adjusted using the evaluation result to obtain a new target contact temperature inversion calculation model. The error evaluation indicators include root mean square error, mean absolute error, mean absolute percentage error, and maximum absolute error; the root mean square error is used to quantify the standard deviation between the predicted value and the true value; the mean absolute error is used to calculate the absolute average of the prediction error; the mean absolute percentage error is used to evaluate the mean of the relative error percentage between the predicted value and the true value; the maximum absolute error is used to identify the most severe deviation in the prediction result; the predicted value is the target contact temperature; and the true value is the initial contact temperature.

8. A device for retrieving the temperature of the vacuum interrupter contact in an environmentally friendly switch, characterized in that, include: The temperature point determination module is used to construct a three-dimensional geometric model of the vacuum circuit breaker based on the environmental protection switch, and to use the three-dimensional geometric model of the vacuum circuit breaker to perform multi-physics field simulation calculations on the vacuum interrupter to obtain the temperature gradient distribution cloud map on the outer surface of the ceramic shell of the interrupter and the gas streamline distribution map in the gas chamber, and to determine the temperature point based on the temperature gradient distribution cloud map, the gas streamline distribution map and the preset heat and mass transfer theory. The correlation determination module is used to determine the correlation between the initial contact temperature of the contact and the temperature point, and then use the XGBoost algorithm to construct an initial contact temperature inversion calculation model based on the point temperature corresponding to the temperature point, the correlation, and the initial contact temperature. The hyperparameter adjustment module is used to adjust the initial hyperparameters corresponding to the initial contact temperature inversion calculation model using the grid search method and cross-validation method to obtain the target hyperparameters, and then use the target hyperparameters to adjust the initial contact temperature inversion calculation model to obtain the target contact temperature inversion calculation model. The contact temperature determination module is used to perform contact temperature inversion operation based on the target contact temperature inversion calculation model and the load current, chamber pressure, ambient temperature, correlation relationship and initial contact temperature of the vacuum interrupter to obtain the target contact temperature.

9. An electronic device, characterized in that, include: Memory, used to store computer programs; A processor is configured to execute the computer program to implement the vacuum interrupter contact temperature inversion method in the environmentally friendly switch as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, Used to store a computer program, wherein the computer program, when executed by a processor, implements the vacuum interrupter contact temperature inversion method in the environmentally friendly switch as described in any one of claims 1 to 7.

Images