Cable joint state evaluation method and system based on digital twinning technology

Abstract

The invention provides a method and a system for evaluating the state of a cable joint based on a digital twinning technology. The method comprises the following steps: constructing a multi-physical field coupling simulation model of an electromagnetic field and a temperature field of the cable joint; simulation data of the cable joint under different working conditions and environmental conditions are obtained, and ultrasonic information of the cable is obtained through a cable entity sensor; using simulation data to train and test the multi-physics coupling simulation model by applying a U-net deep learning algorithm to obtain a simulation agent model; extracting parameter characteristics of the agent model simulation data and the ultrasonic information; constructing a comprehensive state quantitative evaluation model of the cable joint according to a comparison result of the parameter characteristics and a preset threshold value; and obtaining health state evaluation data of the to-be-evaluated cable joint through the multi-physics field coupling simulation model and the comprehensive state quantitative evaluation model of the cable joint. The method can comprehensively and accurately evaluate the state of the cable joint, and has the capabilities of real-time monitoring and fault prediction.

Description

Technical Field

[0001] The present invention relates to the technical field of smart grid and power system management, and in particular to a state assessment method and system for a cable joint based on digital twin technology. Background Art

[0002] With the development of power systems and the increasing demand for power reliability, the evaluation and monitoring of the working status of high-voltage cable joints, as an important part of power transmission, is particularly important.

[0003] However, the existing cable joint status assessment methods have the following problems: First, the single data source problem. Most current status assessment methods rely on a single type of sensor data and lack comprehensive monitoring of the multi-physical fields of cable joints. This single data source may lead to inaccurate assessment results and failure to detect potential faults in a timely manner; second, insufficient data processing and analysis capabilities. Existing assessment systems often lack advanced data processing and analysis capabilities and are unable to effectively extract and utilize key features in sensor data. This results in the judgment of the cable joint status relying on manual experience, which is highly subjective and has low accuracy; third, the lack of dynamic monitoring and prediction capabilities. Traditional methods are mostly regular inspections, which cannot achieve real-time monitoring and dynamic evaluation of cable joints. For sudden failures that may occur in cable joints, it is impossible to provide timely warnings and processing, which poses a safety hazard. Summary of the invention

[0004] In view of the above technical problems, the present invention provides a method and system for evaluating the state of a cable joint based on digital twin technology, the method comprising:

[0005] Construct a multi-physics coupling simulation model of the electromagnetic field and temperature field of the cable joint;

[0006] The simulation data of the cable joint under different working conditions and environmental conditions are obtained through finite element simulation, and the ultrasonic signal inside the cable joint is obtained by arranging sensors on the surface of the cable entity;

[0007] Using the simulation data, training and testing the simulation agent model based on the U-net deep learning algorithm to obtain a simulation agent model;

[0008] Extracting parameter features of the simulation data and ultrasonic information of the simulation agent model; constructing a comprehensive state quantitative evaluation model of the cable joint according to the comparison result of the parameter features and the preset threshold value;

[0009] The health status evaluation data of the cable joint to be evaluated is obtained through the comprehensive status quantitative evaluation model of the cable joint.

[0010] Furthermore, a multi-physics field coupling simulation model of the electromagnetic field and temperature field of the cable connector is constructed, including:

[0011] According to the physical dimensions and material parameters of the cable joint entity, a multi-physics field coupling simulation model of the electromagnetic field and temperature field of the cable joint is constructed.

[0012] Furthermore, finite element simulation is used to obtain simulation data of cable connectors under different working conditions and environmental conditions, including:

[0013] The multi-physical field coupling simulation model is used to solve the coupling problem of the electromagnetic field and the temperature field on a finite element simulation platform to obtain simulation data of the cable connector.

[0014] Furthermore, the multi-physics field coupling simulation model is used to solve the coupling problem of the electromagnetic field and the temperature field on a finite element simulation platform to obtain simulation data of the cable connector, including:

[0015] The control variable method is used to analyze the distribution of the electric field and temperature field of the cable joint under different current carrying conditions, convection heat transfer coefficients and ambient temperatures.

[0016] According to the affected distribution, the cable joint simulation data under different operating and environmental conditions are obtained.

[0017] Further, the simulation data is used to train and test the simulation agent model based on the U-net deep learning algorithm to obtain a simulation agent model, including:

[0018] The simulation data includes temperature distribution simulation data and electromagnetic field distribution simulation data;

[0019] Determine an experimental group training data set and a test group data set according to the temperature simulation data and the electromagnetic field distribution simulation data, and divide the experimental group data into a training set and a validation set according to a preset ratio;

[0020] Obtaining basic parameters corresponding to the temperature distribution simulation data and the electromagnetic field distribution simulation data;

[0021] Using the basic parameters as input parameters of a U-net deep learning algorithm, and training the simulation agent model based on the deep learning algorithm according to the training set and the test set;

[0022] Optimizing the deep learning model using an adaptive moment estimation optimization algorithm;

[0023] The test group data set is used to verify the simulation agent model based on the U-net deep learning algorithm.

[0024] Furthermore, extracting the parameter features of the simulation proxy model simulation data and ultrasound information includes:

[0025] The simulation data and ultrasonic information are processed and feature extracted to obtain key operating parameter characteristics of the cable under real-time operating conditions, wherein the parameter characteristics include temperature gradient, ultrasonic time-frequency signal and electric field strength.

[0026] Furthermore, according to the comparison result between the parameter characteristics and the preset threshold value, a comprehensive state quantitative evaluation model of the cable joint is constructed, including:

[0027] Set a threshold for each feature to determine whether the abnormal state has been reached, and deduct the score based on the comparison between the feature value and the threshold;

[0028] The starting score of the cable joint operation status is set to 100 points, and points are deducted according to the degree to which the characteristic value exceeds the threshold;

[0029] According to the scoring criteria and deduction mechanism, the health status of the cable joint is quantified into a score, and the score is divided into three categories: normal state, warning state and abnormal state according to the preset threshold, so as to construct a comprehensive state quantitative evaluation model of the cable joint.

[0030] The present invention also provides a cable joint status assessment system based on digital twin technology, which is characterized by comprising:

[0031] A simulation model building module is used to build a multi-physics field coupling simulation model of the electromagnetic field and temperature field of the cable joint;

[0032] A simulation data and ultrasonic data acquisition module is used to acquire simulation data of the cable joint under different working conditions and environmental conditions through finite element simulation, and to acquire ultrasonic signals inside the cable joint by arranging sensors on the surface of the cable entity;

[0033] A training and testing module, used to train and test the simulation agent model based on the U-net deep learning algorithm using the simulation data to obtain a simulation agent model;

[0034] An evaluation model building module is used to extract parameter features of the simulation data and ultrasonic information of the simulation proxy model; and to build a comprehensive state quantitative evaluation model of the cable joint according to the comparison results of the parameter features and the preset threshold values;

[0035] The evaluation module is used to obtain health status evaluation data of the cable joint to be evaluated through the comprehensive state quantitative evaluation model of the cable joint.

[0036] Further, training and testing modules include:

[0037] A simulation data determination submodule, used to determine that the simulation data includes temperature distribution simulation data and electromagnetic field distribution simulation data;

[0038] The training set and validation set division submodule is used to determine the experimental group training data set and the test group data set according to the temperature simulation data and the electromagnetic field distribution simulation data, and divide the experimental group data into the training set and the validation set according to a preset ratio;

[0039] A basic parameter acquisition submodule is used to acquire basic parameters corresponding to the temperature distribution simulation data and the electromagnetic field distribution simulation data;

[0040] A training submodule, used for taking the basic parameters as input parameters of the U-net deep learning algorithm, and training the simulation agent model based on the U-net deep learning algorithm according to the training set and the test set;

[0041] An optimization submodule, used to optimize the deep learning model using an adaptive moment estimation optimization algorithm;

[0042] The verification submodule is used to verify the simulation agent model based on the U-net deep learning algorithm using the test group data set.

[0043] Furthermore, the evaluation model building modules include:

[0044] The first score deduction setting submodule is used to set the threshold of each feature, which is used to determine whether the standard of abnormal state is reached, and deduct the score according to the comparison result between the feature value and the threshold;

[0045] A second score deduction setting submodule is used to set the starting score of the operating status of the cable connector to 100 points, and deduct points according to the degree to which the characteristic value exceeds the threshold;

[0046] The scoring quantification submodule is used to quantify the health status of the cable joint into a score according to the scoring criteria and deduction mechanism. The score is divided into three categories: normal state, warning state and abnormal state according to the preset threshold, so as to construct a comprehensive state quantification evaluation model for the cable joint.

[0047] The present invention provides a method and system for evaluating the state of a cable joint based on digital twin technology. By introducing digital twin technology, the method and its affiliated platform effectively solve the problems of single data source, insufficient data processing and analysis capabilities, and lack of dynamic monitoring and prediction capabilities in existing methods. It can provide comprehensive and accurate state evaluation, and has real-time monitoring and fault prediction capabilities, providing a strong guarantee for the safe and reliable operation of the power system. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 It is a flow chart of a method for evaluating the state of a cable joint based on digital twin technology provided by an embodiment of the present invention;

[0049] Figure 2 This is an application scenario diagram of the cable joint status assessment method provided by an embodiment of the present invention;

[0050] Figure 3 It is a multi-physics field calculation flow chart of a cable connector provided by an embodiment of the present invention;

[0051] Figure 4 It is a flow chart of a cable joint multi-physics field deep learning prediction model provided by an embodiment of the present invention;

[0052] Figure 5 It is a schematic diagram of the U-net neural network structure provided by an embodiment of the present invention;

[0053] Figure 6 It is a structural schematic diagram of a cable joint status assessment system based on digital twin technology provided by an embodiment of the present invention; DETAILED DESCRIPTION

[0054] Many specific details are set forth in the following description to facilitate a full understanding of the present invention. However, the present invention can be implemented in many other ways than those described herein, and those skilled in the art can make similar generalizations without violating the connotation of the present invention, so the present invention is not limited to the specific implementation disclosed below.

[0055] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0056] Example 1

[0057] A cable joint status assessment method based on digital twin technology is provided, which can be applied to Figure 2 In the application scenario shown. The method communicates with the server through the network, and the data storage system can store the data that the server needs to process. The data storage system can be integrated on the server, or placed on the cloud or other network servers, all of which belong to the data layer processing.

[0058] In the specific implementation, the server builds an electromagnetic-thermal multi-physics field coupling simulation model for cable joints and uses the model to obtain simulation data of cable joints. Based on the simulation data, a deep learning algorithm is used to train the multi-physics field coupling simulation model to obtain a deep learning prediction model. Then, a comprehensive state quantitative evaluation model for high-voltage cable joints is constructed.

[0059] Based on the deep learning prediction model and the comprehensive state quantitative evaluation model of high-voltage cable joints, the health status data of cable joints can be obtained on the visualization platform. The server can be implemented using a stand-alone server or a server cluster consisting of multiple servers.

[0060] like Figure 1 As shown in the figure, a cable joint status assessment method based on digital twin technology is provided. Figure 2 The server application in is used as an example to illustrate the following steps:

[0061] Step S101, constructing a multi-physics field coupling simulation model of the electromagnetic field and temperature field of the cable connector.

[0062] Specifically, first of all, as a key component in the high-voltage power system, the cable joint is subject to multiple influences of loads and environmental conditions. During operation, the cable joint may be affected by multiple physical fields such as electromagnetic-thermal, resulting in performance degradation or failure. Therefore, the purpose of constructing an electromagnetic-thermal multi-physics field coupling simulation model of the cable joint is to better understand and predict the interactions and influences between these physical fields, so as to guide the design and optimization of the cable joint and improve its operational reliability and service life.

[0063] The multi-physics field coupling simulation model can accurately simulate the actual situation of the cable joint under different working conditions and environments, and analyze the influence of electromagnetic field and thermal field on the temperature distribution and electromagnetic field distribution inside the joint. The voltage level of the high-voltage cable joint is 110kV. In this way, not only can the hot spots and concentrated areas of high electric field strength that may appear in the cable joint during operation be identified, but also the occurrence of these problems can be reduced by adjusting design parameters and improving manufacturing processes, thereby improving the overall performance and safety of the cable joint.

[0064] The multi-physics simulation model can also provide a large amount of training data for the deep learning algorithm. By analyzing and learning these simulation data, the deep learning prediction model and health status assessment model of the cable joint can be constructed. Combined with the actual sensor data, the working status of the cable joint can be monitored in real time, potential faults can be discovered and warned in time, and unexpected shutdowns and losses of the power system can be avoided.

[0065] According to the information such as the physical size and material parameters of the cable joint entity, a three-dimensional geometric model of the cable joint and a multi-physical field coupling simulation model of the electromagnetic field and temperature field are constructed.

[0066] In the simulation model, material parameters are set to be consistent with the physical parameters of the actual cable connector entity; the boundary conditions of the electromagnetic field are set, where the electric field is set using the "current (ec)" and "magnetic field (mf)" components, the cable excitation voltage is set using the "potential", and the ground layer is set using the "grounding" setting. The main functional expressions of the electromagnetic field include:

[0067]

[0068] Where J is the total current density, σ is the conductivity, E is the electric field intensity, Je is the applied current density, Q V is the body charge density, H is the magnetic field intensity, B is the magnetic induction intensity, and A is the magnetic vector potential;

[0069] Set the boundary conditions of the temperature field, and set the initial temperature to 293.15K. Define the entire cable joint area as a generalized heat source, which is generated by the electromagnetic loss in the insulating material; at the same time, the copper conductor connection point in the cable joint is also designated as a heat source. It should be noted that the equivalent conductivity of the crimping tube at the copper conductor connection point at the joint is different from the conductivity of the copper conductor of the cable body. The main functional expressions of the temperature field include:

[0070]

[0071] Where ρ is the density of the material; C p is the constant pressure specific heat capacity of the material; u is the velocity vector. If the material itself is moving, a convection term is introduced; is the temperature gradient; q is the heat flux vector; Q is the heat generated inside the unit volume; Q ted The amount of heat added or removed from an external source per unit volume.

[0072] The key to multi-physics coupling is to deal with the interaction and influence between different physical fields. In order to perform multi-physics coupling simulation, it is necessary to select appropriate simulation software, such as ANSYS, COMSOL Multiphysics, etc. These software usually provide rich physical field modules and powerful solvers, which can facilitate multi-physics modeling and simulation.

[0073] Step S102, obtaining simulation data of the cable joint under different working conditions and environmental conditions through finite element simulation, and obtaining the acoustic wave signal inside the cable joint by arranging sensors on the surface of the cable entity.

[0074] like Figure 3As shown, the multi-physics field coupling simulation model is used to solve the coupling problem of the electromagnetic field and the temperature field on the finite element simulation platform to obtain the simulation data of the cable joint. By running the simulation model, simulation data of the cable joint under different working conditions and environmental conditions can be obtained. These data include electromagnetic field distribution, temperature field distribution, etc., which reflect the internal state of the cable joint during the working process;

[0075] Specifically, the above construction obtains a finite element simulation model of a multi-physical field, and uses the control variable method to analyze the influence of different operating conditions and environmental conditions on the operating state of the cable joint. According to the affected conditions, the simulation data of the cable joint is obtained, including:

[0076] The initial parameters of the multi-physics coupling simulation model are set, including the setting of parameters such as ambient temperature, initial conductivity, and heat flux. The electric field boundary conditions and temperature field boundary conditions are applied first, and the electromagnetic field simulation of the cable joint is analyzed. When current flows through the conductor, the electromagnetic field generates a heat source through Joule loss and dielectric loss, thereby affecting the distribution of the thermal field, and the internal temperature of the cable will affect the resistivity of the conductor. Therefore, in order to obtain the accurate electromagnetic field and temperature distribution of the cable joint under the electromagnetic-thermal multi-physics field coupling, it is necessary to continuously iterate and calculate based on the coupling relationship between the two, and then determine whether the convergence conditions are met; if not, continue to iterate based on the electric field boundary conditions and temperature field boundary conditions. If they are met, obtain the temperature distribution data of each unit node; finally, when the iteration is completed, output the results, which include the temperature distribution data and electromagnetic field distribution data of each unit node.

[0077] In actual operation, the distribution of the electric field and temperature field of the cable will also be different under different loads, convection heat transfer coefficients and environmental conditions. According to IEEE Std 835-1994 "Standard Power Cable Current Carrying Table", the rated current range of 110kV cable is 800-1200A. By adopting the control variable method, the distribution of the electric field and temperature field of the cable joint under different current carrying conditions, convection heat transfer coefficients and ambient temperatures is analyzed respectively; according to the affected distribution, the simulation data of the cable joint under different operating and environmental conditions is obtained. It is also possible to accurately analyze the influence of the cable joint on the electromagnetic field and temperature field, ensuring that only one physical field parameter is changed each time, thereby avoiding interference and errors caused by the simultaneous changes of multiple variables. The simulation analysis of the influence of different load conditions on the temperature distribution and electromagnetic field distribution of the cable joint was carried out. The simulation of the cable joint under different operating conditions was set to 800-1200A, with a gradient of 10A; the simulation analysis of the influence of different convection heat transfer coefficients on the temperature distribution and electromagnetic field distribution of the cable joint was carried out. The simulation of the cable joint under different convection heat transfer coefficients was set to 5-25W / m2·K, with a gradient of 1W / m2·K; at the same time, the simulation analysis of the influence of different ambient temperatures on the temperature distribution and electromagnetic field distribution of the cable joint was carried out. The simulation of the cable joint under different ambient temperatures was set to 293.15-308.15K, with a gradient of 1K. COMSOL performs parametric modeling and analytical calculations

[0078] Step S103: Use the simulation data to train and test the simulation agent model based on the U-net deep learning algorithm to obtain a simulation agent model.

[0079] The simulation data includes temperature distribution simulation data and electromagnetic field distribution simulation data;

[0080] Determine an experimental group training data set and a test group data set according to the temperature simulation data and the electromagnetic field distribution simulation data, and divide the experimental group data into a training set and a validation set according to a preset ratio;

[0081] Obtaining basic parameters corresponding to the temperature distribution simulation data and the electromagnetic field distribution simulation data;

[0082] Using the basic parameters as input parameters, and training the simulation agent model based on the U-net deep learning algorithm according to the training set and the test set;

[0083] Optimizing the deep learning model using an adaptive moment estimation optimization algorithm;

[0084] The test group data set is used to verify the simulation agent model based on the U-net deep learning algorithm.

[0085] The multi-physics coupling simulation model is established based on physical principles and mathematical equations. The mathematical model can well describe the working conditions of the cable connector. The deep learning algorithm is a powerful data-driven modeling method that can automatically extract features from large amounts of data and learn complex mapping relationships.

[0086] The purpose of using deep learning algorithms to train simulation data is to enable the deep learning model to learn the laws and patterns contained in the simulation data, so as to establish a mapping relationship with cable load, convection heat transfer coefficient, and ambient temperature as input and temperature-electromagnetic field distribution as output. The trained multi-physics field coupling simulation model becomes a prediction model that can accept new input data, namely the real-time changing load, convection heat transfer coefficient, and ambient temperature of the cable, and quickly calculate the corresponding output prediction results.

[0087] The above-mentioned simulation data of the cable joint includes temperature distribution simulation data and electromagnetic field distribution simulation data; the deep learning algorithm is an image semantic segmentation neural network; according to the simulation data, a deep learning algorithm is used to train a multi-physics field coupling simulation model, such as Figure 4 As shown, including:

[0088] According to the temperature distribution simulation data and the electromagnetic field distribution simulation data, determine the experimental group training data set and the test group data set, and divide the experimental group data into a training set and a validation set according to a preset ratio;

[0089] Obtaining basic parameters corresponding to temperature distribution simulation data and electromagnetic field distribution simulation data;

[0090] The basic parameters are used as input parameters of the image semantic segmentation neural network, and the multi-physics field coupling simulation model is trained based on the training set and the validation set;

[0091] Adopting an adaptive moment estimation optimization algorithm to optimize the deep learning model;

[0092] The test group data set is used to verify the simulation agent model based on the U-net deep learning algorithm.

[0093] Specifically, before determining the training data set and the test data set, in order to improve the model training efficiency and accelerate the network convergence speed, the Z-score normalization method is used to normalize the simulation data. The formula for Z-score normalization is as follows:

[0094]

[0095] Among them, Z is the standardized feature quantity, X represents the feature data before normalization; μ represents the mean of the feature data; σ represents the standard deviation of the feature data.

[0096] The experimental group data set is used to build and train the model, while the test group data set is used to evaluate the performance of the model. After determining the experimental group data set, it needs to be further divided into a training set and a validation set. The training set is used to train the model by adjusting the model's parameters to minimize the prediction error. The validation set is used to evaluate the performance of the model during the training process to help select the optimal model structure and parameter configuration. The ratio of the training set and the validation set is usually determined based on the specific problem and the size of the data set. In general, the training set should account for the majority of the experimental group data so that the model can fully learn the characteristics of the data; the validation set accounts for a smaller part and is used for performance monitoring and model optimization during the training process.

[0097] Basic parameters include but are not limited to: ambient temperature, load current, and cable connector surface convection heat transfer coefficient.

[0098] like Figure 5 As shown, the image semantic segmentation neural network (U-net neural network) is used for training, which is an improved form of the fully convolutional neural network (FCN). As a deep learning model, U-net effectively extracts high-dimensional nonlinear features in the data through convolution operations. The network mainly consists of a feature extraction part, a feature enhancement part, and a prediction part. The feature extraction and feature enhancement parts have symmetrical structures.

[0099] The feature extraction part consists of convolutional layers and maximum pooling layers. Each convolutional layer is followed by a Relu activation function to downsample the features. During the downsampling process, the number of channels is doubled each time, thereby retaining the important feature information in the downsampling process to the greatest extent. The operation formula of the convolutional layer is as follows:

[0100] Conv(x)=W*x+b

[0101] Where W is the convolution kernel, * represents the convolution operation, x is the input, and b is the bias term. The ReLU activation function formula is:

[0102] ReLU(x)=max(0,x)

[0103] The feature enhancement part consists of a deconvolution layer, a maximum pooling layer, and a Relu activation function. By upsampling the features, the important features obtained by the feature extraction part are upsampled twice and fused, so as to retain the key information in the downsampling process to the maximum extent, and finally form an effective feature layer with the same size as the input. The deconvolution operation formula is as follows:

[0104] Deconv(x)=W deconv *x+b

[0105] Among them, W deconv is the deconvolution kernel.

[0106] Through this structure, the U-net neural network can effectively process multi-scale features while ensuring high accuracy, and is suitable for complex image segmentation tasks.

[0107] The U-net neural network takes the basic parameters as input and the corresponding feature vector values ​​as output information, and uses the training set and validation set to train the multi-physics field coupling simulation model. During the training process, the training loss function and the validation loss function are used as the main evaluation indicators. First, the initial range of the number of hidden layer nodes in the model is determined by the empirical formula, and then the grid search method is used for parameter optimization. Using the mean square error (RMSE) as the loss function, the simple cross-validation method is used to determine the optimal number of hidden layer nodes and the optimal number of training iterations by analyzing the changing trends of the training loss and validation loss. The optimization algorithm directly uses Adam (adaptive moment estimation optimization algorithm) as the optimizer of the deep learning prediction model. Finally, by setting and adjusting hyperparameters such as learning rate, learning rate decay factor, decay cycle, batch size, and number of iterations, the model is ensured to achieve optimal performance.

[0108] The processed test dataset is then input into a data-driven deep learning multi-physics simulation agent model to verify the accuracy and effectiveness of the model.

[0109] Step S104, extracting parameter features of the simulation proxy model simulation data and ultrasonic information; and constructing a comprehensive state quantitative evaluation model of the cable joint according to a comparison result between the parameter features and a preset threshold value.

[0110] The simulation agent model simulation data and ultrasonic information are processed and feature extracted to obtain key operating parameter characteristics of the cable under real-time operating conditions, wherein the parameter characteristics include temperature gradient, ultrasonic time-frequency signal and electric field strength.

[0111] Constructing a comprehensive state quantitative evaluation model for cable joints is a process of modeling the performance degradation and failures of cable joints caused by various factors during actual operation. The comprehensive state quantitative evaluation model is an important tool for evaluating the health status of cable joints. It is based on factors such as the physical properties, working environment conditions, and load conditions of cable joints. It uses physical mathematical models and deep learning algorithms to describe the changes in the state and health status of cable joints during long-term operation. The model extracts features and performs quantitative scoring through the fusion of the simulation agent model based on the U-net deep learning algorithm and actual sensor data, thereby achieving accurate evaluation and real-time monitoring of the comprehensive health status of cable joints.

[0112] The collected data are preprocessed to extract key features from the data, such as temperature gradient, ultrasonic time-frequency signal, and electric field strength.

[0113] According to the common fault types of high-voltage cable joints, temperature gradient, ultrasonic time-frequency information, and electric field strength are selected as characteristic parameters for quantitative evaluation. Among them, the temperature gradient reflects the temperature changes in different areas of the cable joint and is an important indicator for evaluating overheating and insulation aging; ultrasonic time-frequency information is obtained by ultrasonic sensors to obtain the sound wave signal inside the joint. Analysis of its time-frequency characteristics can detect internal structural changes and potential cracks; electric field strength is an important indicator for measuring the electric field distribution inside the cable joint and evaluating the impact of electrical stress on the joint material.

[0114] Fluctuations in these characteristic parameters over time, working environment or load may lead to the occurrence of failures, which can be a cumulative process ranging from minor to significant.

[0115] Among them, the extraction formula of temperature gradient feature is as follows:

[0116]

[0117] Where T is temperature and x, y, z are spatial coordinates.

[0118] The characteristic parameter extraction of ultrasonic time-frequency signals is usually analyzed using Short-Time Fourier Transform (STFT). Its basic form is as follows:

[0119]

[0120] Among them, x(t) is the input signal, usually an ultrasonic signal; w(t) is the window function, which is used to limit the effective length of the signal in time; X(t,f) is the Fourier transform result at time t and frequency f, which represents the spectral content of the signal.

[0121] The extraction formula of the characteristic parameter of electric field intensity is as follows:

[0122]

[0123] Among them, E x , E y , E z are the components of the electric field in each direction.

[0124] Step S105, obtaining health status evaluation data of the cable joint to be evaluated through the multi-physical field coupling simulation model and the comprehensive state quantification evaluation model of the cable joint.

[0125] A threshold for each feature is set to determine whether the standard for abnormal status is met, and the score is deducted based on the comparison result between the feature value and the threshold. The starting score for the operating status of the cable joint is set to 100 points, and points are deducted based on the degree to which the feature value exceeds the threshold. According to the scoring standard and deduction mechanism, the health status of the cable joint is quantified into a score, and the score is divided into three categories: normal state, warning state and abnormal state according to the preset threshold, thereby constructing a comprehensive state quantitative evaluation model for the cable joint.

[0126] The quantitative evaluation of the comprehensive state of high-voltage cable joints mainly conducts quantitative evaluation on the above three characteristic quantities. The deduction values ​​for the specific characteristics of the three characteristic quantities that can reflect the health state of the cable joints are defined as:

[0127]

[0128] Among them, x i is the value of the i-th feature, T i is the threshold of the feature, k i is the deduction coefficient for this feature.

[0129] Then, the deduction value is calculated according to the degree to which the feature value exceeds the threshold. The starting score of each feature is 100 points, and the corresponding deduction is made according to the magnitude of the excess threshold. That is, the state quantitative evaluation formula of the model is as follows:

[0130]

[0131] Among them, S represents the comprehensive status score, n is the number of features, and p i Represents the penalty value of the i-th feature.

[0132] The comprehensive state quantitative assessment model combines these predicted values ​​with real-time sensor data such as ultrasonic information, and then extracts characteristic parameters that can reflect the operating status of the cable, including temperature gradient, electric field strength, and ultrasonic time-frequency information. Through the preset scoring criteria and deduction mechanism, the comprehensive assessment model can classify the health status of the cable joint into three levels: normal, caution, and abnormal.

[0133] In the specific example of the above scheme, the health status data of the cable joint is obtained according to the deep learning prediction model and the state quantification comprehensive evaluation model. In order to characterize the health status of the cable joint, the comprehensive status score S is used to calculate the health status of the cable joint, which is proposed to be calculated by the following formula:

[0134]

[0135] With the comprehensive status score S, the system has a digital standard for dividing the health level, which can vividly reflect the health status assessment results of the cable joint in the form of numerical values. As shown in the following table, a comprehensive status score greater than 80 indicates that the cable joint is in normal health and in good operating condition; when the comprehensive status score is between 60 and 80, the cable joint is in a state of caution, and the load of the cable joint should be adjusted appropriately according to the situation; a comprehensive status score lower than 60 indicates that the cable joint is in an abnormal working state and there is a risk of failure. The higher the comprehensive status score, the better the health status of the cable joint.

[0136]

[0137]

[0138] Specifically, the process of developing the encapsulated cable joint status assessment platform is carried out in the form of software. This platform converts statistical images and cable joint health status data into mathematical modeling form, and further writes it into program running code. Through optimization processing to ensure that the program runs correctly and smoothly, it is finally integrated into the Python environment to form a complete cable joint status assessment platform. After setting the operating conditions, the platform can quickly and automatically complete the evaluation and classification of the cable joint health status.

[0139] When designing the GUI, the program running code is integrated into the black box. In addition to beautifying the interface design, the user interface mainly retains the basic simulation parameter input interface and the result data and image display interface. In this way, users do not need to understand the simulation principle or the state assessment principle, but only need to enter the basic parameters in the GUI software interface. The integrated platform will automatically calculate the temperature distribution simulation data and electromagnetic field distribution simulation data of the cable joint, and then complete the health status assessment and display the results.

[0140] The software adopts a paged display format, including the main interface, cable connector and working condition setting interface, data import interface and status result display interface.

[0141] The main interface serves as the cover, with "Start", "Help" and "Close" buttons embedded in it. Click Help to view the help document, which contains detailed usage tutorials and internal working principles.

[0142] The operating conditions and physical parameters can be set in the cable connector and working condition setting interface. The operating conditions include current, convection heat transfer coefficient, and ambient temperature. The internal cross-section of the cable is provided in the interface for reference.

[0143] The data import module is used to collect and prepare sensor data and simulation data of cable joints, including key information such as temperature distribution and electromagnetic field distribution.

[0144] The status result display interface will display the simulation data processed by the deep learning prediction model and the ultrasonic information collected by the cable entity sensor, as well as statistical charts after data preprocessing and feature extraction. Through the high-voltage cable joint status assessment platform, these statistical charts and the health status data of the cable joint are presented, allowing users to intuitively understand the status of the joint. Under normal working conditions, the prompt bar is always green; when in the attention state, the prompt bar is always yellow; when an abnormality occurs, the prompt bar will flash red to remind users that there may be a risk of failure in the joint.

[0145] Each module of the cable joint health status assessment platform can be implemented by software, hardware or a combination thereof. These modules can be embedded in or independent of the processor of the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the corresponding operations of each module.

[0146] Example 2

[0147] Based on the same inventive concept, the present invention also provides a cable joint status assessment system 600 based on digital twin technology, such as Figure 6 As shown, including:

[0148] A simulation model building module 610 is used to build a multi-physics field coupling simulation model of the electromagnetic field and temperature field of the cable connector;

[0149] The simulation data and ultrasonic data acquisition module 620 is used to acquire simulation data of the cable joint under different working conditions and environmental conditions through finite element simulation, and to acquire ultrasonic signals inside the cable joint by arranging sensors on the surface of the cable entity;

[0150] A training and testing module 630 is used to train and test the simulation agent model based on the U-net deep learning algorithm using the simulation data to obtain a simulation agent model;

[0151] The evaluation model building module 640 is used to extract the parameter characteristics of the simulation data and the ultrasonic information of the simulation agent model; and to build a comprehensive state quantitative evaluation model of the cable joint according to the comparison result between the parameter characteristics and the preset threshold value;

[0152] The evaluation module 650 is used to obtain health status evaluation data of the cable joint to be evaluated through the comprehensive state quantitative evaluation model of the cable joint.

[0153] Further, training and testing modules include:

[0154] A simulation data determination submodule, used to determine that the simulation data includes temperature distribution simulation data and electromagnetic field distribution simulation data;

[0155] The training set and validation set division submodule is used to determine the experimental group training data set and the test group data set according to the temperature simulation data and the electromagnetic field distribution simulation data, and divide the experimental group data into the training set and the validation set according to a preset ratio;

[0156] A basic parameter acquisition submodule is used to acquire basic parameters corresponding to the temperature distribution simulation data and the electromagnetic field distribution simulation data;

[0157] A training submodule, used for training the simulation agent model based on the U-net deep learning algorithm according to the training set and the test set by taking the basic parameters as input parameters;

[0158] An optimization submodule, used to optimize the deep learning model using an adaptive moment estimation optimization algorithm;

[0159] The verification submodule is used to verify the simulation agent model based on the U-net deep learning algorithm using the test group data set.

[0160] Furthermore, the evaluation model building modules include:

[0161] The first score deduction setting submodule is used to set the threshold of each feature, which is used to determine whether the standard of abnormal state is reached, and deduct the score according to the comparison result between the feature value and the threshold;

[0162] A second score deduction setting submodule is used to set the starting score of the operating status of the cable connector to 100 points, and deduct points according to the degree to which the characteristic value exceeds the threshold;

[0163] The scoring quantification submodule is used to quantify the health status of the cable joint into a score according to the scoring criteria and deduction mechanism. The score is divided into three categories: normal state, warning state and abnormal state according to the preset threshold, so as to construct a comprehensive state quantification evaluation model for the cable joint.

[0164] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0165] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0166] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0167] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0168] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the relevant field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents, and any modifications or equivalents that do not depart from the spirit and scope of the present invention should be included in the scope of the claims of the present invention.

Claims

1. A method for evaluating the state of a cable joint based on digital twin technology, characterized in that: include: Construct a multi-physics coupling simulation model of the electromagnetic field and temperature field of the cable joint; The simulation data of the cable joint under different working conditions and environmental conditions are obtained through finite element simulation, and the ultrasonic signal inside the cable joint is obtained by arranging sensors on the surface of the cable entity; Using the simulation data, applying the U-net deep learning algorithm to the multi-physics field coupling simulation model for training and testing to obtain a simulation proxy model; Extracting parameter features of simulation data and ultrasound information of the simulation proxy model; According to the comparison result of the parameter characteristics and the preset threshold value, a comprehensive state quantitative evaluation model of the cable joint is constructed; The health status evaluation data of the cable joint to be evaluated is obtained through the comprehensive status quantitative evaluation model of the cable joint.

2. The method according to claim 1, characterized in that Construct a multi-physics coupling simulation model of the electromagnetic field and temperature field of the cable connector, including: According to the physical dimensions and material parameters of the cable joint entity, a multi-physics field coupling simulation model of the electromagnetic field and temperature field of the cable joint is constructed.

3. The method according to claim 1, characterized in that: Finite element simulation is used to obtain simulation data of cable joints under different working conditions and environmental conditions, including: The multi-physical field coupling simulation model is used to solve the coupling problem of the electromagnetic field and the temperature field on a finite element simulation platform to obtain simulation data of the cable connector.

4. The method according to claim 3, characterized in that The multi-physics field coupling simulation model is used to solve the coupling problem of the electromagnetic field and the temperature field on the finite element simulation platform to obtain simulation data of the cable joint, including: The control variable method is used to analyze the distribution of the electric field and temperature field of the cable joint under different current carrying conditions, convection heat transfer coefficients and ambient temperatures. According to the affected distribution, the cable joint simulation data under different operating and environmental conditions are obtained.

5. The method according to claim 1, characterized in that Using the simulation data, the multi-physics field coupling simulation model is trained and tested by applying the U-net deep learning algorithm to obtain a simulation agent model, including: The simulation data includes temperature distribution simulation data and electromagnetic field distribution simulation data; Determine an experimental group training data set and a test group data set according to the temperature simulation data and the electromagnetic field distribution simulation data, and divide the experimental group data into a training set and a validation set according to a preset ratio; Obtaining basic parameters corresponding to the temperature distribution simulation data and the electromagnetic field distribution simulation data; Using the basic parameters as input parameters of a U-net deep learning algorithm, and training the multi-physics field coupling simulation model according to the training set and the test set; Optimizing the deep learning model using an adaptive moment estimation optimization algorithm; The test group data set is used to verify the simulation agent model based on the U-net deep learning algorithm.

6. The method according to claim 1, characterized in that Extracting parameter features of the simulation proxy model simulation data and ultrasound information includes: The simulation data and ultrasonic information are processed and feature extracted to obtain key operating parameter characteristics of the cable under real-time operating conditions, wherein the parameter characteristics include temperature gradient, ultrasonic time-frequency signal and electric field strength.

7. The method according to claim 1, characterized in that According to the comparison result between the parameter characteristics and the preset threshold value, a comprehensive state quantitative evaluation model of the cable joint is constructed, including: Set a threshold for each feature to determine whether the abnormal state has been reached, and deduct the score based on the comparison between the feature value and the threshold; The starting score of the cable joint operation status is set to 100 points, and points are deducted according to the degree to which the characteristic value exceeds the threshold; According to the scoring criteria and deduction mechanism, the health status of the cable joint is quantified into a score, and the score is divided into three categories: normal state, warning state and abnormal state according to the preset threshold, so as to construct a comprehensive state quantitative evaluation model of the cable joint.

8. A cable joint status assessment system based on digital twin technology, characterized in that: include: A simulation model building module is used to build a multi-physics field coupling simulation model of the electromagnetic field and temperature field of the cable joint; A simulation data and ultrasonic data acquisition module is used to acquire simulation data of the cable joint under different working conditions and environmental conditions through finite element simulation, and to acquire ultrasonic signals inside the cable joint by arranging sensors on the surface of the cable entity; A training and testing module, used to train and test the simulation agent model based on the U-net deep learning algorithm using the simulation data to obtain a simulation agent model; An evaluation model building module is used to extract parameter features of simulation data and ultrasound information of the simulation agent model; According to the comparison result of the parameter characteristics and the preset threshold value, a comprehensive state quantitative evaluation model of the cable joint is constructed; The evaluation module is used to obtain health status evaluation data of the cable joint to be evaluated through the comprehensive state quantitative evaluation model of the cable joint.

9. The system according to claim 8, characterized in that Training and testing modules, including: A simulation data determination submodule, used to determine that the simulation data includes temperature distribution simulation data and electromagnetic field distribution simulation data; The training set and validation set division submodule is used to determine the experimental group training data set and the test group data set according to the temperature simulation data and the electromagnetic field distribution simulation data, and divide the experimental group data into the training set and the validation set according to a preset ratio; A basic parameter acquisition submodule is used to acquire basic parameters corresponding to the temperature distribution simulation data and the electromagnetic field distribution simulation data; A training submodule, used for taking the basic parameters as input parameters of the U-net deep learning algorithm, and training the simulation agent model based on the deep learning algorithm according to the training set and the test set; An optimization submodule, used to optimize the deep learning model using an adaptive moment estimation optimization algorithm; The verification submodule is used to verify the simulation agent model based on the U-net deep learning algorithm using the test group data set.

10. The system according to claim 8, characterized in that Evaluation model building blocks, including: The first score deduction setting submodule is used to set the threshold of each feature, which is used to determine whether the standard of abnormal state is reached, and deduct the score according to the comparison result between the feature value and the threshold; A second score deduction setting submodule is used to set the starting score of the operating status of the cable connector to 100 points, and deduct points according to the degree to which the characteristic value exceeds the threshold; The scoring quantification submodule is used to quantify the health status of the cable joint into a score according to the scoring criteria and deduction mechanism. The score is divided into three categories: normal state, warning state and abnormal state according to the preset threshold, so as to construct a comprehensive state quantification evaluation model for the cable joint.

Images