High-speed Connector Detection Method, Device and Equipment Based on Data Analysis

Abstract

The present application discloses a high-speed connector detection method, device and equipment based on data analysis. The method includes: constructing a multi-physical field coupling analysis model according to the geometric parameters of the high-speed connector; determining characteristic parameters affecting the performance of the high-speed connector according to the multi-physical field coupling analysis model, and generating a standard feature library corresponding to the characteristic parameters and performance indicators; obtaining real-time detection data of the high-speed connector; identifying potential defects of the high-speed connector according to the standard feature library and the real-time detection data. By implementing the solution of the present application, comparing the real-time detection data with the standard feature library can timely detect performance anomalies, and the theoretical basis provided by the multi-physical field model makes defect identification more accurate.

Description

Technical Field

[0001] The present application relates to the field of data processing, and in particular, to a high-speed connector detection method, device, and equipment based on data analysis. Background Art

[0002] With the rapid development of electronic information technology, as a key component for signal transmission between electronic devices, the reliability and performance of high-speed connectors directly affect the stable operation of the entire system. In modern communication, computer, aerospace and other fields, higher and higher requirements are put forward for the signal transmission rate, bandwidth and integrity, which makes the design and manufacture of high-speed connectors face huge challenges. Traditional high-speed connector detection methods mainly rely on manual experience judgment and single physical field analysis, and it is difficult to comprehensively evaluate the performance characteristics of connectors in complex working environments. In practical applications, high-speed connectors are often simultaneously affected by the coupling effects of multiple physical fields such as mechanical stress, electromagnetic field, and temperature field. These factors interact and restrict each other, resulting in a very complex formation mechanism of connector performance degradation and potential defects. In addition, the existing detection means lack a systematic and standardized characteristic parameter evaluation system, and cannot accurately identify and predict potential faults of connectors. Summary of the Invention

[0003] The present application provides a high-speed connector detection method, device, and equipment based on data analysis, which are used to solve the problem in the related art that potential faults of connectors cannot be accurately identified and predicted.

[0004] In the first aspect of the present application, a high-speed connector detection method based on data analysis is provided. The high-speed connector detection method based on data analysis includes:

[0005] Construct a multi-physical field coupling analysis model according to the geometric parameters of the high-speed connector;

[0006] Determine the characteristic parameters affecting the performance of the high-speed connector according to the multi-physical field coupling analysis model, and generate a standard characteristic library corresponding to the characteristic parameters and performance indicators;

[0007] Obtain the real-time detection data of the high-speed connector;

[0008] Identify potential defects of the high-speed connector according to the standard characteristic library and the real-time detection data.

[0009] Optionally, in the first implementation manner of the first aspect of the present application, before the step of constructing a multi-physical field coupling analysis model according to the geometric parameters of the high-speed connector, the following is further included:

[0010] Obtain the geometric data of the high-speed connector by performing three-dimensional laser scanning on the high-speed connector;

[0011] Determine a parametric geometric model according to the geometric data;

[0012] Construct a digital twin model according to the geometric parameters corresponding to the parametric geometric model.

[0013] Optionally, in the second implementation manner of the first aspect of the present application, the step of determining the characteristic parameters affecting the performance of the high-speed connector according to the multi-physical-field coupling analysis model and generating a standard characteristic library corresponding to the characteristic parameters and performance indicators includes:

[0014] Obtain mechanical parameters, electrical parameters, and thermal parameters from the digital twin model respectively;

[0015] Construct a mechanical model according to the mechanical parameters;

[0016] Construct an electrical model according to the electrical parameters;

[0017] Construct a thermal model according to the thermal parameters;

[0018] Determine the performance parameters of the high-speed connector by coupling the data of the mechanical model, the electrical model, and the thermal model;

[0019] Identify the characteristic parameters affecting the high-speed connector among the performance parameters through parameter sensitivity analysis;

[0020] Determine the mapping relationship between the characteristic parameters and the performance indicators;

[0021] Generate a standard characteristic library according to the mapping relationship.

[0022] Optionally, in the third implementation manner of the first aspect of the present application, the step of obtaining the real-time detection data of the high-speed connector includes:

[0023] Determine the contact pressure measurement data of the high-speed connector according to the force-sensitive sensor array;

[0024] Determine the hot spot distribution data of the high-speed connector according to the temperature sensor array;

[0025] Determine the deformation amount data at the contact interface of the high-speed connector according to the micro-displacement sensor;

[0026] Preprocess the contact pressure measurement data, the hot spot distribution data, and the deformation amount data to generate physical parameter data.

[0027] Optionally, in the fourth implementation manner of the first aspect of the present application, the method further includes:

[0028] Determine the electric field distribution data of the high-speed connector according to the electric field probe array;

[0029] Determine the magnetic field distribution data of the high-speed connector according to the magnetic field probe array;

[0030] Fuse the electric field distribution data and the magnetic field distribution data through a multi-dimensional data fusion algorithm to form electromagnetic field data.

[0031] Optionally, in the fifth implementation manner of the first aspect of the present application, the real-time detection data includes the physical parameter data and the electromagnetic field data, and the step of identifying potential defects of the high-speed connector according to the standard feature library and the real-time detection data includes:

[0032] Generate a defect training sample for the deep learning model according to the standard feature library and the real-time detection data;

[0033] Train the deep learning model according to the defect training sample;

[0034] Extract features from the physical parameter data and the electromagnetic field data respectively according to the multi-stream convolutional neural network of the trained deep learning model to determine the spatial features of the physical parameter data and the electromagnetic field data;

[0035] Determine the temporal features of the physical parameter data and the electromagnetic field data according to the long short-term memory network of the trained deep learning model;

[0036] Generate a multi-dimensional feature vector by splicing the spatial features and the temporal features;

[0037] Identify potential defects existing in the high-speed connector according to the multi-dimensional feature vector.

[0038] Optionally, in the sixth implementation manner of the first aspect of the present application, the method further includes:

[0039] Obtain the historical operation data of the high-speed connector;

[0040] Establish a performance degradation model of the high-speed connector according to the historical operation data and the real-time detection data;

[0041] Predict the remaining service life of the high-speed connector according to the performance degradation model.

[0042] The second aspect of the present application provides a high-speed connector detection device based on data analysis, and the high-speed connector detection device based on data analysis includes:

[0043] A construction module for constructing a multi-physical field coupling analysis model according to the geometric parameters of the high-speed connector;

[0044] A generation module, configured to determine characteristic parameters affecting the performance of the high-speed connector according to the multi-physical-field coupling analysis model, and generate a standard feature library corresponding to the characteristic parameters and performance indicators;

[0045] An acquisition module, configured to acquire real-time detection data of the high-speed connector;

[0046] An identification module, configured to identify potential defects of the high-speed connector according to the standard feature library and the real-time detection data.

[0047] A third aspect of the embodiments of the present application provides an electronic device, including a memory and a processor. The processor is configured to execute a computer program stored on the memory. When the processor executes the computer program, each step in the method for detecting a high-speed connector based on data analysis provided in the first aspect of the embodiments of the present application is implemented.

[0048] A fourth aspect of the embodiments of the present application provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, each step in the method for detecting a high-speed connector based on data analysis provided in the first aspect of the embodiments of the present application is implemented.

[0049] In summary, according to the method, device and equipment for detecting a high-speed connector based on data analysis provided by the solution of the present application, a multi-physical-field coupling analysis model is constructed according to the geometric parameters of the high-speed connector; characteristic parameters affecting the performance of the high-speed connector are determined according to the multi-physical-field coupling analysis model, and a standard feature library corresponding to the characteristic parameters and performance indicators is generated; real-time detection data of the high-speed connector is acquired; potential defects of the high-speed connector are identified according to the standard feature library and the real-time detection data. By implementing the solution of the present application, the real-time detection data is compared with the standard feature library, and performance anomalies can be found in time. The theoretical basis provided by the multi-physical-field model makes defect identification more accurate. Description of the Drawings

[0050] Figure 1 It is a schematic flowchart of the method for detecting a high-speed connector based on data analysis provided by the embodiments of the present application;

[0051] Figure 2 It is a schematic diagram of program modules of the device for detecting a high-speed connector based on data analysis provided by the embodiments of the present application;

[0052] Figure 3 It is a schematic structural diagram of the electronic device provided by the embodiments of the present application. Detailed Embodiments

[0053] In order to make the invention object, features, and advantages of the present application more obvious and understandable, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative efforts belong to the scope of protection of the present application.

[0054] In order to solve the problem in the related art that the potential faults of the connector cannot be accurately identified and predicted, the embodiments of the present application provide a high-speed connector detection method based on data analysis, as Figure 1 is a schematic flowchart of the high-speed connector detection method based on data analysis provided in this embodiment. The high-speed connector detection method based on data analysis includes the following steps:

[0055] Step 110: Construct a multi-physics field coupling analysis model according to the geometric parameters of the high-speed connector.

[0056] Specifically, in this embodiment, a high-precision 3D scanner is used to scan the high-speed connector to obtain its accurate geometric parameters, including the diameter and length of the male end pins, the thickness and elastic modulus of the female end reeds, etc. The scanned data is imported into computer-aided design (CAD) software to establish a geometric model of the connector. Among them, the multi-physics field coupling analysis model includes a mechanical model, an electrical model, a thermal model, and an electromagnetic field model.

[0057] In an optional implementation manner of this embodiment, before the step of constructing a multi-physics field coupling analysis model according to the geometric parameters of the high-speed connector, it further includes: obtaining the geometric data of the high-speed connector through three-dimensional laser scanning of the high-speed connector; determining a parameterized geometric model according to the geometric data; constructing a digital twin model according to the geometric parameters corresponding to the parameterized geometric model.

[0058] Specifically, in this embodiment, during the digital modeling process of the high-speed connector, it is first necessary to use a high-precision three-dimensional laser scanner to scan the connector in all directions. The laser scanner used in this process usually adopts blue light structured light technology, and the scanning accuracy can reach the order of 0.01 mm. During the scanning process, the connector is fixed on a precision rotary workbench, and the integrity and accuracy of the collected data are ensured by scanning from multiple angles and directions. During the scanning process, the structured light emitted by the laser scanner hits the surface of the connector to form a stripe pattern. At the same time, these patterns are collected by a high-resolution camera, and the three-dimensional surface contour of the object is reconstructed through the principle of triangulation. Subsequently, the obtained original scanning data is processed by professional point cloud processing software, including operations such as noise removal, data alignment, and mesh reconstruction. At this stage, special attention needs to be paid to the data integrity of the key feature parts of the connector, such as the pin and the reed contact area. Through reverse engineering software, the processed point cloud data is converted into an editable three-dimensional solid model. During this process, the surface needs to be accurately fitted to ensure the geometric accuracy of the model. Then, a parametric geometric model is established based on the obtained geometric data. This process requires identifying the key dimensional parameters of the connector, such as the diameter and length of the pin, and the thickness and bending angle of the reed. Through parametric modeling software (such as CATIA or NX), these key dimensions are set as variable parameters to establish a parameter-driven geometric model. Such a parametric model has strong adjustability and can facilitate design optimization and variant design. Finally, a digital twin model is constructed based on the parametric geometric model. The digital twin model not only contains geometric information but also needs to integrate multi-dimensional information such as physical properties, material attributes, and assembly relationships. By associating the sensor data on the physical connector with the digital model, real-time mapping and interaction between the physical world and the digital world are realized. The digital twin model can reflect the working state of the connector in real time, predict the performance change trend, and provide support for product optimization and maintenance decisions.

[0059] Step 120: Determine the characteristic parameters affecting the performance of the high-speed connector according to the multi-physical field coupling analysis model, and generate a standard characteristic library corresponding to the characteristic parameters and performance indicators.

[0060] Specifically, in this embodiment, according to the coupling analysis results, the stress distribution and stress concentration points in the key area are determined. Characteristic parameters such as the maximum stress value, stress distribution law, and stress concentration coefficient are recorded. The current density distribution characteristics are determined, and characteristic parameters such as the maximum current density, current density change in the key area, and current density gradient are recorded. According to the thermal analysis results, the temperature distribution characteristics are determined, and characteristic parameters such as the maximum temperature rise, temperature change in the key area, and thermal stress are recorded. According to the actual usage requirements of the connector, performance indicators such as contact resistance, mechanical strength, and high-temperature resistance are determined. Through experiments and simulations, the corresponding relationship between the characteristic parameters and performance indicators is established. For example, the maximum stress and mechanical strength, the maximum current density and contact resistance, the maximum temperature rise and high-temperature resistance, etc. The multi-physical field coupling analysis results and experimental data are collected, and the corresponding relationship between the characteristic parameters and performance indicators is sorted out. A standard characteristic library is established, and the corresponding relationship between the characteristic parameters and performance indicators is entered into the database to form a complete standard characteristic library. This characteristic library will be used as a benchmark for subsequent real-time detection and defect identification.

[0061] In an alternative implementation manner of this embodiment, the steps of determining the characteristic parameters affecting the performance of the high-speed connector according to the multi-physical field coupling analysis model and generating a standard characteristic library corresponding to the characteristic parameters and performance indicators include:

[0062] Obtain mechanical parameters, electrical parameters, and thermal parameters from the digital twin model respectively; construct a mechanical model according to the mechanical parameters; construct an electrical model according to the electrical parameters; construct a thermal model according to the thermal parameters; determine the performance parameters of the high-speed connector by coupling the data of the mechanical model, electrical model, and thermal model; identify the characteristic parameters affecting the high-speed connector among the performance parameters through parameter sensitivity analysis; determine the mapping relationship between the characteristic parameters and performance indicators; generate a standard characteristic library according to the mapping relationship.

[0063] Specifically, in this embodiment, during the multi-physics field analysis of the high-speed connector, relevant physical parameters need to be extracted from the digital twin model first. By parsing the digital twin model, mechanical parameters (such as elastic modulus, Poisson's ratio, yield strength, etc.), electrical parameters (such as conductivity, dielectric constant, etc.), and thermal parameters (such as thermal conductivity, specific heat capacity, etc.) can be obtained. During the parameter extraction process, attention should be paid to the interdependent relationships between parameters, such as the influence of temperature on the elastic modulus of materials, the influence of stress on conductivity, etc. A mechanical model is constructed based on the obtained mechanical parameters, and this process is mainly achieved through the finite element method. In the mechanical model, the force conditions of the connector under actual working conditions need to be considered, including insertion and extraction force, contact pressure, vibration load, etc. At the same time, an electrical model is constructed based on the electrical parameters, focusing on characteristics such as current distribution, contact resistance, and parasitic parameters. In addition, a thermal model is constructed using the thermal parameters to analyze the temperature field distribution and thermal stress effect. After establishing each physical field model, a coupled analysis is performed through a multi-physics field coupling solver. During the coupling process, there are complex interactions between physical fields. For example, the Joule heat generated by the passing current will affect the temperature field distribution, and the change in the temperature field will cause thermal stress, thereby changing the contact pressure and resistance. Through iterative solution, a stable coupled solution is finally obtained to determine the comprehensive performance parameters of the connector. Through parameter sensitivity analysis, the characteristic parameters that have a significant impact on the connector performance can be identified. This process is achieved by changing the input parameters and observing the changes in the output response. By calculating the sensitivity coefficient, the importance degree of different parameters can be quantified. After determining the key characteristic parameters, the mapping relationship between these parameters and the performance indicators is established, and finally a standard characteristic library is generated.

[0064] Step 130: Obtain the real-time detection data of the high-speed connector.

[0065] Specifically, in this embodiment, according to the characteristic parameters, appropriate sensors are selected, such as force-sensitive sensors, temperature sensors, current density sensors, etc. The sensors are arranged in the key areas of the connector, such as stress concentration points, current density change areas, temperature rise areas, etc., to ensure that important parameters can be monitored in real time. The obtained real-time detection data is cleaned to remove noise and outliers to ensure the accuracy of the data.

[0066] In an optional implementation manner of this embodiment, the step of obtaining the real-time detection data of the high-speed connector includes: determining the contact pressure measurement data of the high-speed connector according to the force-sensitive sensor array; determining the hot spot distribution data of the high-speed connector according to the temperature sensor array; determining the deformation data at the contact interface of the high-speed connector according to the micro-displacement sensor; and preprocessing the contact pressure measurement data, hot spot distribution data, and deformation data to generate physical parameter data.

[0067] Specifically, in this embodiment, the force-sensitive sensor array is used to measure the pressure distribution at the contact interface of the high-speed connector. The force-sensitive sensor can convert physical pressure into an electrical signal. Through the sensor array, pressure data at multiple points can be obtained to form a complete pressure distribution map. For example, when a connector is inserted into the corresponding slot, the force-sensitive sensor array can detect the pressure value at each contact point to determine the uniformity and firmness of the contact. These data can be transmitted to the central processing unit through the data acquisition system for analysis. While acquiring the contact pressure data, the temperature sensor array is used to monitor the heat distribution generated by the high-speed connector during operation. When the high-speed connector operates under high-frequency and high-current conditions, it is prone to generate heat, resulting in a temperature increase. The temperature sensor array can detect the temperature of each part of the connector in real time and generate a hot spot distribution map. For example, if the temperature of a certain contact point is significantly higher than other areas, it may indicate that there is poor contact or excessive resistance at this contact point. These temperature data are also transmitted to the central processing unit through the data acquisition system for analysis. The micro-displacement sensor is used to measure the deformation data at the contact interface of the high-speed connector. The micro-displacement sensor can detect minute displacement changes, reflecting the deformation situation of the connector during the contact process. For example, during the insertion of the connector, some contact points may deform, affecting the contact quality. The micro-displacement sensor array can accurately measure these deformation amounts and provide detailed deformation data.

[0068] After obtaining the contact pressure measurement data, hot spot distribution data, and deformation data, these data need to be preprocessed. The data preprocessing includes steps such as noise filtering, data calibration, and data fusion. First, noise filtering is to remove random noise and interference signals in the sensor data to improve the accuracy of the data. For example, digital filtering techniques such as mean filtering and Kalman filtering can be used to smooth the sensor data. Second, data calibration is to ensure the consistency and comparability of the data of each sensor. For example, through the method of experimental calibration, the response curve and sensitivity of the sensor can be determined, and the measurement data can be corrected. Finally, data fusion is to integrate the data of different sensors to generate more comprehensive and reliable physical parameter data. For example, multi-sensor data fusion techniques such as weighted average method and Bayesian estimation can be used to combine the contact pressure data, hot spot distribution data, and deformation data to generate a complete physical parameter model.

[0069] It can be understood that in practical applications, when a high-speed connector transmits high-frequency signals, problems such as uneven contact pressure, local overheating, and deformation may occur simultaneously. If relying solely on a single type of sensor, it is difficult to comprehensively and accurately diagnose these problems. However, by jointly using a force-sensitive sensor array, a temperature sensor array, and a micro-displacement sensor, the working state of the connector can be comprehensively monitored, and detailed physical parameter data can be obtained. For example, through the force-sensitive sensor array, it can be found that the pressure at certain contact points is significantly lower than that in other areas, indicating the possibility of poor contact; through the temperature sensor array, it can be found that the temperature in certain areas rises abnormally, indicating the problem of local overheating; through the micro-displacement sensor, the deformation of the connector during the insertion process can be monitored, indicating potential mechanical stress problems. Through data preprocessing and fusion, a comprehensive physical parameter model can be generated, providing an important basis for fault diagnosis and performance optimization.

[0070] In an optional implementation manner of this embodiment, the electric field distribution data of the high-speed connector is determined according to the electric field probe array; the magnetic field distribution data of the high-speed connector is determined according to the magnetic field probe array; the electric field distribution data and the magnetic field distribution data are fused through a multi-dimensional data fusion algorithm to form electromagnetic field data.

[0071] Specifically, in this embodiment, the electric field probe array is used to measure the electric field distribution around the high-speed connector. The electric field probe can convert the electric field intensity into an electrical signal through the capacitance effect, thereby realizing the accurate measurement of the electric field distribution. Similar to the electric field probe array, the magnetic field probe array is used to measure the magnetic field distribution around the high-speed connector. The magnetic field probe can convert the magnetic field intensity into an electrical signal through the inductance effect, thereby realizing the accurate measurement of the magnetic field distribution. After the data preprocessing is completed, various data fusion algorithms can be used for the fusion of electric field and magnetic field data. For example, the weighted average method can be used to perform weighted summation according to the weights of the electric field and magnetic field data to obtain the fused electromagnetic field data. The Bayesian estimation method can also be used to perform data fusion and update by establishing a probability model of the electric field and magnetic field data. In addition, deep learning algorithms can be used to automatically fuse and predict the electric field and magnetic field data by training a neural network model. Through the data fusion algorithm, the electric field and magnetic field data can be combined to obtain a more complete and accurate electromagnetic field distribution data.

[0072] Step 140: Identify potential defects of the high-speed connector according to the standard feature library and the real-time detection data.

[0073] Specifically, in this embodiment, key feature parameters are extracted from the real-time detection data, such as stress values, temperature values, current density values, etc. The extracted feature parameters are compared with the data in the standard feature library to determine whether there is a deviation. It should be noted that during the comparison process, machine learning algorithms (such as support vector machines, random forests, neural networks, etc.) are used for defect identification. When training the model, the data in the standard feature library is used as the training set, and the real-time detection data is used as the test set. According to the detection data obtained by the machine learning algorithm, the specific location and type of the defect are determined.

[0074] In an alternative implementation manner of this embodiment, the steps of identifying potential defects of a high-speed connector according to the standard feature library and real-time detection data include: generating defect training samples for a deep learning model according to the standard feature library and real-time detection data; training the deep learning model according to the defect training samples; extracting features of physical parameter data and electromagnetic field data respectively by using the multi-stream convolutional neural network of the trained deep learning model to determine the spatial features of the physical parameter data and electromagnetic field data; determining the temporal features of the physical parameter data and electromagnetic field data by using the long short-term memory network of the trained deep learning model; generating a multi-dimensional feature vector by splicing the spatial features and temporal features; and identifying potential defects existing in the high-speed connector according to the multi-dimensional feature vector.

[0075] Specifically, in this embodiment, representative historical data and their corresponding characteristic parameters are extracted from the standard feature library. These data include physical parameters and electromagnetic field data under various known normal and fault states. The physical parameters and electromagnetic field data of the current state collected by the sensor array and electromagnetic probes for real-time detection data also need to be synchronously obtained and compared with the data in the standard feature library to determine whether there are abnormalities or potential defects. When generating defect training samples, first, the standard feature library is matched with the real-time detection data, and data samples with typical fault characteristics are screened out. Specifically, by comparing the characteristic parameters of the real-time detection data with those of the normal state in the standard feature library, abnormal data points can be identified. For example, if the contact pressure of a certain connector in the real-time detection data is significantly lower than the normal range, and abnormal fluctuations also occur in the electric field strength and magnetic field strength, it can be preliminarily determined that there may be a potential defect of poor contact in this connector. These abnormal data points, together with their corresponding annotation information, constitute the defect training samples. The deep learning model is trained according to the generated defect training samples. First, a multi-stream convolutional neural network (CNN) structure is designed to process physical parameter data and electromagnetic field data respectively. The convolutional layer of the CNN performs convolutional operations on the input data through a sliding window method to extract the spatial features of the data. For physical parameter data, such as contact pressure, temperature, and deformation amount, the CNN can extract the spatial distribution features of the data, such as the pressure concentration phenomenon in certain areas or the temperature hot spot areas. For electromagnetic field data, the CNN can identify the distribution patterns and variation rules of the electric and magnetic fields, such as abnormal electric field strength or magnetic field leakage phenomena in certain areas. During the model training process, the weights of the convolutional layer are continuously adjusted through the backpropagation algorithm to minimize the prediction error. The features extracted by the convolutional layer are downsampled through the pooling layer to reduce the data dimension and retain important features. Finally, the features are mapped to the output space through the fully connected layer to obtain the classification result. Through a large number of defect training samples, the CNN can gradually learn the spatial features under various fault states and achieve accurate classification of physical parameter data and electromagnetic field data. In addition to spatial features, physical parameter data and electromagnetic field data also have temporal features. In the trained deep learning model, a long short-term memory network (LSTM) is used to capture the temporal features of the data. The LSTM can effectively process time series data and capture the long-term dependence relationship of the data through its special gating mechanism. For example, for the contact pressure data of a certain connector, the LSTM can identify the change trend of the pressure over time and determine whether it is a stable state or a gradually deteriorating trend. For electromagnetic field data, the LSTM can identify the periodic changes or sudden changes in the electric and magnetic field strengths, thereby determining potential faults such as electromagnetic interference or short circuits. By splicing the spatial features and temporal features, a multi-dimensional feature vector can be generated.Specifically, the spatial features extracted by the CNN and the temporal features extracted by the LSTM are concatenated to form a high-dimensional feature vector, which contains comprehensive information of physical parameter data and electromagnetic field data. During the feature vector concatenation process, the weighted average method or the attention mechanism can be adopted to assign different weights to different features to highlight the importance of key features. Finally, potential defects existing in the high-speed connector are identified based on the multi-dimensional feature vector. By analyzing the concatenated feature vector, the fault type and severity of the connector can be accurately identified. For example, if some spatial features and temporal features in the feature vector deviate significantly from the normal range, it can be determined that the connector may have faults such as poor contact, overheating, or electromagnetic interference.

[0076] It should be noted that in this embodiment, by combining the multi-stream convolutional neural network (CNN) and the long short-term memory network (LSTM), spatial features and temporal features are extracted from physical parameter data and electromagnetic field data, and a multi-dimensional feature vector is generated:

[0077] ,

[0078] where V is the multi-dimensional feature vector, is the activation function, represents the weighted sum of the spatial feature vectors extracted by the CNN, M is the number of physical parameter data, is the weight coefficient of the i-th physical parameter data, the i-th physical parameter data [[ID=2nd]]the spatial feature vector extracted by the CNN, represents the weighted sum of the temporal feature vectors extracted by the LSTM, N is the number of electromagnetic field data, is the weight coefficient of the j-th electromagnetic field data, is the j-th electromagnetic field data combining the previous moment's hidden state the temporal feature vector extracted by the LSTM, is the feature concatenation operation, which combines the spatial feature vector and the temporal feature vector into a multi-dimensional feature vector.

[0079] It should be noted that the formula of the trained classification model can be expressed as:

[0080] ,

[0081] where, is the final classification result, is the activation function used to convert the input vector into a probability distribution, is the weight matrix of the output layer, is the bias vector of the output layer, Is an element - level multiplication operation used to introduce the attention mechanism, Is the attention vector, representing the importance weights of each feature. This formula integrates the spatial and temporal features of physical parameter data and electromagnetic field data. After being weighted by the attention mechanism, a multi - dimensional feature vector is generated. Finally, through The layer outputs the classification result, which is used to identify potential defects existing in the high - speed connector.

[0082] In an optional implementation manner of this embodiment, historical operation data of the high - speed connector is obtained; a performance degradation model of the high - speed connector is established according to the historical operation data and real - time detection data; the remaining service life of the high - speed connector is predicted according to the performance degradation model.

[0083] Specifically, in this embodiment, historical operation data is obtained through a multi - type sensor array installed on the high - speed connector. These sensors include force - sensitive sensors, temperature sensors, electric field probes, and magnetic field probes, etc., so as to obtain the operation parameters of the connector under different working conditions and environmental conditions, including multi - dimensional historical data such as contact pressure, temperature distribution, electric field strength, and magnetic field strength. At the same time, combined with operation and maintenance data such as the maintenance record, fault record, and replacement record of the connector, a complete historical database is established. On the basis of obtaining the historical operation data, the real - time detection data is associated and analyzed with the historical operation data, and a performance degradation model of the high - speed connector is established through a deep learning algorithm. Based on the established performance degradation model, the remaining service life of the connector can be predicted by inputting the real - time detection data. The prediction process first standardizes the real - time data to make its scale consistent with the historical data, and then inputs the processed data into the trained performance degradation model. The model will compare the current performance parameters with the historical degradation trend, and combine with the Weibull distribution model to calculate the reliability function of the high - speed connector and predict the remaining service life.

[0084] Optionally, the calculation formula of the performance degradation model in this embodiment is expressed as:

[0085] ,

[0086] Wherein, Is the performance degradation index vector at time t, Is a non - linear mapping function, K is the number of feature extractors. A feature extractor is an algorithm or model used to extract useful feature information from raw data, mainly used to extract key features that can reflect the performance state of the connector from the complex data collected by sensors, Is the weight coefficient of the k - th feature extractor, Is the k - th deep feature extractor, based on an improved LSTM network, Is the historical data matrix at time t, is the real-time data matrix at time t, is the tensor product operator for feature fusion, λ is the attenuation coefficient, is the time-varying stress function, is the adaptive weight function, is the environmental factor influence vector. It can be understood that in the data preparation stage, historical data matrices and real-time data matrices are prepared, including but not limited to pressure, temperature, electromagnetic field strength, etc. Feature extraction is to extract useful features from historical data and real-time data through a deep learning model (such as an improved LSTM network), and weights are assigned to each feature extractor according to the importance of the features. The time-varying stress function in the stress function calculation stage is used to describe the stress changes suffered by the device during operation, and the stress changes will affect the degradation rate of the device. In the environmental impact analysis stage, the adaptive weight function and the environmental factor influence vector are used to comprehensively consider the impact of environmental factors (such as temperature, humidity, dust, etc.) on the device degradation. Finally, all the extracted features and weights are non-linearly mapped, and combined with the stress attenuation factor and environmental impact, the final performance degradation index is calculated.

[0087] According to a high-speed connector detection method based on data analysis provided by the solution of the present application, a multi-physical field coupling analysis model is constructed according to the geometric parameters of the high-speed connector; the characteristic parameters affecting the performance of the high-speed connector are determined according to the multi-physical field coupling analysis model, and a standard feature library corresponding to the characteristic parameters and performance indicators is generated; the real-time detection data of the high-speed connector is obtained; the potential defects of the high-speed connector are identified according to the standard feature library and the real-time detection data. Through the implementation of the solution of the present application, by comparing the real-time detection data with the standard feature library, performance anomalies can be discovered in a timely manner, and the theoretical basis provided by the multi-physical field model makes defect identification more accurate.

[0088] Figure 2 This is a high-speed connector detection device provided by an embodiment of the present application. This high-speed connector detection device based on data analysis can be used to implement the high-speed connector detection method based on data analysis in the foregoing embodiments. As Figure 2 shown, this high-speed connector detection device based on data analysis mainly includes:

[0089] A construction module 10 for constructing a multi-physical field coupling analysis model according to the geometric parameters of the high-speed connector;

[0090] A generation module 20 for determining the characteristic parameters affecting the performance of the high-speed connector according to the multi-physical field coupling analysis model and generating a standard feature library corresponding to the characteristic parameters and performance indicators;

[0091] An acquisition module 30 for acquiring the real-time detection data of the high-speed connector;

[0092] An identification module 40, configured to identify potential defects of the high-speed connector according to a standard feature library and real-time detection data.

[0093] In an optional implementation manner of this embodiment, the high-speed connector detection device further includes: a determination module. The acquisition module is configured to: obtain geometric data of the high-speed connector by performing three-dimensional laser scanning on the high-speed connector. The determination module is configured to: determine a parametric geometric model according to the geometric data. The construction module is configured to: construct a digital twin model according to geometric parameters corresponding to the parametric geometric model.

[0094] In an optional implementation manner of this embodiment, the generation module is specifically configured to: respectively obtain mechanical parameters, electrical parameters, and thermal parameters from the digital twin model; construct a mechanical model according to the mechanical parameters; construct an electrical model according to the electrical parameters; construct a thermal model according to the thermal parameters; determine performance parameters of the high-speed connector by coupling data of the mechanical model, the electrical model, and the thermal model; identify characteristic parameters affecting the high-speed connector in the performance parameters through parameter sensitivity analysis; determine the mapping relationship between the characteristic parameters and the performance indicators; and generate a standard feature library according to the mapping relationship.

[0095] In an optional implementation manner of this embodiment, the acquisition module is specifically configured to: determine contact pressure measurement data of the high-speed connector according to a force-sensitive sensor array; determine hot spot distribution data of the high-speed connector according to a temperature sensor array; determine deformation data at the contact interface of the high-speed connector according to a micro-displacement sensor; and preprocess the contact pressure measurement data, the hot spot distribution data, and the deformation data to generate physical parameter data.

[0096] In an optional implementation manner of this embodiment, the determination module is further configured to: determine electric field distribution data of the high-speed connector according to an electric field probe array; determine magnetic field distribution data of the high-speed connector according to a magnetic field probe array; and fuse the electric field distribution data and the magnetic field distribution data through a multi-dimensional data fusion algorithm to form electromagnetic field data.

[0097] In an optional implementation manner of this embodiment, the identification module is specifically configured to: generate defect training samples for a deep learning model according to the standard feature library and real-time detection data; train the deep learning model according to the defect training samples; respectively extract features of the physical parameter data and the electromagnetic field data through a multi-stream convolutional neural network of the trained deep learning model to determine spatial features of the physical parameter data and the electromagnetic field data; determine temporal features of the physical parameter data and the electromagnetic field data according to a long short-term memory network of the trained deep learning model; splice the spatial features and the temporal features to generate a multi-dimensional feature vector; and identify potential defects existing in the high-speed connector according to the multi-dimensional feature vector.

[0098] In an alternative embodiment of this embodiment, the high-speed connector detection device further includes: a prediction module. The acquisition module is further configured to: acquire the historical operation data of the high-speed connector. The construction module is further configured to: establish a performance degradation model of the high-speed connector according to the historical operation data and the real-time detection data. The prediction module is configured to: predict the remaining service life of the high-speed connector according to the performance degradation model.

[0099] According to a high-speed connector detection device based on data analysis provided by the solution of the present application, a multi-physical field coupling analysis model is constructed according to the geometric parameters of the high-speed connector; characteristic parameters affecting the performance of the high-speed connector are determined according to the multi-physical field coupling analysis model, and a standard feature library corresponding to the characteristic parameters and performance indicators is generated; real-time detection data of the high-speed connector is acquired; potential defects of the high-speed connector are identified according to the standard feature library and the real-time detection data. By implementing the solution of the present application and comparing the real-time detection data with the standard feature library, performance anomalies can be discovered in a timely manner, and the theoretical basis provided by the multi-physical field model makes defect identification more accurate.

[0100] According to what is provided by the solution of the present application Figure 3 An electronic device provided for an embodiment of the present application. This electronic device can be used to implement the data analysis-based high-speed connector detection method in the foregoing embodiments, and mainly includes:

[0101] A memory 301, a processor 302, and a computer program 303 stored on the memory 301 and executable on the processor 302. The memory 301 and the processor 302 are communicatively connected. When the processor 302 executes the computer program 303, the data analysis-based high-speed connector detection method in the foregoing embodiments is implemented. Among them, the number of processors can be one or more.

[0102] The memory 301 can be a high-speed random access memory (RAM, Random Access Memory) or a non-volatile memory, such as a disk memory. The memory 301 is used to store executable program code, and the processor 302 is coupled to the memory 301.

[0103] Furthermore, an embodiment of the present application also provides a computer-readable storage medium, which can be disposed in the electronic device in the foregoing embodiments, and the computer-readable storage medium can be the memory in the foregoing Figure 3 illustrated embodiments.

[0104] A computer program is stored on the computer-readable storage medium, and when the program is executed by a processor, it implements the high-speed connector detection method based on data analysis in the foregoing embodiments. Further, the computer-readable storage medium may also be various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a RAM, a magnetic disk, or an optical disc.

[0105] Those skilled in the art can clearly understand that for the convenience and simplicity of description, the specific working processes of the systems, devices, and units described above can refer to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0106] If the 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-readable storage medium. Based on this understanding, the technical solution of the present application, in essence, or the part that contributes to the prior art, or all or part of this technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present application. The foregoing storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disc.

[0107] As described above, the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements on some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the various embodiments of the present application.

Claims

1. A high-speed connector detection method based on data analysis, characterized in that Including: Construct a multi - physical - field coupling analysis model according to the geometric parameters of the high - speed connector; Determine the characteristic parameters affecting the performance of the high - speed connector according to the multi - physical - field coupling analysis model, and generate a standard feature library corresponding to the characteristic parameters and performance indicators; Obtain the real - time detection data of the high - speed connector; Identify potential defects of the high - speed connector according to the standard feature library and the real - time detection data; The method further includes: Obtain the historical operation data of the high - speed connector; Establish a performance degradation model of the high - speed connector according to the historical operation data and the real - time detection data; where the calculation formula of the performance degradation model is expressed as: ， Among them, is the performance degradation index vector at time t, is the non - linear mapping function, K is the number of feature extractors. A feature extractor is an algorithm or model used to extract useful feature information from raw data, mainly used to extract key features that can reflect the performance state of the connector from the complex data collected by sensors, is the weight coefficient of the k - th feature extractor, is the k - th deep feature extractor, based on the improved LSTM network, is the historical data matrix at time t, is the real - time data matrix at time t, is the tensor product operator for feature fusion, λ is the attenuation coefficient, is the time - varying stress function, is the adaptive weight function, is the environmental factor influence vector; Extract the key features reflecting the performance of the high - speed connector from the historical operation data and the real - time detection data according to the performance degradation model, perform non - linear mapping on all the extracted features and weights, and combine the stress attenuation factor and environmental impact to calculate the final performance degradation index, so as to predict the remaining service life of the high - speed connector.

2. The high-speed connector detection method based on data analysis according to claim 1, wherein Before the step of constructing a multi - physical - field coupling analysis model according to the geometric parameters of the high - speed connector, it further includes: Obtain the geometric data of the high - speed connector by performing three - dimensional laser scanning on the high - speed connector; Determine a parametric geometric model according to the geometric data; Construct a digital twin model according to the geometric parameters corresponding to the parametric geometric model.

3. The high-speed connector detection method based on data analysis according to claim 2, wherein The step of determining the characteristic parameters affecting the performance of the high - speed connector according to the multi - physical - field coupling analysis model, and generating a standard feature library corresponding to the characteristic parameters and performance indicators includes: Respectively obtain mechanical parameters, electrical parameters, and thermal parameters from the digital twin model; Construct a mechanical model according to the mechanical parameters; Construct an electrical model according to the electrical parameters; Construct a thermal model according to the thermal parameters; Determine the performance parameters of the high - speed connector by coupling the data of the mechanical model, the electrical model, and the thermal model; Identify the characteristic parameters affecting the high - speed connector among the performance parameters through parameter sensitivity analysis; Determine the mapping relationship between the characteristic parameters and the performance indicators; Generate a standard feature library according to the mapping relationship.

4. The high-speed connector detection method based on data analysis according to claim 1, wherein The step of obtaining the real - time detection data of the high - speed connector includes: Determine the contact pressure measurement data of the high - speed connector according to the force - sensitive sensor array; Determine the hot - spot distribution data of the high - speed connector according to the temperature sensor array; Determine the deformation data at the contact interface of the high - speed connector according to the micro - displacement sensor; Pre - process the contact pressure measurement data, the hot - spot distribution data, and the deformation data to generate physical parameter data.

5. The method for detecting a high-speed connector based on data analysis according to claim 4, wherein The method further includes: Determine the electric - field distribution data of the high - speed connector according to the electric - field probe array; Determine the magnetic - field distribution data of the high - speed connector according to the magnetic - field probe array; Fuse the electric - field distribution data and the magnetic - field distribution data through a multi - dimensional data fusion algorithm to form electromagnetic - field data.

6. The method for detecting a high-speed connector based on data analysis according to claim 5, characterized in that, The real-time detection data includes the physical parameter data and the electromagnetic field data. The step of identifying potential defects of the high-speed connector according to the standard feature library and the real-time detection data includes: Generating a defect training sample for a deep learning model according to the standard feature library and the real-time detection data; Training the deep learning model according to the defect training sample; Extracting features from the physical parameter data and the electromagnetic field data respectively by using the multi-stream convolutional neural network of the trained deep learning model to determine the spatial features of the physical parameter data and the electromagnetic field data; Determining the temporal features of the physical parameter data and the electromagnetic field data by using the long short-term memory network of the trained deep learning model; Generating a multi-dimensional feature vector by splicing the spatial features and the temporal features; Identifying potential defects existing in the high-speed connector according to the multi-dimensional feature vector.

7. A high-speed connector detection device based on data analysis, characterized in that, The high-speed connector detection device based on data analysis includes: A construction module for constructing a multi-physical field coupling analysis model according to the geometric parameters of the high-speed connector; A generation module for determining characteristic parameters affecting the performance of the high-speed connector according to the multi-physical field coupling analysis model and generating a standard feature library corresponding to the characteristic parameters and performance indicators; An acquisition module for acquiring real-time detection data of the high-speed connector; An identification module for identifying potential defects of the high-speed connector according to the standard feature library and the real-time detection data; The prediction module is used for: acquiring historical operation data of the high-speed connector; establishing a performance degradation model of the high-speed connector according to the historical operation data and the real-time detection data; wherein, the calculation formula of the performance degradation model is expressed as: ， Among them, is the performance degradation index vector at time t, is the non - linear mapping function, K is the number of feature extractors. A feature extractor is an algorithm or model used to extract useful feature information from raw data, mainly used to extract key features that can reflect the performance state of the connector from the complex data collected by sensors, is the weight coefficient of the k - th feature extractor, is the k - th deep feature extractor, based on the improved LSTM network, is the historical data matrix at time t, is the real - time data matrix at time t, is the tensor product operator, used for feature fusion, λ is the attenuation coefficient, is the time - varying stress function, is the adaptive weight function, is the environmental factor influence vector; according to the performance degradation model, extract the key features that can reflect the performance of the high - speed connector from the historical operation data and the real - time detection data, perform non - linear mapping on all the extracted features and weights, and combine the stress attenuation factor and environmental influence to calculate the final performance degradation index, so as to predict the remaining service life of the high - speed connector.

8. An electronic device, characterized in that, Including a memory and a processor, wherein: The processor is used for executing a computer program stored on the memory; When the processor executes the computer program, the steps in the high-speed connector detection method based on data analysis according to any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, the steps in the high-speed connector detection method based on data analysis according to any one of claims 1 to 6 are implemented.

Images