Method, system, device and storage medium for predicting thermal defects of electrical connectors

Abstract

The present invention discloses a method, system, device and storage medium for predicting thermal defects of electrical connectors, which relate to the field of equipment detection technology. The key points of its technical solution are: the present invention takes into account that since the change of the operating parameters of the electrical equipment at the next moment is uncertain, the temperature obtained by predicting the current operating parameters of the electrical connector based on the neural network temperature prediction model will have deviations in accuracy. Therefore, the power load of the power system at the next moment is predicted based on the load prediction model, and the temperature predicted by the current operating parameters is corrected based on the predicted power load, thereby reducing the degree of deviation of the temperature at the next moment predicted by the current operating parameters, thereby improving the accuracy of temperature prediction, and then using the corrected temperature to predict the degree of thermal defects of different categories of electrical connectors, providing a theoretical basis for regular maintenance and testing of the connectors, thereby reducing the frequency of emergency repairs caused by overheating of electrical connectors during the peak summer period.

Description

Technical Field

[0001] The present invention relates to the technical field of equipment detection, and more particularly to a method, system, equipment and storage medium for predicting thermal defects of electrical connectors. Background Art

[0002] The planned maintenance of electrical connectors is still in the mode of taking power outages and arranging planned maintenance when infrared temperature measurement and other means are used to find that the connectors have critical defects. The emergency repair time of electrical connectors with thermal defects is long, and there are problems such as reverse load and reverse mother, which may cause long power outages.

[0003] However, electrical connectors that overheat during high-load conditions such as peak summer demand are often difficult to predict and discover in advance with existing technical means under normal loads. Therefore, how to predict the thermal defects caused by temperature changes in the connectors of the operated and maintained equipment can provide a basis for arranging regular inspections and tests of the connectors, and reduce emergency repairs caused by overheating of electrical connectors during peak summer demand. Summary of the Invention

[0004] To address the shortcomings of the prior art, the present application provides a method, system, device, and storage medium for predicting thermal defects in electrical connectors. These methods are capable of identifying target areas containing electrical connectors and the types of electrical connectors in the target areas within acquired historical infrared images of electrical equipment, thereby ensuring the accuracy of subsequent predictions of the degree of thermal defects. The method predicts the current operating parameters of the electrical connector based on a pre-trained neural network temperature prediction model to obtain the temperature of the electrical connector at the next moment. Since the changes in the operating parameters of the electrical equipment at the next moment are uncertain, the temperature predicted based on the current operating parameters of the electrical connector based on the neural network temperature prediction model may have deviations in accuracy. Therefore, the power load of the power system at the next moment is predicted based on a load prediction model. The temperature predicted based on the current operating parameters is corrected based on the predicted power load, thereby reducing the deviation in the temperature predicted based on the current operating parameters at the next moment, thereby improving the accuracy of the temperature prediction. The corrected temperatures are then used to predict the degree of thermal defects of different types of electrical connectors, providing a theoretical basis for regular inspection and testing of the connectors, thereby reducing the frequency of emergency repairs due to overheating of electrical connectors during the peak summer season.

[0005] The above technical objectives of the present invention are achieved through the following technical solutions:

[0006] In a first aspect, the present application provides a method for predicting thermal defects of an electrical connector, comprising:

[0007] Obtain historical infrared images of electrical equipment and power load sequences of power systems;

[0008] Identify the target area containing electrical connectors in the historical infrared image using the trained electrical connector recognition model, and identify the category of the electrical connectors in the target area;

[0009] extracting temperature information of the electrical connector according to the target area, and determining a temperature state of the electrical connector based on the temperature information of the electrical connector, wherein the temperature state is a normal state or an abnormal state;

[0010] When the temperature state of the electrical connector is abnormal, obtaining current operating parameters of the electrical connector, inputting the current operating parameters into a pre-trained neural network temperature prediction model, and predicting the temperature of the electrical connector at a next moment, wherein the neural network temperature prediction model is trained based on historical temperature parameters of the electrical connector determined based on historical operating parameters of the electrical connector;

[0011] Input the power load sequence into the pre-trained load forecasting model to predict the power load of the power system at the next moment;

[0012] Correcting the temperature of the electrical connector at the next moment using the power load of the power system at the next moment to obtain a corrected temperature of the electrical connector;

[0013] The thermal defect levels of different types of electrical connectors are predicted based on the corrected temperatures of the electrical connectors.

[0014] In one embodiment, the electrical connector recognition model is obtained by training a convolutional neural network using a training image set, wherein the training image set includes infrared images of electrical connectors of different categories;

[0015] The load forecasting model consists of a graph convolutional neural network module, a time series convolutional neural network module and a fully connected layer.

[0016] In one embodiment, determining a temperature state of the electrical connector based on temperature information of the electrical connector includes:

[0017] The temperature state of the electrical connector is determined by comparing the standard temperature information corresponding to the category of the electrical connector in the database with the temperature information of the electrical connector; wherein, when the temperature information of the electrical connector is greater than the standard temperature information, the electrical temperature state is an abnormal state, and when the temperature information of the electrical connector is less than and equal to the standard temperature information, the electrical temperature state is a normal state.

[0018] In one embodiment, the neural network temperature prediction model is obtained by the following training method:

[0019] Obtain historical operating parameters of electrical connectors within a preset time;

[0020] At a preset interval within the preset time, determining a historical temperature parameter of the electrical connector based on the historical operating parameters;

[0021] The neural network temperature prediction model to be trained is trained using historical temperature parameters of the electrical connector to obtain a trained neural network temperature prediction model.

[0022] In one embodiment, the historical operating parameters include a maximum current value and a minimum current value of the electrical connector within a preset time period, and a maximum temperature and a minimum temperature of an environment in which the electrical connector is located;

[0023] Determining the historical temperature parameters of the electrical connector according to the historical operating parameters includes:

[0024] calculating a maximum heat dissipation and a minimum heat dissipation of the electrical connector within a preset time period based on a maximum current value, a minimum current value, and a resistance value of the electrical connector, and calculating a first temperature fluctuation range of the electrical connector within a preset time interval based on the maximum heat dissipation and the minimum heat dissipation;

[0025] A first temperature fluctuation range of the electrical connector within a preset time interval is coupled and compensated by the maximum temperature and the minimum temperature of the environment in which the electrical connector is located, thereby obtaining a second temperature fluctuation range of the electrical connector within the preset time interval;

[0026] A historical temperature parameter of the electrical connector is determined based on the second temperature fluctuation range.

[0027] In one embodiment, coupling compensation is performed on a first temperature fluctuation range of the electrical connector within a preset time interval using a maximum temperature and a minimum temperature of an environment in which the electrical connector is located, to obtain a second temperature fluctuation range of the electrical connector within the preset time interval, including:

[0028] A heat conduction model is established according to the structural parameters of the electrical connector. The maximum temperature and the minimum temperature of the environment in which the electrical connector is located are coupled and compensated according to the heat conduction model of the electrical connector to obtain a second temperature fluctuation range of the electrical connector within a preset interval time.

[0029] In one embodiment, the current operating parameters of the electrical connector are the current value flowing through the electrical connector and the temperature value of the environment at the current moment;

[0030] The current value flowing through the electrical connector and the ambient temperature value at the current moment are input into a pre-trained neural network temperature prediction model to predict the temperature of the electrical connector at the next moment;

[0031] Predict the degree of thermal defects of different types of electrical connectors based on the corrected temperature of the electrical connector, including:

[0032] Determining a predicted critical temperature of the electrical connector according to the type of the electrical connector;

[0033] The degree of thermal defect of the electrical joint is predicted based on the difference between the corrected temperature of the electrical joint and the critical temperature.

[0034] In a second aspect, the present application provides an electrical connector thermal defect prediction system, comprising:

[0035] A data acquisition module is used to obtain historical infrared images of electrical equipment and power load sequences of the power system;

[0036] an electrical connector recognition module, configured to identify a target area of ​​the historical infrared image containing electrical connectors using a trained electrical connector recognition model, and to identify the type of electrical connector in the target area;

[0037] a temperature state determining module, configured to extract temperature information of the electrical connector according to the target area, and determine a temperature state of the electrical connector based on the temperature information of the electrical connector, wherein the temperature state is a normal state or an abnormal state;

[0038] a temperature prediction module for obtaining current operating parameters of the electrical connector when the temperature of the electrical connector is abnormal, inputting the current operating parameters into a pre-trained neural network temperature prediction model, and predicting the temperature of the electrical connector at a next moment, wherein the neural network temperature prediction model is trained based on historical temperature parameters of the electrical connector determined based on historical operating parameters of the electrical connector;

[0039] The load forecasting module is used to input the power load sequence into the pre-trained load forecasting model to predict the power load of the power system at the next moment;

[0040] a temperature correction module, configured to correct the temperature of the electrical connector at the next moment using the power load of the power system at the next moment, thereby obtaining the corrected temperature of the electrical connector;

[0041] The defect prediction module is used to predict the degree of thermal defects of different types of electrical connectors based on the corrected temperature of the electrical connectors.

[0042] In a third aspect, the present application provides an electronic device comprising: a memory and a processor, wherein a computer program is stored on the memory, and the computer program can be executed by the processor so that the processor implements a method for predicting thermal defects of electrical connectors as described in any one of the first aspects.

[0043] In a fourth aspect, the present application provides a computer-readable storage medium, in which a computer program is stored. The computer program is suitable for being loaded and executed by a processor, so that the processor implements a method for predicting thermal defects of electrical connectors as described in any one of the first aspects.

[0044] Compared with the prior art, the present invention has the following beneficial effects:

[0045] 1. The present invention provides a method for predicting thermal defects of electrical connectors. Taking into account the uncertainty of the changes in the operating parameters of the electrical equipment at the next moment, the temperature obtained by predicting the current operating parameters of the electrical connector based on the neural network temperature prediction model will have deviations in accuracy. Therefore, the power load of the power system at the next moment is predicted based on the load prediction model, and the temperature predicted by the current operating parameters is corrected based on the predicted power load, thereby reducing the degree of deviation of the temperature at the next moment predicted by the current operating parameters, thereby improving the accuracy of temperature prediction. The corrected temperature is then used to predict the degree of thermal defects of different types of electrical connectors, providing a theoretical basis for regular inspection and testing of the connectors, thereby reducing the frequency of emergency repairs caused by overheating of electrical connectors during the peak summer period.

[0046] 2. The present invention can identify the target area containing electrical connectors and the type of electrical connectors in the target area in the acquired historical infrared images of the electrical equipment, thereby ensuring the accuracy of the subsequent prediction of the degree of thermal defects.

[0047] 3. The present invention only predicts the temperature of electrical connectors with abnormal temperature status, and does not predict the temperature of electrical connectors with normal temperature status, so as to avoid the problem that the time to output the prediction results becomes longer due to the low processing efficiency of the electronic device processing a large number of images, thereby affecting the user experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] The drawings described herein are used to provide a further understanding of the embodiments of the present invention, constitute a part of this application, and do not constitute a limitation of the embodiments of the present invention. In the drawings:

[0049] Figure 1 A schematic flow chart of a method for predicting thermal defects in electrical connectors provided in an embodiment of the present application;

[0050] Figure 2 A schematic diagram of the process of training a neural network temperature prediction model provided in an embodiment of the present application;

[0051] Figure 3 A structural block diagram of an electrical connector thermal defect prediction system provided in an embodiment of the present application;

[0052] Figure 4A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0053] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with examples and drawings. The exemplary embodiments of the present invention and their descriptions are only used to explain the present invention and are not intended to limit the present invention.

[0054] The planned maintenance of electrical connectors is still in the mode of taking power outages and arranging planned maintenance when infrared temperature measurement and other means are used to find that the connectors have critical defects. The emergency repair time of electrical connectors with thermal defects is long, and there are problems such as reverse load and reverse mother, which may cause long power outages.

[0055] However, electrical connectors that overheat during high-load conditions such as peak summer demand are often difficult to predict and discover in advance with existing technical means under normal loads. Therefore, how to predict the thermal defects caused by temperature changes in the connectors of the operated and maintained equipment can provide a basis for arranging regular inspections and tests of the connectors, and reduce emergency repairs caused by overheating of electrical connectors during peak summer demand.

[0056] To solve the above technical problems, the present application discloses a method, system, device, and storage medium for predicting thermal defects in electrical connectors, which can improve the accuracy of electrical connector temperature prediction and then use the corrected temperature to predict the degree of thermal defects in different types of electrical connectors, providing a theoretical basis for regular inspection and testing of connectors, thereby reducing the frequency of emergency repairs caused by overheating of electrical connectors during the peak summer season. The following is a detailed description with reference to the accompanying drawings.

[0057] Please refer to Figure 1 As shown, Figure 1 FIG. 1 is a flow chart of a method for predicting thermal defects of an electrical connector disclosed in an embodiment of the present application. Figure 1 As shown, the electrical joint thermal defect prediction method includes the following steps.

[0058] Step S110 , obtaining historical infrared images of electrical equipment and power load sequences of the power system.

[0059] In the embodiments of this application, electrical equipment is a general term for equipment used in power systems to ensure the normal operation and transmission of electricity. Electrical equipment may include generators, transformers, power lines, circuit breakers, and other equipment. An infrared image is formed by capturing the intensity of infrared light from an object. This image is obtained by measuring the heat radiated from the object. It is formed by capturing the infrared radiation of the target in the infrared band using infrared imaging equipment.

[0060] In an embodiment of the present application, the historical infrared image is an infrared image of the electrical equipment in the area to be inspected collected by an infrared imaging device, wherein the electrical equipment in the area to be inspected includes one or more electrical equipment that needs to be inspected, and the infrared imaging device may include an infrared camera, an infrared camera, etc.

[0061] In an embodiment of the present application, the electrical connector thermal defect prediction method is applicable to electronic devices such as terminal devices or servers, wherein the operating system of the electronic device may include but is not limited to Android operating system, IOS operating system, Windows Phone 8 operating system, etc., and is not limited in the embodiment of the present application.

[0062] In this embodiment, the power load sequence of the power system is obtained by taking time as a line and recording the power load at each moment.

[0063] Step S120 , identifying a target area containing electrical connectors in the historical infrared image using the trained electrical connector recognition model, and identifying the type of the electrical connectors in the target area.

[0064] In this embodiment, the electrical connector recognition model can be a neural network model, which includes at least a backbone network, such as a Resnet50 network. The last layer of the network is a fully connected layer, and its dimension can be 1024 or 2048. What is obtained here is the feature vector of the electrical connector. In the embodiment of the present application, the trained electrical connector recognition model can be a trained instance segmentation model, which is convenient for segmenting the target area containing electrical connectors in the historical infrared image. The instance segmentation model may include but is not limited to the R-CNN (Region with CNN feature) model, the Fast R-CNN model, the Mask RCNN model, etc. This is a conventional technical means for constructing a recognition model in the field of image recognition technology, so no further explanation and illustration will be given here.

[0065] In some embodiments, the electrical connector recognition model is obtained by training a convolutional neural network using a training image set, wherein the training image set includes infrared images of various categories of electrical connectors. Specifically, the training image set includes infrared images of various components of an electrical device, and each component is annotated with information about different categories of electrical connectors as an infrared image. The electronic device can use infrared images of various components of multiple different electrical devices as electrical connectors, which can better identify and segment electrical connectors in the electrical devices in the inspection area and effectively indicate whether there are any problems with the electrical connectors. This improves the accuracy of electrical connector recognition while also increasing the efficiency of the electrical connector recognition process.

[0066] Since electrical connectors are key components connecting some electrical equipment, such as cable connectors, knife switches, connectors on both sides of transformers, etc., their locations are relatively hidden, and it is inconvenient to use some drones equipped with infrared cameras to capture corresponding infrared images, and the workload of data collection is also large. Therefore, in step S110, a historical infrared image of the electrical equipment in a to-be-detected area is obtained, and the electrical connector recognition model is used to identify the acquired historical infrared image, so as to determine the target area containing the electrical connector and confirm the category of the electrical connector in the target area. The purpose of confirming the category of the electrical connector is to better predict the thermal defects of the electrical connector in the future, because different electrical connectors are made of different materials, resulting in different heat resistance, resulting in different upper and lower temperature limits of their respective thermal defects. If a unified temperature threshold range is used, it cannot be applied to actual prediction scenarios.

[0067] Step S130 : extracting temperature information of the electrical connector according to the target area, and determining a temperature state of the electrical connector based on the temperature information of the electrical connector, wherein the temperature state is a normal state or an abnormal state.

[0068] In this embodiment, the electronic device extracts temperature information of the electrical connector in the target area based on the historical infrared image. Specifically, the historical infrared image may be converted into a grayscale image, and the grayscale value of each pixel may be converted into a corresponding temperature value based on the linear relationship between grayscale values ​​and temperature values. The electronic device may also convert each pixel in the first infrared image into a corresponding temperature value based on the relationship between RGB values ​​and temperature values. This application does not specifically limit the method for obtaining temperature information based on the historical infrared image.

[0069] Since the category of the electrical connector has been identified in step S120, and the basic parameters of the electrical connector in the area to be inspected are known, such as model, rated voltage, rated current, operating temperature, etc., the temperature state of the electrical connector is determined based on the temperature information of the electrical connector, where the temperature state can be normal or abnormal.

[0070] Step S140, when the temperature state of the electrical connector is abnormal, obtain the current operating parameters of the electrical connector, input the current operating parameters into a pre-trained neural network temperature prediction model, and predict the temperature of the electrical connector at the next moment, wherein the neural network temperature prediction model is trained based on the historical temperature parameters of the electrical connector determined based on the historical operating parameters of the electrical connector.

[0071] In this embodiment, since historical image data is obtained, if the temperature state of a certain electrical connector has been in a normal state during the historical time period, it can be said that the electrical connector has not experienced high temperature, that is, the electrical connector has been at a standard operating temperature, which is in line with normal equipment usage. That is, there is no need to continue predicting the thermal defects of the electrical connector, thereby avoiding the time required to output the prediction results being prolonged due to low processing efficiency caused by the electronic equipment processing a large number of images, thereby affecting the user experience.

[0072] Based on the above specific considerations, in this embodiment, specifically, the historical temperature parameters of the electrical connector are confirmed through the historical operating parameters of the electrical connector in the power system, and a neural network temperature prediction model is trained based on the historical temperature parameters of the electrical connector. When the temperature state of the electrical connector is abnormal, the current operating parameters of the electrical connector are obtained, and the current operating parameters of the electrical connector are input into the pre-trained neural network temperature prediction model to predict the temperature of the electrical connector at the next moment.

[0073] Step S150: input the power load sequence into a pre-trained load forecasting model to predict the power load of the power system at the next moment.

[0074] In this embodiment, the power load sequence is input into a pre-trained load forecasting model to predict the power load of the power system at the next moment. This is a routine operation and no redundant explanation is given.

[0075] In one embodiment, the load forecasting model is composed of a graph convolutional neural network module, a time series convolutional neural network module, and a fully connected layer. Specifically, the load forecasting model is constructed in sequence using the graph convolutional neural network module, the time series convolutional neural network module, and the fully connected layer. The load forecasting model decomposes the power load sequence to obtain features of different scales. Then, in order to learn the potential relationship between the variables in the power load sequence, a graph convolutional neural network is constructed, and the dimensions of the time series of the power load are used as the nodes of the graph convolutional neural network. A multi-attention head mechanism is used to learn the relationship between the nodes to improve the nonlinear mapping ability. In addition, the graph convolutional neural network is also used to update the node features to generate the node embedding to fully consider the relationship between the variables, so as to avoid falling into the local optimal state. Finally, the time relationship of the node embedding is established through the time series convolutional neural network to further enhance the time modeling ability of the model. Finally, a fully connected layer is connected to the output of the time series convolutional neural network to realize the power load sequence prediction at the next moment.

[0076] Training a load forecasting model is conventional. For example, the power load sequence obtained in step S110 serves as the input sequence and target output for training samples, equivalent to a training set and a validation set. The training set accounts for 80% and the validation set accounts for 20%, respectively, as in conventional practice, to train the model in step S102. By training the load forecasting model based on the input sequence and target output, a trained load forecasting model can be obtained.

[0077] Step S160 , using the power load of the power system at the next moment to correct the temperature of the electrical connector at the next moment, to obtain the corrected temperature of the electrical connector.

[0078] Specifically, different power loads will generate different amounts of heat, which will affect temperature changes. Therefore, the predicted temperature is corrected based on the temperature matched by the electrical connector under the power load. Specifically, a regression analysis algorithm can be used to correct the temperature of the electrical connector at the next moment to obtain the corrected temperature, thereby reducing the degree of deviation of the temperature at the next moment predicted by the current operating parameters, thereby improving the accuracy of temperature prediction.

[0079] In step S170 , the thermal defect levels of different types of electrical connectors are predicted based on the corrected temperatures of the electrical connectors.

[0080] Specifically, how to predict the degree of thermal defects is a relatively conventional technical means, such as a ratio method, a difference method, etc., and this embodiment does not make any specific limitations.

[0081] In summary, the thermal defect prediction method of the embodiment of the present application can identify a target area containing an electrical connector and the type of electrical connector in the target area in the acquired historical infrared images of the electrical equipment, thereby ensuring the accuracy of the subsequent prediction of the degree of thermal defect. The current operating parameters of the electrical connector are predicted based on a pre-trained neural network temperature prediction model to obtain the temperature of the electrical connector at the next moment. Since the change in the operating parameters of the electrical equipment at the next moment is uncertain, the accuracy of the temperature predicted by the neural network temperature prediction model based on the current operating parameters of the electrical connector will be biased. Therefore, the power load of the power system at the next moment is predicted based on the load prediction model. The temperature predicted by the current operating parameters is corrected based on the predicted power load, thereby reducing the degree of deviation of the temperature predicted by the current operating parameters at the next moment, thereby improving the accuracy of the temperature prediction. The corrected temperature is then used to predict the degree of thermal defect of different types of electrical connectors, providing a theoretical basis for regular inspection and testing of the connectors, thereby reducing the frequency of emergency repairs caused by overheating of electrical connectors during the peak summer period.

[0082] In some embodiments, determining the temperature state of the electrical connector based on the temperature information of the electrical connector includes:

[0083] The temperature state of the electrical connector is determined by comparing the standard temperature information corresponding to the type of the electrical connector in the database with the temperature information of the electrical connector. When the temperature information of the electrical connector is greater than the standard temperature information, the electrical temperature state is abnormal; when the temperature information of the electrical connector is less than or equal to the standard temperature information, the electrical temperature state is normal. Specifically, in this embodiment, a numerical comparison of the temperature information is used to determine whether the electrical connector is in an abnormal state. Specifically, the temperature value of the temperature information can be determined by checking whether the temperature value of the temperature information is within the temperature value of the standard temperature information. If so, the connector is in a normal state; if so, the connector is in an abnormal state.

[0084] In some embodiments, please refer to Figure 2 As shown, Figure 2 This is a flow chart of the training of a neural network temperature prediction model provided in the embodiment of the present application. Figure 2 As shown, the training of the neural network temperature prediction model includes the following steps a:

[0085] Step S210, obtaining historical operating parameters of the electrical connector within a preset time;

[0086] Step S220, determining historical temperature parameters of the electrical connector according to the historical operating parameters at a preset interval within the preset time;

[0087] Step S230 , using the historical temperature parameters of the electrical connector to train the neural network temperature prediction model to be trained, to obtain a trained neural network temperature prediction model.

[0088] Specifically, in step S210, the historical operating parameters acquired within a preset time period may include a curve of changes in ambient temperature and a curve of changes in current values ​​flowing through the electrical connector.

[0089] In step S220, the preset time period can generally be half a month, one month, one and a half months, or two months under a high-temperature summer load. The preset interval can be one day, i.e., 24 hours, which facilitates reflecting daily load variations. During the preset interval, historical temperature parameters of the electrical connector can be determined based on historical operating parameters. Specifically, in some embodiments, the historical operating parameters include the maximum and minimum current values ​​of the electrical connector within the preset time period, as well as the maximum and minimum temperatures of the environment in which the electrical connector is located. It will be appreciated that because the historical operating parameters are two changing curves, i.e., with a maximum and a minimum point, the maximum and minimum current values ​​of the electrical connector, as well as the maximum and minimum temperatures of the environment in which the electrical connector is located, within the preset time period can be obtained.

[0090] Thus, determining the historical temperature parameters of the electrical connector based on the historical operating parameters includes:

[0091] Based on the maximum current value, minimum current value, and resistance value of the electrical connector, the maximum and minimum heat dissipation of the electrical connector within a preset time interval are calculated, and a first temperature fluctuation range of the electrical connector within a preset time interval is calculated based on the maximum and minimum heat dissipation. Specifically, based on the three parameters required by existing heat calculation formulas for circuit components, namely, current, resistance, and time, the maximum and minimum heat dissipation of the electrical connector can be determined. Since the material and mass of the electrical connector are known parameters, the maximum and minimum temperature values ​​can be determined based on the heat dissipation, the material, and the mass of the electrical connector, thereby obtaining the first temperature fluctuation range of the electrical connector within the preset time interval.

[0092] The first temperature fluctuation range of the electrical connector within the preset time interval is coupled and compensated by the maximum temperature and the minimum temperature of the environment in which the electrical connector is located, thereby obtaining the second temperature fluctuation range of the electrical connector within the preset time interval.

[0093] In this embodiment, since the electrical connector needs to undergo heat conduction reaction with the environment, the environment will reduce the actual temperature of the electrical connector generated by the current. Therefore, the first temperature fluctuation range is coupled and compensated based on the maximum temperature and minimum temperature of the environment. It should be understood here that the electrical connector, as a medium, exchanges heat with the environment of the area to be detected, thereby causing the surface temperature of the electrical connector to be reduced to a certain extent. This is a relatively mature technology, so it will not be explained in detail. It should be understood that under normal circumstances, the operating temperature of the electrical connector can reach about 65 degrees Celsius to 80 degrees Celsius, and the ambient temperature in a normal area to be detected is lower than the operating temperature of the electrical connector. Therefore, under actual circumstances, the boundary value of the second temperature fluctuation range is smaller than the boundary value of the first temperature fluctuation range.

[0094] A historical temperature parameter of the electrical connector is determined based on the second temperature fluctuation range.

[0095] In this embodiment, since the prediction interval time is generally 24 hours a day, that is, one hour is one interval unit, the second temperature fluctuation range corresponds to a temperature value for each hour on the horizontal and vertical axes. After solving the standard deviation of these temperature values, the historical temperature parameters of the electrical connector are obtained.

[0096] In step S230, the historical temperature parameters determined in step S220 are used to train the neural network temperature prediction model to be trained to obtain a trained neural network temperature prediction model. This is a conventional operation method, mainly to improve the prediction accuracy of the neural network temperature prediction model. The training is to enable the model to better predict the temperature of the electrical connector at the next moment.

[0097] In some embodiments, a heat conduction model is established based on the structural parameters of the electrical connector, and coupling compensation is performed on the maximum temperature and minimum temperature of the environment in which the electrical connector is located based on the heat conduction model of the electrical connector to obtain a second temperature fluctuation range of the electrical connector within a preset interval.

[0098] Specifically, for the structural parameters of the electrical connector, except for the electrical conductivity of the copper material, the fluid density and the viscosity coefficient, all other component materials are regarded as isotropic homogeneous media and their physical parameters are constants; the connecting tube and the conductor of the connector are combined into one component, and its resistance is regarded as a uniform resistance; the thermal, phase and flow interactions between the heat sources are ignored; the physical field coupling effects of the electromagnetic, temperature and gas flow of the electrical connector are fully considered, and the solution domain is selected with the electrical connector connecting tube as the center; the solution model volume is controlled to avoid the interference of the boundary layer on the physical field calculation.

[0099] The electromagnetic-thermal coupling is achieved by constructing the governing equations for the electromagnetic field of the electrical connector and the relationship between the electrical conductivity and temperature of the metal material in the cable intermediate connector. The heat source for the electrical connector is generated by electromagnetic heat loss in the conductive material, while the heat source for the external environment is generated by heat dissipation from the electrical connector itself. This allows for the construction of a heat source loss formula. The fluid model uses the average Reynolds number equation, closed using the governing equations for turbulent kinetic energy and turbulent dissipation rate, and the Boussinesq approximation for air density to enhance the convergence of natural convection problems. The result is an electromagnetic-thermal-fluid coupled computational model for the electrical connector. This model, also known as a heat conduction model, is constructed using a heat transfer coefficient calculation method characterized by operating current and ambient temperature. This allows for the calculation of the heat transfer coefficient on the intermediate connector surface and the effective evaluation of the heat dissipation effect. This allows for the calculation of the operating temperature of the electrical connector, thereby determining the second temperature fluctuation range.

[0100] In some embodiments, the current operating parameters of the electrical connector are the current value flowing through the electrical connector and the temperature value of the environment at the current moment.

[0101] Specifically, please refer to the historical operating parameters of step S210. Since the current operating parameters are to be input into the neural network temperature prediction model, the current operating parameters of the electrical connector in this embodiment are the current value flowing through the electrical connector and the ambient temperature value at the current moment.

[0102] The current value flowing through the electrical connector and the ambient temperature at the current moment are input into a pre-trained neural network temperature prediction model to predict the temperature of the electrical connector at the next moment.

[0103] Specifically, since the training of the neural network temperature prediction model has been described in detail in the above embodiment, it is only necessary to input the current value flowing through the electrical connector and the ambient temperature value at the current moment into the neural network temperature prediction model to output the temperature of the electrical connector at the next moment.

[0104] Predict the degree of thermal defects of different types of electrical connectors based on the corrected temperature of the electrical connector, including:

[0105] Determining a predicted critical temperature of the electrical connector according to the type of the electrical connector;

[0106] The degree of thermal defect of the electrical joint is predicted based on the difference between the corrected temperature of the electrical joint and the critical temperature.

[0107] It should be understood here that different types of electrical connectors have critical temperatures. When the critical temperature is exceeded, thermal defects will occur in the electrical connectors, causing problems such as load reversal and mother reversal in the power grid. Since the power grid needs to operate continuously at all times to ensure that there is no power outage at the user end, maintenance of some electrical connectors with lower thermal defects can be delayed. This is because any electrical connector will be set with some safety margins when leaving the factory. For example, although the operating temperature exceeds the critical temperature, the excess is not very large, and the electrical connector can operate normally. Therefore, the degree of thermal defect of the electrical connector is predicted based on the difference between the predicted temperature of the electrical connector at the next moment and the critical temperature. For example, a temperature difference of 10℃ to 20℃ is a general thermal defect; a temperature difference of 20℃ to 40℃ is a major thermal defect; and a temperature difference ≥40℃ is a critical thermal defect. Of course, specific adjustments can be made according to the actual situation of the electrical connector, and this embodiment does not make specific limitations.

[0108] See also Figure 3 , Figure 3 This is a principle block diagram of an electrical connector thermal defect prediction system disclosed in an embodiment of the present application. Figure 3 As shown, the thermal defect prediction system may include:

[0109] The data acquisition module 310 is used to acquire historical infrared images of electrical equipment and power load sequences of the power system;

[0110] an electrical connector recognition module 320 for identifying a target area of ​​the historical infrared image containing an electrical connector using a trained electrical connector recognition model, and identifying a type of the electrical connector in the target area;

[0111] a temperature state determining module 330 for extracting temperature information of the electrical connector according to the target area, and determining a temperature state of the electrical connector based on the temperature information of the electrical connector, wherein the temperature state is a normal state or an abnormal state;

[0112] a temperature prediction module 340 for obtaining current operating parameters of the electrical connector when the temperature of the electrical connector is abnormal, inputting the current operating parameters into a pre-trained neural network temperature prediction model, and predicting the temperature of the electrical connector at a next moment, wherein the neural network temperature prediction model is trained based on historical temperature parameters of the electrical connector determined based on historical operating parameters of the electrical connector;

[0113] The load forecasting module 350 is used to input the power load sequence into a pre-trained load forecasting model to predict the power load of the power system at the next moment;

[0114] The temperature correction module 360 ​​is used to correct the temperature of the electrical connector at the next moment using the power load of the power system at the next moment to obtain the corrected temperature of the electrical connector;

[0115] The defect prediction module 370 is configured to predict the degree of thermal defects of different types of electrical connectors based on the corrected temperatures of the electrical connectors.

[0116] The electrical connector thermal defect prediction system provided by the above-described embodiment can identify target areas containing electrical connectors and the types of electrical connectors in the target areas within historical infrared images of acquired electrical equipment, thereby ensuring the accuracy of subsequent predictions of the degree of thermal defects. The system then predicts the current operating parameters of the electrical connector based on a pre-trained neural network temperature prediction model to obtain the temperature of the electrical connector at the next moment. Since the changes in the operating parameters of the electrical equipment at the next moment are uncertain, the temperature predicted based on the current operating parameters of the electrical connector using the neural network temperature prediction model may have a certain degree of accuracy. Therefore, the power load of the power system at the next moment is predicted based on the load prediction model. Based on the predicted power load, the temperature predicted based on the current operating parameters is corrected to reduce the degree of deviation in the temperature predicted based on the current operating parameters at the next moment, thereby improving the accuracy of the temperature prediction. The corrected temperatures are then used to predict the degree of thermal defects of different types of electrical connectors, providing a theoretical basis for regular inspection and testing of the connectors, thereby reducing the frequency of emergency repairs due to electrical connector overheating during the peak summer season.

[0117] See also Figure 4 , Figure 4 This is a structural diagram of an electronic device disclosed in an embodiment of the present application.

[0118] like Figure 4 As shown, the electronic device may include:

[0119] A memory 901 storing executable program code;

[0120] a processor 902 coupled to the memory 901;

[0121] The processor 902 calls the executable program code stored in the memory 901 and executes Figure 1 、 Figure 2 A method for predicting thermal defects in electrical connectors.

[0122] The present invention discloses a computer-readable storage medium storing a computer program, wherein the computer program enables a computer to execute Figure 1 、 Figure 2 A method for predicting thermal defects in electrical connectors is shown in FIG.

[0123] An embodiment of the present application further discloses an application publishing platform, wherein the application publishing platform is used to publish a computer program product, wherein when the computer program product runs on a computer, the computer executes part or all of the steps of the method in the above method embodiments.

[0124] It should be understood that the references to "one embodiment" or "an embodiment" throughout the specification mean that the specific features, structures, or characteristics associated with the embodiment are included in at least one embodiment of the present application. Therefore, the phrases "in some embodiments" or "in an embodiment" appearing throughout the specification do not necessarily refer to the same embodiment. In addition, these specific features, structures, or characteristics may be combined in any suitable manner in one or more embodiments. Those skilled in the art should also be aware that the embodiments described in the specification are all optional embodiments, and the actions and modules involved are not necessarily required for the present application.

[0125] In the various embodiments of the present application, it should be understood that the size of the sequence numbers of the above-mentioned processes does not necessarily mean the order of execution. The order of execution of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application. The units described above as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the scheme of this embodiment.

[0126] In addition, the functional units in the embodiments of the present application may be integrated into a single processing unit, or each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.

[0127] If the above-mentioned integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-accessible memory. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a memory and includes several requests for a computer device (which can be a personal computer, server or network device, etc., specifically a processor in a computer device) to execute some or all of the steps of the above-mentioned methods of various embodiments of the present application.

[0128] Those skilled in the art will appreciate that all or part of the steps in the various methods of the above embodiments can be completed by instructing related hardware through a program. The program can be stored in a computer-readable storage medium, and the storage medium includes a read-only memory (ROM), a random access memory (RAM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), a one-time programmable read-only memory (OTPROM), an electrically erasable programmable read-only memory (EEPROM), a compact disc read-only memory (CD-ROM) or other optical disc storage, magnetic disk storage, magnetic tape storage, or any other computer-readable medium capable of carrying or storing data.

[0129] The above is a detailed introduction to a method for determining defects in electrical equipment and a terminal device disclosed in an embodiment of the present application. Specific examples are used herein to illustrate the principles and implementation methods of the present application. The description of the above embodiments is only used to help understand the method of the present application and its core idea; at the same time, for general technical personnel in this field, based on the ideas of the present application, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as a limitation on the present application.

Claims

1. A method for predicting thermal defects of electrical connectors, characterized in that: include: Obtain historical infrared images of electrical equipment and power load sequences of power systems; Identify the target area containing electrical connectors in the historical infrared image using the trained electrical connector recognition model, and identify the category of the electrical connectors in the target area; extracting temperature information of the electrical connector according to the target area, and determining a temperature state of the electrical connector based on the temperature information of the electrical connector, wherein the temperature state is a normal state or an abnormal state; When the temperature state of the electrical connector is abnormal, obtaining current operating parameters of the electrical connector, inputting the current operating parameters into a pre-trained neural network temperature prediction model, and predicting the temperature of the electrical connector at a next moment, wherein the neural network temperature prediction model is trained based on historical temperature parameters of the electrical connector determined based on historical operating parameters of the electrical connector; Input the power load sequence into the pre-trained load forecasting model to predict the power load of the power system at the next moment; Correcting the temperature of the electrical connector at the next moment using the power load of the power system at the next moment to obtain a corrected temperature of the electrical connector; The thermal defect levels of different types of electrical connectors are predicted based on the corrected temperatures of the electrical connectors.

2. The method for predicting thermal defects of an electrical connector according to claim 1, wherein: The electrical connector recognition model is obtained by training a convolutional neural network using a training image set, wherein the training image set includes infrared images of electrical connectors of different categories; The load forecasting model consists of a graph convolutional neural network module, a time series convolutional neural network module and a fully connected layer.

3. The method for predicting thermal defects of an electrical connector according to claim 1, wherein: Determining a temperature state of the electrical connector based on the temperature information of the electrical connector includes: The temperature state of the electrical connector is determined by comparing the standard temperature information corresponding to the category of the electrical connector in the database with the temperature information of the electrical connector; wherein, when the temperature information of the electrical connector is greater than the standard temperature information, the electrical temperature state is an abnormal state, and when the temperature information of the electrical connector is less than and equal to the standard temperature information, the electrical temperature state is a normal state.

4. The method for predicting thermal defects of an electrical connector according to claim 1, wherein: The neural network temperature prediction model is obtained through the following training method: Obtain historical operating parameters of electrical connectors within a preset time; At a preset interval within the preset time, determining a historical temperature parameter of the electrical connector based on the historical operating parameters; The neural network temperature prediction model to be trained is trained using historical temperature parameters of the electrical connector to obtain a trained neural network temperature prediction model.

5. The method for predicting thermal defects of an electrical connector according to claim 4, wherein: The historical operating parameters include the maximum current value and the minimum current value of the electrical connector within a preset time, and the maximum temperature and the minimum temperature of the environment in which the electrical connector is located; Determining the historical temperature parameters of the electrical connector according to the historical operating parameters includes: calculating a maximum heat dissipation and a minimum heat dissipation of the electrical connector within a preset time period based on a maximum current value, a minimum current value, and a resistance value of the electrical connector, and calculating a first temperature fluctuation range of the electrical connector within a preset time interval based on the maximum heat dissipation and the minimum heat dissipation; A first temperature fluctuation range of the electrical connector within a preset time interval is coupled and compensated by the maximum temperature and the minimum temperature of the environment in which the electrical connector is located, thereby obtaining a second temperature fluctuation range of the electrical connector within the preset time interval; A historical temperature parameter of the electrical connector is determined based on the second temperature fluctuation range.

6. The method for predicting thermal defects of an electrical connector according to claim 5, characterized in that: The first temperature fluctuation range of the electrical connector within a preset time interval is coupled and compensated by the maximum temperature and the minimum temperature of the environment in which the electrical connector is located, thereby obtaining a second temperature fluctuation range of the electrical connector within the preset time interval, including: A heat conduction model is established according to the structural parameters of the electrical connector. The maximum temperature and the minimum temperature of the environment in which the electrical connector is located are coupled and compensated according to the heat conduction model of the electrical connector to obtain a second temperature fluctuation range of the electrical connector within a preset interval time.

7. The method for predicting thermal defects of an electrical connector according to claim 1, wherein: The current operating parameters of the electrical connector are the current value flowing through the electrical connector and the ambient temperature value at the current moment; The current value flowing through the electrical connector and the ambient temperature value at the current moment are input into a pre-trained neural network temperature prediction model to predict the temperature of the electrical connector at the next moment; Predict the degree of thermal defects of different types of electrical connectors based on the corrected temperature of the electrical connector, including: Determining a predicted critical temperature of the electrical connector according to the type of the electrical connector; The degree of thermal defect of the electrical joint is predicted based on the difference between the corrected temperature of the electrical joint and the critical temperature.

8. An electrical connector thermal defect prediction system, characterized in that: include: A data acquisition module is used to obtain historical infrared images of electrical equipment and power load sequences of the power system; an electrical connector recognition module, configured to identify a target area of ​​the historical infrared image containing electrical connectors using a trained electrical connector recognition model, and to identify the type of electrical connector in the target area; a temperature state determining module, configured to extract temperature information of the electrical connector according to the target area, and determine a temperature state of the electrical connector based on the temperature information of the electrical connector, wherein the temperature state is a normal state or an abnormal state; a temperature prediction module for obtaining current operating parameters of the electrical connector when the temperature of the electrical connector is abnormal, inputting the current operating parameters into a pre-trained neural network temperature prediction model, and predicting the temperature of the electrical connector at a next moment, wherein the neural network temperature prediction model is trained based on historical temperature parameters of the electrical connector determined based on historical operating parameters of the electrical connector; The load forecasting module is used to input the power load sequence into the pre-trained load forecasting model to predict the power load of the power system at the next moment; a temperature correction module, configured to correct the temperature of the electrical connector at the next moment using the power load of the power system at the next moment, thereby obtaining the corrected temperature of the electrical connector; The defect prediction module is used to predict the degree of thermal defects of different types of electrical connectors based on the corrected temperature of the electrical connectors.

9. An electronic device comprising: A memory and a processor, wherein a computer program is stored on the memory, and wherein the computer program can be executed by the processor so that the processor implements a method for predicting thermal defects of an electrical connector as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, which is suitable for being loaded and executed by a processor, so that the processor implements the method for predicting thermal defects of an electrical connector according to any one of claims 1 to 7.

Images