ZnO piezoresistor lightning impulse temperature rise prediction method and device, terminal equipment and storage medium

Abstract

The invention discloses a ZnO varistor lightning impulse temperature rise prediction method and device, terminal equipment and a storage medium, and belongs to the technical field of varistors, and the method comprises the steps: obtaining lightning impulse parameters and microstructure parameters of a to-be-measured ZnO varistor; wherein the microstructure parameters comprise the grain size, the grain boundary layer thickness and the microcosmic uniformity; the microcosmic uniformity is used for quantifying the difference degree of the grain sizes; and inputting the lightning impulse parameters and the microstructure parameters into the trained varistor temperature rise prediction model for prediction, and outputting the maximum surface temperature rise value of the ZnO varistor to be measured. According to the method, the influence of three types of microstructure parameters including the grain size, the grain boundary layer thickness and the microcosmic uniformity on the lightning impulse temperature rise is comprehensively considered, and therefore the problem that in the prior art, due to the fact that the microcosmic influence of the microstructure parameters is neglected, ZnO piezoresistor temperature rise prediction is not accurate is solved.

Description

Technical Field

[0001] The present invention relates to the technical field of varistors, and in particular to a method, device, terminal equipment and storage medium for predicting the lightning impulse temperature rise of a ZnO varistor. Background Art

[0002] As an important nonlinear resistor component, ZnO varistors are widely used in overvoltage protection of power systems, especially in surge protection devices (SPDs). Their excellent nonlinear characteristics can effectively limit overvoltages and protect electrical equipment from damage caused by lightning overvoltages and switching overvoltages. However, ZnO varistors are easily damaged by thermal effects under lightning strikes, which in turn affects their protective performance and can even cause safety accidents such as fires. Therefore, accurately predicting the temperature rise characteristics of ZnO varistors under lightning strikes is extremely important for optimizing their design, improving their reliability, and ensuring the safe operation of power systems.

[0003] The temperature rise characteristics of ZnO varistors are not only affected by lightning impulse parameters but are also closely related to their microstructure. Traditional temperature rise prediction methods often ignore the impact of microstructural parameters (grain size, grain boundary thickness, and microuniformity) on temperature rise, focusing only on the impact of lightning impulse parameters on macroscopic heating and damage. This results in inaccurate temperature rise predictions for varistors. Summary of the Invention

[0004] The present invention provides a method, device, terminal equipment and storage medium for predicting the lightning impulse temperature rise of a ZnO varistor, which can solve the problem of inaccurate temperature rise prediction results of varistor in the prior art.

[0005] An embodiment of the present invention provides a method for predicting the lightning impulse temperature rise of a ZnO varistor, comprising:

[0006] Obtaining lightning impulse parameters and microstructure parameters of the ZnO varistor to be tested; wherein the microstructure parameters include: grain size, grain boundary layer thickness and microuniformity; the microuniformity is used to quantify the degree of difference in grain size;

[0007] Inputting the lightning impulse parameters and the microstructure parameters into a trained varistor temperature rise prediction model, so that the varistor temperature rise prediction model performs prediction based on the lightning impulse parameters and the microstructure parameters, and outputs the maximum surface temperature rise of the ZnO varistor;

[0008] The training process of the varistor temperature rise prediction model includes:

[0009] Acquire a plurality of data samples; wherein each of the data samples includes: a lightning impulse parameter sample, a microstructure parameter sample, and a surface temperature rise maximum value sample; wherein the surface temperature rise maximum value sample is obtained by measuring after applying a lightning impulse to the ZnO varistor sample; and the microstructure parameter sample is the microstructure parameter of the ZnO varistor sample;

[0010] Repeat the model training operation until the mean square error of the prediction results in the preset number of consecutive model training operations is less than a preset error threshold, thereby obtaining a trained varistor temperature rise prediction model;

[0011] The model training operation includes:

[0012] For each data sample, inputting the microstructure parameter sample of the lightning impulse parameter sample of the data sample into the constructed varistor temperature rise prediction model, so that the varistor temperature rise prediction model performs prediction based on the microstructure parameter sample of the lightning impulse parameter sample and outputs a prediction result;

[0013] Calculate the loss function value of the prediction results based on the prediction results of all data samples and the maximum surface temperature rise sample;

[0014] The varistor temperature rise prediction model is adjusted according to the loss function value.

[0015] Further, obtaining the grain size of the ZnO varistor to be tested includes:

[0016] Obtaining a scanning electron microscope image of the ZnO varistor and a magnification of the scanning electron microscope image;

[0017] Measuring the diagonal length of the scanning electron microscope image using a line-intercepting method, and counting the number of grain boundaries on the line-intercepting image;

[0018] Calculating the grain size of the ZnO varistor according to the magnification, the diagonal length, and the number of grain boundaries;

[0019] The calculation formula of the grain size is:

[0020]

[0021] Where d represents the grain size; X represents the magnification; S represents the diagonal length; and Y represents the number of grain boundaries.

[0022] Furthermore, the microscopic uniformity of the ZnO varistor to be tested is obtained, including:

[0023] Obtaining microscopic grain sizes of a plurality of grains in the ZnO varistor;

[0024] Calculating a micrograin size average and a micrograin size standard deviation based on all the micrograin sizes;

[0025] Calculating the microscopic uniformity of the ZnO varistor according to the microscopic grain size average value and the microscopic grain size standard deviation;

[0026] The calculation formula of the micro uniformity is:

[0027]

[0028] Where U d Indicates microscopic uniformity; represents the average value of micro grain size; σ represents the standard deviation of micro grain size.

[0029] Furthermore, the grain boundary layer thickness is the thickness of the ZnO varistor at the thinnest point of the grain boundary layer.

[0030] Furthermore, after obtaining the lightning impulse parameters and the microstructure parameters of the ZnO varistor to be measured, the method further includes:

[0031] Performing outlier detection on the lightning impulse parameters and the microstructure parameters, and eliminating outliers;

[0032] Filling missing values ​​for the lightning impulse parameters and the microstructure parameters after removing outliers;

[0033] The lightning impulse parameters and the microstructure parameters after filling in the missing values ​​are normalized.

[0034] Furthermore, the varistor temperature rise prediction model is a BP neural network model;

[0035] The varistor temperature rise prediction model includes: an input layer, a hidden layer and an output layer; the number of neurons in the input layer is 3 to 7, the number of neurons in the hidden layer is 3 to 10, and the number of neurons in the output layer is 1.

[0036] Furthermore, the initial weights of the varistor temperature rise prediction model are randomly selected from a normal distribution with a mean of 0 and a standard deviation of 0.1, and the initial learning rate is set to a random value between 0.01 and 0.1;

[0037] The varistor temperature rise prediction model is trained using a gradient descent method, and the parameters of the varistor temperature rise prediction model are adjusted using a back propagation algorithm.

[0038] Another embodiment of the present invention further provides a ZnO varistor lightning impulse temperature rise prediction device, comprising: a data acquisition module and a prediction module;

[0039] The data acquisition module is used to obtain lightning impulse parameters and microstructure parameters of the ZnO varistor to be tested; wherein the microstructure parameters include: grain size, grain boundary layer thickness and microuniformity; the microuniformity is used to quantify the degree of difference in grain size;

[0040] The prediction module is used to input the lightning impulse parameters and the microstructure parameters into a trained varistor temperature rise prediction model, so that the varistor temperature rise prediction model predicts according to the lightning impulse parameters and the microstructure parameters and outputs the maximum surface temperature rise of the ZnO varistor; wherein the training process of the varistor temperature rise prediction model includes: obtaining a plurality of data samples; wherein each of the data samples includes: a lightning impulse parameter sample, a microstructure parameter sample and a surface temperature rise maximum value sample; wherein the surface temperature rise maximum value sample is obtained by measuring after applying a lightning impulse to the ZnO varistor sample; the microstructure parameter sample is the microstructure of the ZnO varistor sample. structure parameters; repeatedly performing the model training operation until the mean square error of the prediction results in the model training operation for a consecutive preset number of times is less than a preset error threshold, thereby obtaining a trained varistor temperature rise prediction model; the model training operation includes: for each data sample, inputting the microstructure parameter sample of the lightning impulse parameter sample of the data sample into the constructed varistor temperature rise prediction model, so that the varistor temperature rise prediction model performs prediction based on the microstructure parameter sample of the lightning impulse parameter sample and outputs a prediction result; calculating the loss function value of the prediction result based on the prediction results of all data samples and the surface temperature rise maximum value sample; and adjusting the varistor temperature rise prediction model based on the loss function value.

[0041] Another embodiment of the present invention also provides a terminal device, including: a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, the steps of the ZnO varistor lightning impulse temperature rise prediction method are implemented.

[0042] Another embodiment of the present invention further provides a computer-readable storage medium item, comprising: a stored computer program, which controls the device where the computer-readable storage medium is located to execute the steps of the ZnO varistor lightning impulse temperature rise prediction method when the computer program is running.

[0043] The following beneficial effects are achieved by implementing the present invention:

[0044] The present invention obtains lightning impulse parameters and microstructural parameters of the ZnO varistor to be tested; inputs these parameters into a trained varistor temperature rise prediction model for prediction, and outputs the maximum surface temperature rise of the ZnO varistor to be tested. This method comprehensively considers the impact of three microstructural parameters, namely grain size, grain boundary layer thickness, and microuniformity, on lightning impulse temperature rise. This addresses the problem of inaccurate ZnO varistor temperature rise predictions, which occur in existing technologies, which focus solely on the macroscopic impact of lightning impulse parameters on the surface temperature rise of the varistor while ignoring the microscopic impact of microstructural parameters. Compared to existing technologies, the present invention can improve the accuracy of prediction results. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] In order to more clearly illustrate the technical solution of the present application, the following is a brief introduction to the drawings required for use in the implementation. Obviously, the drawings described below are only some implementation methods of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0046] Figure 1 This is a flow chart of a method for predicting the lightning impulse temperature rise of a ZnO varistor provided by one embodiment of the present invention;

[0047] Figure 2 This is a schematic diagram of the microscopic grain boundary structure of the ZnO varistor;

[0048] Figure 3 The present invention provides a schematic structural diagram of a ZnO varistor lightning impulse temperature rise prediction device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0049] To make the objectives, technical solutions, and advantages of this application more clear, the technical solutions in this application will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of this application. Obviously, the described embodiments are only part of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making any creative efforts are within the scope of protection of this application.

[0050] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which this application belongs; the terms used herein are only for the purpose of describing specific embodiments and are not intended to limit this application; the term "include" and any variations thereof in the specification and claims of this application and the above-mentioned figure descriptions are intended to cover non-exclusive inclusions.

[0051] References herein to "embodiments" mean that a particular feature, structure, or characteristic described in connection with the embodiments may be included in at least one embodiment of the present application. The appearance of this phrase in various places in the specification does not necessarily refer to the same embodiment, nor does it constitute an independent or alternative embodiment that is mutually exclusive of other embodiments. It is understood, both explicitly and implicitly, by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0052] See also Figure 1 To solve the problem of inaccurate temperature rise prediction results of varistors in the prior art, an embodiment of the present invention provides a method for predicting lightning impulse temperature rise of ZnO varistors, comprising:

[0053] S1. Obtaining lightning impulse parameters and microstructure parameters of the ZnO varistor to be tested; wherein the microstructure parameters include: grain size, grain boundary layer thickness and microuniformity; the microuniformity is used to quantify the degree of difference in grain size.

[0054] like Figure 2 FIG1 is a schematic diagram of the microscopic grain boundary structure of the ZnO varistor. In step S1, the microstructure of the ZnO varistor is characterized using a high-precision scanning electron microscope (SEM) and an X-ray diffractometer (XRD) to extract three parameters: grain size, grain boundary layer thickness, and microscopic uniformity.

[0055] It should be noted that lightning impulse parameters generally include: lightning impulse current amplitude, lightning impulse waveform, lightning impulse duration, number of lightning impulses, lightning impulse repetition frequency, and lightning impulse energy. Grain size refers to the average grain size estimated by the intercept method; grain boundary layer thickness refers to the thickness of the thinnest grain boundary layer; and microuniformity quantifies the degree of grain size variation.

[0056] Under multi-pulse lightning strikes, the thermal effects within ZnO varistors are complex, involving multiple aspects such as energy accumulation, heat diffusion, and microstructural changes. Traditional methods struggle to effectively address these complex relationships and are unable to provide high-precision temperature rise predictions. Grain size influences the varistor potential gradient, which in turn affects its shock resistance and flow rate. The thickness of the grain boundary layer determines the key location for intergranular conductivity, and its minimum value is crucial for studying microscopic properties. Microuniformity influences internal current paths and local temperature rise. The combined effect of these parameters reflects the impact of microstructural changes on thermal effects. Temperature rise is closely related to energy accumulation and heat diffusion, thus linking microstructural changes with energy accumulation and heat diffusion, and further linking microstructural parameters with lightning strike temperature rise.

[0057] In a preferred embodiment, obtaining the grain size of the ZnO varistor to be tested includes:

[0058] Obtaining a scanning electron microscope image of the ZnO varistor and a magnification of the scanning electron microscope image;

[0059] Measuring the diagonal length of the scanning electron microscope image using a line-intercepting method, and counting the number of grain boundaries on the line-intercepting image;

[0060] Calculating the grain size of the ZnO varistor according to the magnification, the diagonal length, and the number of grain boundaries;

[0061] The calculation formula of the grain size is:

[0062]

[0063] Where d represents the grain size; X represents the magnification; S represents the diagonal length; and Y represents the number of grain boundaries.

[0064] In this embodiment, the grain size is calculated using the line-intercept method to determine the average grain size, ensuring the comprehensiveness and accuracy of the microstructural parameters. S is the diagonal length of the SEM image when measured using the line-intercept method, X is the magnification of the SEM image, and Y is the number of grain boundaries on the line-intercepted SEM image.

[0065] In a preferred embodiment, obtaining the microscopic uniformity of the ZnO varistor to be measured includes:

[0066] Obtaining microscopic grain sizes of a plurality of grains in the ZnO varistor;

[0067] Calculating a micrograin size average and a micrograin size standard deviation based on all the micrograin sizes;

[0068] Calculating the microscopic uniformity of the ZnO varistor according to the microscopic grain size average value and the microscopic grain size standard deviation;

[0069] The calculation formula of the micro uniformity is:

[0070]

[0071] Where U d Indicates microscopic uniformity; represents the average value of micro grain size; σ represents the standard deviation of micro grain size.

[0072] It should be noted that the grain size mentioned above is a statistical value, which refers to the average size of the grains calculated by the intercept method. The microscopic grain size mentioned in this embodiment is a measured value, which is the actual size of the grains obtained by actual measurement.

[0073] In this embodiment, a large amount of micro grain size data is obtained and statistical analysis is performed to calculate the average micro grain size. The standard deviation of micro grain size σ is calculated to measure the degree of dispersion of micro grain size data. Calculation of the microscopic uniformity U of ZnO varistor using σ d To quantify the degree of difference in grain size. d The value is between 0 and 1. The closer it is to 1, the smaller the difference in grain size and the more consistent the lattice size; the closer it is to 0, the larger the difference in uniform grain size and the more uneven the lattice size.

[0074] In a preferred embodiment, the grain boundary layer thickness is the thickness of the ZnO varistor at the thinnest point of the grain boundary layer.

[0075] In a preferred embodiment, after obtaining the lightning impulse parameters and the microstructure parameters of the ZnO varistor to be measured, the method further includes:

[0076] Performing outlier detection on the lightning impulse parameters and the microstructure parameters, and eliminating outliers;

[0077] Filling missing values ​​for the lightning impulse parameters and the microstructure parameters after removing outliers;

[0078] The lightning impulse parameters and the microstructure parameters after filling in the missing values ​​are normalized.

[0079] In this example, data was detected and processed for outliers, using the Z-score method to identify and remove outliers to ensure data reliability. Missing values ​​were imputed using mean imputation or K-nearest neighbor imputation to ensure data integrity. Microstructural parameters were normalized, scaling all parameter values ​​to between 0 and 1 to eliminate the effects of different dimensions and improve model training efficiency.

[0080] S2. Inputting the lightning impulse parameters and the microstructure parameters into a trained varistor temperature rise prediction model, so that the varistor temperature rise prediction model performs prediction based on the lightning impulse parameters and the microstructure parameters, and outputs the maximum surface temperature rise of the ZnO varistor;

[0081] The training process of the varistor temperature rise prediction model includes:

[0082] Acquire a plurality of data samples; wherein each of the data samples includes: a lightning impulse parameter sample, a microstructure parameter sample, and a surface temperature rise maximum value sample; wherein the surface temperature rise maximum value sample is obtained by measuring after applying a lightning impulse to the ZnO varistor sample; and the microstructure parameter sample is the microstructure parameter of the ZnO varistor sample;

[0083] Repeat the model training operation until the mean square error of the prediction results in the preset number of consecutive model training operations is less than a preset error threshold, thereby obtaining a trained varistor temperature rise prediction model;

[0084] The model training operation includes:

[0085] For each data sample, inputting the microstructure parameter sample of the lightning impulse parameter sample of the data sample into the constructed varistor temperature rise prediction model, so that the varistor temperature rise prediction model performs prediction based on the microstructure parameter sample of the lightning impulse parameter sample and outputs a prediction result;

[0086] Calculate the loss function value of the prediction results based on the prediction results of all data samples and the maximum surface temperature rise sample;

[0087] The varistor temperature rise prediction model is adjusted according to the loss function value.

[0088] It should be noted that under multiple-pulse lightning strikes, different lightning strike parameters can cause microstructural changes in ZnO varistors, which in turn affect their temperature rise. For example, multiple-pulse strikes can reduce the grain size of ZnO varistors, increase grain boundaries, and create crack channels. These microstructural changes are related to grain size, grain boundary layer thickness, and microuniformity—microstructural parameters that are the input parameters of the varistor temperature rise prediction model. The varistor temperature rise prediction model indirectly accounts for the impact of lightning strike parameters on the temperature rise of ZnO varistors through these microstructural parameters.

[0089] During training, an early stopping mechanism is used to prevent overfitting. Training is stopped when the error on the validation set stops decreasing over multiple iterations. Specifically, after each iteration, the model's mean squared error (MSE) is checked. Training is stopped when the MSE is less than a set value (2.38°C) for five consecutive iterations. The model is considered to have converged to meet the preset accuracy requirements, and training is terminated.

[0090] It should be noted that the ZnO varistors were selected from the same batch, using the same process and formulation, ensuring basic consistency in the experiments. The only difference between the samples was in their microstructure. This screening method effectively controls variables, minimizing interference from other factors when studying the relationship between microstructural parameters and temperature rise, ensuring the accuracy and reliability of the research results.

[0091] The maximum surface temperature rise sample refers to the maximum surface temperature rise of the ZnO varistor sample under lightning impulse. An infrared thermal imager is used to scan the sample surface immediately after the multi-pulse impact to obtain a thermal imaging diagram of the surface temperature distribution. The maximum surface temperature rise is read and recorded from the diagram.

[0092] In a preferred embodiment, the varistor temperature rise prediction model is a BP neural network model;

[0093] The varistor temperature rise prediction model includes: an input layer, a hidden layer and an output layer; the number of neurons in the input layer is 3 to 7, the number of neurons in the hidden layer is 3 to 10, and the number of neurons in the output layer is 1.

[0094] In this embodiment, the varistor temperature rise prediction model includes an input layer, a hidden layer, and an output layer. The input parameters of the input layer are preprocessed lightning impulse parameters and microstructure parameters. The output parameter of the output layer is the predicted maximum surface temperature rise of the ZnO varistor under lightning impulse.

[0095] In a preferred embodiment, the initial weights of the varistor temperature rise prediction model are randomly selected from a normal distribution with a mean of 0 and a standard deviation of 0.1, and the initial learning rate is set to a random value between 0.01 and 0.1;

[0096] The varistor temperature rise prediction model is trained using a gradient descent method, and the parameters of the varistor temperature rise prediction model are adjusted using a back propagation algorithm.

[0097] It should be noted that the present invention aims to address the problem of inaccurate temperature rise predictions for ZnO varistors, which primarily focuses on the macroscopic effects of lightning impulse parameters on the surface temperature rise of varistors while ignoring the microscopic effects of microstructural parameters. Therefore, the present invention comprehensively considers the effects of three microstructural parameters on the temperature rise caused by lightning impulses: grain size, grain boundary layer thickness, and microuniformity. This method provides a method for predicting the maximum surface temperature rise of a varistor under lightning impulses based on these parameters: grain size, microuniformity of grain boundary layer thickness, and lightning impulse parameters. Compared to the existing technology, the present invention incorporates consideration of varistor microstructural parameters, which can improve the accuracy of prediction results.

[0098] In one embodiment, the method for predicting the lightning impulse temperature rise of a ZnO varistor further includes:

[0099] The prediction results were analyzed using mean squared error (MSE) to assess the model's prediction accuracy under different microstructural parameters and ensure the reliability of the prediction results. By calculating the MSE, the degree of discrepancy between the model's predicted maximum surface temperature rise and the actual maximum surface temperature rise can be quantified. A smaller MSE value indicates that the model's prediction is closer to the actual value and more accurate. Conversely, a larger MSE value indicates a poorer prediction.

[0100] To optimize prediction accuracy based on the error analysis results, adjust network parameters: First, adjust the learning rate. The learning rate affects the step size of weight updates during model training. If the mean squared error decreases slowly during training, the learning rate may be too small. Increase the learning rate appropriately to accelerate model convergence. If the mean squared error fluctuates or even increases, the learning rate may be too large. Reduce the learning rate to make model training more stable. In the original model, the learning rate is 1. Try different learning rates, such as 0.1 and 0.01, to observe changes in the mean squared error and select the one that minimizes the mean squared error. Reinitialize the weights. Different initial weights will cause the model to converge to different results. When the mean squared error is large, randomly reinitialize the weights and train the model multiple times to select the initial weight combination that minimizes the mean squared error to improve the model's prediction accuracy.

[0101] This invention can be used for lightning protection design optimization and fault warning and prevention. It provides a scientific basis for the design optimization of ZnO varistors, helping to improve material formulations and production processes, and enhancing their ability to withstand multiple pulse shocks. By predicting temperature rise, the microstructure of ZnO varistors can be optimized, improving their protective performance in power systems. This invention uses real-time temperature rise prediction to provide early warning of potential faults such as thermal damage and fire. Research has shown that the maximum surface temperature rise threshold for thermal damage or fire in ZnO varistors is 180°C. Under operating load, ZnO varistors can exceed this threshold, leading to thermal imbalance and fire. Approaching or reaching 180°C can be used as an important warning threshold. When the model-predicted maximum surface temperature rise approaches or exceeds this threshold, a warning signal for thermal damage or fire is issued. When the surface temperature reaches this threshold, the thermal stress within the ZnO varistor may exceed its tolerance limit, significantly changing material properties and posing a significant risk of thermal damage and fire. This provides support for fault diagnosis and prevention in power systems. The prediction results can be used to guide the operation and maintenance of power systems and reduce safety risks caused by lightning.

[0102] like Figure 3 As shown, based on the above method embodiment, a corresponding device embodiment is provided;

[0103] An embodiment of the present invention provides a ZnO varistor lightning impulse temperature rise prediction device, comprising: a data acquisition module and a prediction module;

[0104] The data acquisition module is used to obtain the grain size, grain boundary layer thickness and micro-uniformity of the ZnO varistor to be tested; wherein the micro-uniformity is used to quantify the degree of difference in grain size;

[0105] The prediction module is used to input the grain size, the grain boundary layer thickness and the micro-uniformity into a trained varistor temperature rise prediction model, so that the varistor temperature rise prediction model predicts according to the grain size, the grain boundary layer thickness and the micro-uniformity, and outputs the maximum surface temperature rise of the ZnO varistor; wherein the training process of the varistor temperature rise prediction model includes: obtaining grain size samples, grain boundary layer thickness samples, micro-uniformity samples and surface temperature rise maximum samples of a plurality of ZnO varistor samples; wherein the surface temperature rise maximum sample is obtained by measuring after applying a lightning impulse to the ZnO varistor sample; repeating the model training operation until the model training operation is performed for a preset number of consecutive times. The mean square error of the prediction result is less than a preset error threshold, and a trained varistor temperature rise prediction model is obtained; the model training operation includes: for each ZnO varistor sample, the grain size sample, the grain boundary layer thickness sample, and the micro-uniformity sample of the ZnO varistor sample are input into the constructed varistor temperature rise prediction model, so that the varistor temperature rise prediction model predicts according to the grain size sample, the grain boundary layer thickness sample, and the micro-uniformity sample of the ZnO varistor sample, and outputs the prediction result; based on the prediction results of all ZnO varistor samples and the surface temperature rise maximum value sample, the mean square error and the loss function value of the prediction result are calculated; and according to the loss function value, the varistor temperature rise prediction model is adjusted.

[0106] It can be understood that the above-mentioned device embodiment corresponds to the method embodiment of the present invention, which can implement the ZnO varistor lightning impulse temperature rise prediction method provided by any of the above-mentioned method embodiments of the present invention.

[0107] It should be noted that the device embodiments described above are merely illustrative, and some or all of the modules may be selected according to actual needs to achieve the purpose of the present embodiment. Furthermore, in the drawings of the device embodiments provided by the present invention, the connection relationship between modules indicates that they have a communication connection, which may be implemented as one or more communication buses or signal lines. Those skilled in the art can understand and implement the present invention without inventive effort.

[0108] Based on the above-mentioned method embodiment, another embodiment of the present invention provides a terminal device, which includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, it implements the ZnO varistor lightning impulse temperature rise prediction method of any embodiment of the present invention.

[0109] For example, in this embodiment, the computer program may be divided into one or more modules, which are stored in the memory and executed by the processor to implement the present invention. The one or more module elements may be a series of computer program instruction segments capable of performing specific functions, and the instruction segments are used to describe the execution process of the computer program in the terminal device.

[0110] The terminal device may be a computing device such as a desktop computer, a notebook computer, a PDA, a cloud server, etc. The terminal device may include, but is not limited to, a processor and a memory.

[0111] The processor may be a central processing unit (CPU), other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor, etc. The processor is the control center of the terminal device, connecting various parts of the entire terminal device using various interfaces and lines.

[0112] Based on the above method embodiments, another embodiment of the present invention provides a computer-readable storage medium, including a stored computer program, wherein when the computer program is running, the device where the computer-readable storage medium is located is controlled to execute the ZnO varistor lightning impulse temperature rise prediction method described in any one of the above method embodiments of the present invention.

[0113] Wherein, the module / unit integrated in the device / terminal equipment, if implemented in the form of a software functional unit and sold or used as an independent product, can be stored in a computer-readable storage medium. Based on this understanding, the present invention implements all or part of the process in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and when the computer program is executed by the processor, it can implement the steps of the above-mentioned various method embodiments. Wherein, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may include: any entity or device that can carry the computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electric carrier signal, telecommunication signal and software distribution medium, etc.

[0114] The above is a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications are also considered to be within the scope of protection of the present invention.

Claims

1. A method for predicting the lightning impulse temperature rise of a ZnO varistor, characterized in that: include: Obtaining lightning impulse parameters and microstructure parameters of the ZnO varistor to be tested; wherein the microstructure parameters include: grain size, grain boundary layer thickness and microuniformity; the microuniformity is used to quantify the degree of difference in grain size; Inputting the lightning impulse parameters and the microstructure parameters into a trained varistor temperature rise prediction model, so that the varistor temperature rise prediction model performs prediction based on the lightning impulse parameters and the microstructure parameters, and outputs the maximum surface temperature rise of the ZnO varistor; The training process of the varistor temperature rise prediction model includes: Acquire a plurality of data samples; wherein each of the data samples includes: a lightning impulse parameter sample, a microstructure parameter sample, and a surface temperature rise maximum value sample; wherein the surface temperature rise maximum value sample is obtained by measuring after applying a lightning impulse to the ZnO varistor sample; and the microstructure parameter sample is the microstructure parameter of the ZnO varistor sample; Repeat the model training operation until the mean square error of the prediction results in the preset number of consecutive model training operations is less than a preset error threshold, thereby obtaining a trained varistor temperature rise prediction model; The model training operation includes: For each data sample, inputting the microstructure parameter sample of the lightning impulse parameter sample of the data sample into the constructed varistor temperature rise prediction model, so that the varistor temperature rise prediction model performs prediction based on the microstructure parameter sample of the lightning impulse parameter sample and outputs a prediction result; Calculate the loss function value of the prediction results based on the prediction results of all data samples and the maximum surface temperature rise sample; The varistor temperature rise prediction model is adjusted according to the loss function value.

2. The method for predicting the temperature rise of a ZnO varistor during lightning strike according to claim 1, wherein: Obtaining the grain size of the ZnO varistor to be tested, including; Obtaining a scanning electron microscope image of the ZnO varistor and a magnification of the scanning electron microscope image; Measuring the diagonal length of the scanning electron microscope image using a line-intercepting method, and counting the number of grain boundaries on the line-intercepting image; Calculating the grain size of the ZnO varistor according to the magnification, the diagonal length, and the number of grain boundaries; The calculation formula of the grain size is: Where d represents the grain size; X represents the magnification; S represents the diagonal length; and Y represents the number of grain boundaries.

3. The method for predicting the temperature rise of a ZnO varistor under lightning impulse as claimed in claim 1, wherein: Obtain the microscopic uniformity of the ZnO varistor to be tested, including: Obtaining microscopic grain sizes of a plurality of grains in the ZnO varistor; Calculating a micrograin size average and a micrograin size standard deviation based on all the micrograin sizes; Calculating the microscopic uniformity of the ZnO varistor according to the microscopic grain size average value and the microscopic grain size standard deviation; The calculation formula of the micro uniformity is: Where U d Indicates microscopic uniformity; represents the average value of micro grain size; σ represents the standard deviation of micro grain size.

4. The method for predicting the lightning impulse temperature rise of a ZnO varistor according to claim 1, wherein: The grain boundary layer thickness is the thickness of the ZnO varistor at the thinnest grain boundary layer.

5. The method for predicting the temperature rise of a ZnO varistor during lightning strike according to claim 1, wherein: After obtaining the lightning impulse parameters and the microstructure parameters of the ZnO varistor to be tested, the following steps are also included: Performing outlier detection on the lightning impulse parameters and the microstructure parameters, and eliminating outliers; Filling missing values ​​for the lightning impulse parameters and the microstructure parameters after removing outliers; The lightning impulse parameters and the microstructure parameters after filling in the missing values ​​are normalized.

6. The method for predicting the temperature rise of a ZnO varistor during lightning strike according to claim 1, wherein: The varistor temperature rise prediction model is a BP neural network model; The varistor temperature rise prediction model includes: an input layer, a hidden layer and an output layer; the number of neurons in the input layer is 3 to 7, the number of neurons in the hidden layer is 3 to 10, and the number of neurons in the output layer is 1.

7. The method for predicting the temperature rise of a ZnO varistor during lightning strike according to claim 1, wherein: The initial weights of the varistor temperature rise prediction model are randomly selected from a normal distribution with a mean of 0 and a standard deviation of 0.1, and the initial learning rate is set to a random value between 0.01 and 0.1; The varistor temperature rise prediction model is trained using a gradient descent method, and the parameters of the varistor temperature rise prediction model are adjusted using a back propagation algorithm.

8. A ZnO varistor lightning impulse temperature rise prediction device, characterized in that: include: Data acquisition module and prediction module; The data acquisition module is used to obtain lightning impulse parameters and microstructure parameters of the ZnO varistor to be tested; wherein the microstructure parameters include: grain size, grain boundary layer thickness and microuniformity; the microuniformity is used to quantify the degree of difference in grain size; The prediction module is used to input the lightning impulse parameters and the microstructure parameters into a trained varistor temperature rise prediction model, so that the varistor temperature rise prediction model predicts according to the lightning impulse parameters and the microstructure parameters and outputs the maximum surface temperature rise of the ZnO varistor; wherein the training process of the varistor temperature rise prediction model includes: obtaining a plurality of data samples; wherein each of the data samples includes: a lightning impulse parameter sample, a microstructure parameter sample and a surface temperature rise maximum value sample; wherein the surface temperature rise maximum value sample is obtained by measuring after applying a lightning impulse to the ZnO varistor sample; the microstructure parameter sample is the microstructure of the ZnO varistor sample. structure parameters; repeatedly performing the model training operation until the mean square error of the prediction results in the model training operation for a consecutive preset number of times is less than a preset error threshold, thereby obtaining a trained varistor temperature rise prediction model; the model training operation includes: for each data sample, inputting the microstructure parameter sample of the lightning impulse parameter sample of the data sample into the constructed varistor temperature rise prediction model, so that the varistor temperature rise prediction model performs prediction based on the microstructure parameter sample of the lightning impulse parameter sample and outputs a prediction result; calculating the loss function value of the prediction result based on the prediction results of all data samples and the surface temperature rise maximum value sample; and adjusting the varistor temperature rise prediction model based on the loss function value.

9. A terminal device, characterized in that: The method comprises a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, the method for predicting the lightning impulse temperature rise of a ZnO varistor according to any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium, characterized in that include: A stored computer program, wherein when the computer program is running, the device where the computer-readable storage medium is located is controlled to execute the ZnO varistor lightning impulse temperature rise prediction method according to any one of claims 1 to 7.

Images