Automobile wire harness terminal aging diagnosis auxiliary system and method

Abstract

The invention belongs to the technical field of automobile wire harness terminal diagnosis, and discloses an automobile wire harness terminal aging diagnosis auxiliary system and method, and the method comprises the steps: collecting the state data of a wire harness terminal, and collecting the data of a working environment where the wire harness terminal is located; correcting the electrical parameter data according to the working environment data, and further correcting the physical state data to obtain an actual state data set; constructing an aging diagnosis model, setting the framework of the aging diagnosis model as a sharing layer and branch layers, setting the branch layers as a first branch layer and a second branch layer, reconstructing a loss function for the first branch layer and the second branch layer, and taking the actual state data set as input to train the aging diagnosis model; outputting the predicted aging trend and the remaining service life of the wire harness terminal based on the aging diagnosis model; and according to the predicted aging trend and the residual service life of the wire harness terminal, analyzing whether the wire harness terminal has potential abnormity, and sending a maintenance prompt to a user.

Description

Technical Field

[0001] The present invention relates to the technical field of automobile wiring harness terminal diagnosis, and more specifically, to an automobile wiring harness terminal aging diagnosis auxiliary system and method. Background Art

[0002] The aging problem of automotive wiring harness terminals directly affects the safety and reliability of the vehicle. As the use time increases, the wiring harness terminals may gradually age due to the influence of their own electrical parameters and physical state. Traditional detection methods mainly rely on regular manual inspections or simple threshold alarm systems. These methods often fail to detect potential aging problems in time, resulting in processing after the failure occurs.

[0003] There are many shortcomings in the existing wiring harness terminal monitoring and diagnosis systems: first, these systems usually only focus on electrical parameters (such as contact resistance, current intensity, etc.), but ignore the important impact of environmental factors (such as temperature, humidity) on the performance of wiring harness terminals; second, since sensors may produce measurement errors under different environmental conditions, the existing systems lack an effective environmental correction mechanism, resulting in insufficient data accuracy; in addition, many existing systems rely on a single or a few key performance indicators to evaluate the aging status, ignoring the comprehensive analysis of multi-dimensional data, and it is difficult to fully reflect the actual health status of the equipment; finally, most traditional prediction models use simple regression analysis methods, which fail to fully consider the complex nonlinear relationships in the aging process, thus affecting the accuracy of the prediction. Summary of the invention

[0004] In order to overcome the above-mentioned defects of the prior art and to achieve the above-mentioned purpose, the present invention provides the following technical solution: an automobile wiring harness terminal aging diagnosis auxiliary system, comprising: Multi-dimensional information perception unit: captures the status data of the wiring harness terminal in real time, and collects the working environment data of the wiring harness terminal operation scene; Environmental coupling compensation unit: used to correct electrical parameter data according to working environment data, use the corrected electrical parameter data to correct physical state data, and obtain the actual state data set according to the corrected electrical parameter data and physical state data combined with the working environment data; Aging intelligent assessment unit: construct an aging diagnosis model, set the aging diagnosis model architecture as a shared layer and a branch layer, set the branch layer to the first branch layer and the second branch layer, and reconstruct the loss function for the first branch layer and the second branch layer, use the actual state data set as input to train the aging diagnosis model, and output the predicted aging trend and remaining service life of the wiring harness terminal based on the aging diagnosis model; Real-time monitoring and early warning unit: Analyzes whether there are potential abnormalities in the wiring harness terminals based on the predicted aging trend and remaining service life of the wiring harness terminals, and issues preventive maintenance instructions to users.

[0005] Further, the state data includes electrical parameter data and physical state data; Electrical parameter data include contact resistance, current intensity and voltage drop; contact resistance refers to the resistance of the connection point between the wiring harness terminal and its matching conductive part, current intensity refers to the current value passing through the cross section of the wiring harness terminal, and voltage drop refers to the voltage difference generated by the resistance of the wiring harness terminal itself and the contact resistance when the current passes through the wiring harness terminal; Physical status data includes harness terminal temperature and mechanical stress; Working environment data include external temperature, external humidity and corrosive gas concentration; Perform data cleaning, missing value filling and standardization on electrical parameter data, physical status data and working environment.

[0006] Furthermore, the actual status data set is obtained in the following manner: According to all the processed data, the contact resistance is corrected by the external humidity using the humidity correction resistance formula to obtain the humidity correction resistance. The humidity correction resistance is corrected by the external temperature using the temperature correction resistance formula to obtain the temperature and humidity correction resistance. The temperature and humidity correction resistance is corrected by the concentration of corrosive gas to obtain the final correction resistance. Based on the final corrected resistance, the current intensity and the voltage drop are corrected to obtain a corrected current and a corrected voltage; The temperature and mechanical stress of the wiring harness terminal are corrected by using the final correction resistance, correction current and correction voltage combined with the external humidity to obtain the corrected temperature and corrected mechanical stress of the wiring harness terminal; The working environment data is integrated with the corrected electrical parameter data and physical state data into an actual state data set.

[0007] Furthermore, the humidity correction resistor is obtained in the following manner: In the test environment, test the contact resistance of the wiring harness terminals under different humidity conditions and record the corresponding humidity-resistance data; Define the average outdoor humidity of the area as the initial reference humidity. According to the humidity-resistance data, take the contact resistance value under the initial reference humidity as the initial contact resistance, and take the contact resistance value corresponding to the maximum humidity as the initial maximum contact resistance value. At the same time, set the initial control parameters. The set initial reference humidity, initial contact resistance value, initial maximum contact resistance value and initial control parameters are used as the initial parameter combination of the humidity correction resistance formula, and the nonlinear regression tool is used to iterate the parameters of the humidity correction resistance formula starting from the initial parameter combination according to the humidity-resistance data, and the humidity correction resistance estimation value under the current parameter combination is calculated in each iteration; The least square method is used to calculate the sum of square errors between the estimated value of humidity-corrected resistance under the current parameter combination and the actual contact resistance value, and then the parameter combination is adjusted by the gradient descent method to reduce the error; When the error between the humidity-corrected resistance estimate and the actual contact resistance is less than a preset resistance error threshold or the maximum number of iterations is reached, the iteration stops and the optimal parameter combination is obtained; The optimal parameter combination is substituted into the humidity correction resistance formula, and the humidity correction resistance of the wiring harness terminal at time t is calculated according to the external humidity at time t using the humidity correction resistance formula.

[0008] Furthermore, the final corrected resistance is obtained in the following manner: Based on the humidity correction resistance, the temperature and humidity correction resistance is calculated by using the temperature correction resistance formula through the external temperature at time t; Define room temperature as the reference temperature, test the contact resistance of the wiring harness terminal at different temperatures under the test environment, and record the corresponding temperature-resistance data; The initial temperature sensitivity coefficient is set. According to the temperature-resistance data, the initial temperature sensitivity coefficient and the humidity correction resistance are substituted into the temperature correction resistance formula to calculate the predicted temperature and humidity correction resistance value at each temperature point. The least square method is used to calculate and minimize the sum of square errors between the predicted temperature and humidity correction resistance value and the actual resistance value in the temperature-resistance data, and the optimal temperature sensitivity coefficient that minimizes the sum of square errors is obtained. Substitute the optimal temperature sensitivity coefficient into the temperature correction resistance formula, and use the temperature correction resistance formula to calculate the temperature and humidity correction resistance through the external temperature and humidity correction resistance at time t; Under the test environment, test the contact resistance of the wiring harness terminals under different corrosive gas concentrations and record the corresponding corrosion-resistance data; The initial parameters of the corrosion gas compensation formula are set, and according to the corrosion-resistance data, the initial parameters are substituted into the corrosion gas compensation formula to calculate the predicted final corrected resistance value under each corrosion concentration gas, and the least square method is used to calculate and minimize the sum of square errors between the predicted final corrected resistance value and the actual resistance value in the corrosion-resistance data, so as to obtain the optimal parameters that minimize the sum of square errors; Substitute the optimal parameters of the corrosion gas compensation formula into the corrosion gas compensation formula, correct the resistance based on temperature and humidity, and use the corrosion gas compensation formula to calculate the final corrected resistance through the corrosion gas concentration at time t.

[0009] Further, the correction temperature and the correction mechanical stress of the wiring harness terminal are obtained by: According to Ohm's law, the ratio of the contact resistance at time t to the final corrected resistance is calculated, and the result is multiplied by the current intensity to obtain the corrected current; Take the ratio of the final corrected resistance to the contact resistance, multiply it by the voltage drop, and get the corrected voltage; Based on the corrected resistance and current, the temperature change of the wiring harness terminal due to the Joule heating effect is calculated; The sum of the outside temperature and the temperature change of the wiring harness terminal is taken as the correction temperature of the wiring harness terminal; The product of the corrected temperature of the harness terminal and the elastic modulus and thermal expansion coefficient of the harness terminal is taken as the mechanical stress change caused by thermal expansion; The sum of the mechanical stress and the mechanical stress change is taken as the corrected mechanical stress of the wiring harness terminal.

[0010] Furthermore, the training method of the aging diagnosis model includes: Step 71: Use the LSTM model to build an aging diagnosis model, set the aging diagnosis model architecture to a shared layer and a branch layer, use the LSTM layer as the shared layer, set the branch layer to the first branch layer and the second branch layer, use the actual state sample data set as input, input it into the shared layer, and the output of the shared layer is input into the first branch layer and the second branch layer respectively; The first branch layer extracts information specific to the aging trend from the output of the shared layer, and predicts it through the fully connected layer to output the aging trend of the wiring harness terminal; The second branch layer extracts information specific to the remaining useful life from the output of the shared layer, and predicts it through the fully connected layer to output the remaining useful life of the wiring harness terminal; Step 72: Collect a training sample set, which includes an actual state data set and a corresponding aging trend label and a remaining service life label; the aging trend label indicates the value of the corrected state data of the wiring harness terminal changing over time, and the remaining service life label indicates the remaining normal working time of the wiring harness terminal; Step 73: Divide the training sample set into a training set and a validation set in proportion; Step 74: define a loss function of the first branch layer and a loss function of the second branch layer; Combine the loss functions of the first branch layer and the second branch layer to obtain the total loss function; Set the initial loss parameter combination, including ,pc, , and , use network search to optimize parameters, and use cross-validation method to evaluate different loss function parameter combinations to obtain the optimal loss function parameter combination; Step 75: Use the training set to forward propagate the model to obtain the predicted values ​​of aging trend and remaining service life, use the corresponding loss functions to calculate the losses of the first branch layer and the second branch layer respectively, and then calculate the total loss function, and perform back propagation through the total loss function. In each iteration of the model, use the Adam optimization algorithm to update the model parameters, including the parameters of the shared layer and the two branch layers; Step 76: For each iteration, use AUC as the evaluation metric and calculate the AUC value on the validation set; According to the AUC value on the validation set, calculate the difference between the AUC value after the current iteration and the AUC value of the previous iteration, which is recorded as the iteration difference; Step 77: setting an iteration difference threshold, if the iteration difference is greater than the iteration difference threshold, it is determined that the performance of the model is improved; If the iteration difference is less than or equal to the iteration difference threshold, it is determined that the performance of the model has not been improved; If the performance of the model on the validation set does not improve in consecutive DD iterations, the training is stopped to obtain a trained aging diagnosis model.

[0011] Furthermore, the method of analyzing whether there is a potential abnormality in the wiring harness terminal includes: Acquire the state data of the wiring harness terminal at the current moment ut, and use the aging prediction model to calculate and output the aging trend of the corrected state data of the wiring harness terminal over time starting from the current moment; Set the time window uu, and take the current moment uu as the moment before the time window. In the time window, continuously collect the status data of the wiring harness terminal. At the moment (ut+uu), obtain the latest status data for correction. Perform difference calculation between the corrected latest status data and the predicted value of the status data at the same moment in the aging trend to obtain the status deviation value. If the state deviation value is greater than the preset state deviation threshold, it is determined that the actual aging speed is greater than the predicted aging trend; If the state deviation value is less than or equal to the preset state deviation threshold, it is determined that the actual aging speed is less than or equal to the predicted aging trend; If it is determined that the actual aging rate is greater than the predicted aging trend, further analysis is conducted to determine whether it is caused by sudden environmental factors. If not, analysis is conducted to determine whether there are potential abnormalities in the wiring harness terminals.

[0012] Further, the analysis is whether it is caused by sudden environmental factors. If not, the way to analyze whether there is a potential abnormality in the wiring harness terminal includes: If it is determined that the actual aging speed is greater than the predicted aging trend, the time period in which the actual aging speed begins to increase is identified and obtained, and is marked as an increasing time period, and a time period of equal length in which the actual aging speed does not begin to increase before the increasing time period is marked as a normal time period; Calculate the absolute difference between the average working environment data in the increased time period and the average working environment data in the normal time period. If the absolute difference is greater than the preset environment difference threshold, it is determined that the actual aging speed is greater than the predicted aging trend, which is caused by sudden environmental factors. If the absolute difference is less than or equal to the preset environmental difference threshold, it is determined that the actual aging speed is greater than the predicted aging trend and is not caused by sudden environmental factors; If it is determined that it is not caused by sudden environmental factors, use the latest status data at time (ut+uu) to recalculate the latest predicted remaining service life through the aging prediction model, set the safety margin qp%, and if the latest predicted remaining service life is greater than or equal to qp% of the predicted remaining service life at time ut, it is determined that there is no potential abnormality in the harness terminal; If the latest predicted remaining service life is less than qp% of the predicted remaining service life at time ut, it is determined that there is a potential abnormality in the wiring harness terminal.

[0013] Furthermore, an auxiliary method for diagnosing aging of automobile wiring harness terminals comprises: S1, real-time capture of wiring harness terminal status data, and simultaneously collection of wiring harness terminal operating scenario working environment data; S2, correcting electrical parameters according to the working environment data, and using the corrected electrical parameters to correct the physical state data to obtain an actual state data set; S3, constructing an aging diagnosis model, setting the aging diagnosis model architecture to a shared layer and a branch layer, setting the branch layer to the first branch layer and the second branch layer, and reconstructing the loss function for the first branch layer and the second branch layer, using the actual state data set as input to train the aging diagnosis model, and outputting the predicted aging trend and remaining service life of the wiring harness terminal based on the aging diagnosis model; S4. Analyze whether there are potential abnormalities in the wiring harness terminals based on the predicted aging trend and remaining service life of the wiring harness terminals, and issue preventive maintenance instructions to the user.

[0014] The technical effects and advantages of the automotive wiring harness terminal aging diagnosis auxiliary system and method of the present invention are as follows: The present invention comprehensively reflects the actual health status of the wiring harness terminals by collecting electrical parameters (such as contact resistance, current intensity, voltage drop) and physical state data (such as temperature, mechanical stress) of the wiring harness terminals, as well as their working environment data (such as external temperature, humidity, and corrosive gas concentration), eliminating the limitations of single-dimensional data monitoring; secondly, a multi-level data correction formula (including humidity correction resistance formula, temperature correction resistance formula, and corrosive gas compensation formula) is used to reconstruct the original contact resistance data with errors caused by environmental influences, reduce the impact of environmental factors on the measurement results, ensure that the final resistance value is closer to the true value, and correct other electrical parameters and physical state data, thereby improving the accuracy and reliability of the data; then, the LSTM model is used to construct an aging diagnosis model, which includes two branch layers, Focusing on aging trend prediction and remaining service life prediction respectively, the prediction accuracy and pertinence are significantly improved through joint optimization through customized loss functions. Finally, based on the aging trend and remaining service life prediction values ​​output by the aging diagnosis model, the status data of the wiring harness terminal is monitored in real time and compared with the predicted trend. When the actual aging rate is greater than the predicted trend, it is analyzed whether it is caused by sudden environmental factors, which helps to eliminate external interference factors and achieve global optimization and flexible adjustment of the health status of the wiring harness terminal. It avoids the expansion of problems and reduces the impact of abnormal situations on the system. It can not only effectively prevent the occurrence of potential problems, but also maximize resource utilization efficiency, reduce operating costs, ensure the stable operation of the wiring harness terminal, improve the stability and reliability of the entire system, and ensure instant response and high-precision anomaly detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 A schematic diagram of an automotive wiring harness terminal aging diagnosis auxiliary system of the present invention; Figure 2 A schematic diagram of an auxiliary method for diagnosing aging of automobile wiring harness terminals according to the present invention; Figure 3 The present invention is a schematic diagram of an abnormality detection decision process of an automobile wiring harness terminal aging diagnosis auxiliary system. DETAILED DESCRIPTION

[0016] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0017] Embodiment 1 See also Figure 1 and Figure 3As shown, this embodiment provides an automotive wiring harness terminal aging diagnosis auxiliary system, including: Multi-dimensional information perception unit: captures the status data of the wiring harness terminal in real time, and collects the working environment data of the wiring harness terminal operation scene; Environmental coupling compensation unit: used to correct electrical parameter data according to working environment data, including: using humidity correction resistance formula to correct contact resistance to obtain humidity correction resistance, using temperature correction resistance formula to correct humidity correction resistance to obtain temperature and humidity correction resistance, correcting temperature and humidity correction resistance by corrosive gas concentration to obtain final correction resistance; based on the final correction resistance, correcting current intensity and voltage drop to obtain correction current and correction voltage; The physical state data is corrected using the corrected electrical parameter data, including: using the final corrected resistance, corrected current and corrected voltage in combination with the external humidity to correct the temperature and mechanical stress of the wiring harness terminal to obtain the corrected temperature and corrected mechanical stress of the wiring harness terminal; According to the corrected electrical parameter data and physical state data, combined with the working environment data, the actual state data set is obtained; Aging intelligent assessment unit: construct an aging diagnosis model, set the aging diagnosis model architecture as a shared layer and a branch layer, set the branch layer to the first branch layer and the second branch layer, and reconstruct the loss function for the first branch layer and the second branch layer, use the actual state data set as input to train the aging diagnosis model, and output the predicted aging trend and remaining service life of the wiring harness terminal based on the aging diagnosis model; Real-time monitoring and early warning unit: Analyzes whether there are potential abnormalities in the wiring harness terminals based on the predicted aging trend and remaining service life of the wiring harness terminals, and issues preventive maintenance instructions to users; The status data includes electrical parameter data and physical status data; Electrical parameter data include contact resistance (measured using the four-wire method), current intensity (measured using a Hall effect sensor or shunt), and voltage drop (measured using a high-precision voltmeter or differential amplifier); contact resistance represents the resistance of the connection point between the harness terminal and its paired conductive component (the resistance value at the connection point or interface between the two conductive components. This resistance not only includes the resistance of the terminal material itself, but more importantly, it reflects the effective resistance of the actual contact surface due to factors such as surface roughness, oxide layer, and contact pressure. Contact resistance is very important for evaluating the quality of electrical connections, especially in high current applications. Excessive contact resistance may cause local overheating or even fire, which is different from ordinary electrical The resistance value is somewhat different), the current intensity indicates the current value passing through the cross section of the wiring harness terminal (it can also be expressed as the current value flowing through the terminal and its change over time, usually refers to the flow rate of electricity through a certain conductor cross section, that is, the amount of charge passing through the cross section per unit time. The current intensity can be understood as the current size passing through the terminal cross section. The concept of ordinary current is more extensive, and the current intensity is more refined). The voltage drop indicates the voltage difference caused by the resistance and contact resistance of the wiring harness terminal itself when the current passes through the wiring harness terminal (this voltage drop can directly reflect the quality of the terminal connection and whether there is an aging problem. If the aging of the terminal causes its resistance to increase, the corresponding voltage drop will also increase, which may cause the performance of the entire circuit to decline); It should be noted that contact resistance, current intensity and voltage drop appear more often in engineering practice, especially in the fields of electrical connection reliability assessment and troubleshooting; while ordinary resistance, current and voltage are more inclined to basic physical concepts and are more common in theory and technical design stages; contact resistance, current intensity and voltage drop pay more attention to performance and safety considerations in actual operation; in contrast, ordinary resistance, current and voltage focus more on principle analysis and theoretical calculation; Physical condition data includes harness terminal temperature (measured by thermocouples, RTDs, or thermistors) and mechanical stress (measured by strain gauges or other forms of pressure / strain sensors); Working environment data include external temperature, external humidity and corrosive gas concentration (measured by dedicated gas sensors such as electrochemical sensors or infrared absorption sensors); Perform data cleaning, missing value filling and standardization on electrical parameter data, physical status data and working environment; The electrical parameter data is corrected according to the working environment data, and the physical state data is corrected using the corrected electrical parameter data to obtain the actual state data set. The method includes: Since the sensor will have some measurement errors under different environmental conditions (such as temperature, humidity and corrosive gas concentration), by correcting the measurement data in combination with the working environment data, the measurement errors caused by changes in these environmental factors can be compensated to reduce or eliminate the impact of environmental factors on the measurement results; After all the processed data, the contact resistance is corrected by the external humidity using the humidity correction resistance formula to obtain the humidity correction resistance. The humidity correction resistance is corrected by the external temperature using the temperature correction resistance formula to obtain the temperature and humidity correction resistance. The temperature and humidity correction resistance is corrected by the concentration of corrosive gas to obtain the final correction resistance. Based on the final corrected resistance, the current intensity and the voltage drop are corrected to obtain a corrected current and a corrected voltage; The temperature and mechanical stress of the wiring harness terminal are corrected by using the final correction resistance, correction current and correction voltage combined with the external humidity to obtain the corrected temperature and corrected mechanical stress of the wiring harness terminal; Integrate the working environment data with the corrected electrical parameter data and physical state data into an actual state data set; Ways to obtain humidity correction resistance include: The humidity correction resistance of the wiring harness terminal at time t is calculated according to the external humidity at time t using the humidity correction resistance formula; The humidity corrected resistance formula is defined as: ,in, Indicates the humidity correction resistance at the external humidity H, Indicates the maximum contact resistance value (as humidity increases, the material of the wiring harness terminal may absorb moisture, causing the resistance to increase until a saturation point is reached. This is not a single data point that can be obtained by direct measurement and needs to be estimated and fitted). Indicates reference humidity The contact resistance value under this condition, e represents the natural constant, Indicates the control parameter (used to control the steepness of the formula curve, reflecting the rate of change of the effect of humidity on resistance); It should be noted that the temperature-corrected resistance formula represents a curve. is the maximum value of the curve, that is, when H tends to infinity, The limit value of It is the midpoint of the curve, that is, when the humidity reaches this value, the contact resistance is in the middle state; This formula expresses that when the humidity H is less than When the humidity H is equal to When the contact resistance is in the middle state, , when the humidity H is greater than When the contact resistance begins to approach the maximum value As the humidity increases further, the contact resistance gradually approaches ; The humidity-corrected resistance formula can well simulate the nonlinear effect of humidity on contact resistance, and show the saturation effect under high humidity conditions, making the curve closer to the actual situation; the general form of the formula is usually linear or exponential, assuming that the resistance changes at a fixed ratio for every unit increase in humidity, while the humidity-corrected resistance formula is nonlinear and can better capture the complex changes in practical applications. The general form of the formula does not take into account the saturation effect, which may cause the predicted value to be too high or too low under high humidity. The humidity-corrected resistance formula can naturally show that the resistance approaches a certain limit value as the humidity increases, which not only improves the accuracy, but also enhances the adaptability and flexibility, making it suitable for humidity correction under different materials and environmental conditions; Under a test environment (such as a constant temperature and humidity chamber), test the contact resistance of the wiring harness terminals at different humidity levels (for example, from 20% to 95% relative humidity) and record the corresponding humidity-resistance data; Define the average outdoor humidity of the area as the initial reference humidity. According to the humidity-resistance data, take the contact resistance value under the initial reference humidity as the initial contact resistance, and take the contact resistance value corresponding to the maximum humidity as the initial maximum contact resistance value. At the same time, set the initial control parameters (based on experience, such as 0.1). Set the initial reference humidity, initial contact resistance value, initial maximum contact resistance value and initial control parameters As the initial parameter combination of the formula, a nonlinear regression tool (such as scipy.optimize.curve_fit in Python, fitnlm in MATLAB, or nls function in R language) is used to iterate the parameters of the formula starting from the initial parameter combination according to the humidity-resistance data, and the humidity-corrected resistance estimate under the current parameter combination is calculated in each iteration; The least square method is used to calculate the sum of square errors between the estimated value of humidity-corrected resistance under the current parameter combination and the actual contact resistance value, and then the parameter combination is adjusted by the gradient descent method to reduce the error; When the error between the humidity-corrected resistance estimate and the actual contact resistance is less than a preset resistance error threshold or the maximum number of iterations is reached, the iteration stops and the optimal parameter combination is obtained; It should be noted that the purpose of this step is to calibrate the data collected by the sensor and remove the influence of external humidity on the contact resistance, so as to obtain a more accurate contact resistance value; Ways to get the final trimmed resistance include: Based on the humidity correction resistor, the temperature correction resistor is calculated by the external temperature at time t using the temperature correction resistor formula. ,in, It represents the temperature and humidity correction resistance at the external temperature T at time t. represents the reference temperature, T represents the external temperature at time t, represents the temperature sensitivity coefficient; It should be noted that the temperature and humidity corrected resistance formula is obtained by transforming the Arrhenius equation. The general form of the Arrhenius equation is: , where k is the reaction rate constant, A is the frequency factor (also called the pre-exponential factor), which represents the frequency and orientation factors of molecular collisions, is the activation energy, i.e., the minimum energy required for the reaction, R is the ideal gas constant (approximately 8.14 J / (mol.K)), and T is the absolute temperature (in Kelvin). In this method, the original Arrhenius equation is deformed and introduced , is an empirical parameter, similar to the activation energy The combination of the general form and the frequency factor A determines the degree of influence of temperature on the resistance change. The symbols and letters used to represent the general form formula here are independent and have nothing to do with the symbols and letters used in other formulas in the embodiments. When the letters used in the general form are repeated in the embodiments, the embodiments shall prevail. The formula here only serves as a reference for explanation. The original Arrhenius equation is mainly used to describe the relationship between chemical reaction rate and temperature, while the deformed formula can be directly applied to the resistance correction problem in the field of electrical engineering. It provides a simple method to calculate the contact resistance value at different temperatures. The deformed formula directly gives the effect of temperature on contact resistance, avoiding the complex physical derivation and conversion process, and the temperature correction of resistance can be completed through simple exponential calculation. Parameter It can be adjusted according to specific materials and application scenarios, so that the model can flexibly adapt to different environmental conditions. It has high versatility for different types of wiring harness terminals and working environments. The formula structure is simple and intuitive, which is easy for engineers and technicians to understand and implement. You only need to know the humidity correction resistance value and the current temperature T, the corrected resistance value can be quickly calculated; Define room temperature as the reference temperature, test the contact resistance of the wiring harness terminal at different temperatures (for example, from -20°C to 80°C) in the test environment, and record the corresponding temperature-resistance data; Set the initial temperature sensitivity coefficient, substitute the initial temperature sensitivity coefficient into the temperature correction resistance formula to calculate the predicted temperature and humidity correction resistance value at each temperature point according to the temperature-resistance data, use the least squares method to calculate and minimize the square sum of the error between the predicted temperature and humidity correction resistance value and the actual resistance value in the temperature-resistance data, and obtain the optimal temperature sensitivity coefficient that minimizes the square sum of the error ; Based on the temperature and humidity corrected resistance, the influence of the corrosion gas concentration on the contact resistance is further considered, and the corrosion gas compensation formula is used to calculate the final corrected resistance through the corrosion gas concentration at time t; The corrosion gas compensation formula is defined as: ,in, represents the final corrected resistance at time t, represents the Lambert W function, which is used to describe the nonlinear effect of corrosive gas concentration on resistance and is defined as satisfying = The equation is: C represents the current corrosive gas concentration. It represents the corrosion influence coefficient, which reflects the degree of influence of the corrosive gas on the resistance. K represents a constant, which is used to standardize the change of the concentration of the corrosive gas. It should be noted that the corrosion gas compensation formula is to deal with the effect of corrosion gas concentration on resistance by introducing Lambert W function. The effect of corrosion gas on resistance is usually highly nonlinear. The Lambert W function can better describe this complex relationship. Lambert W function is a special function that is often used to solve equations involving exponential terms and can handle complex nonlinear effects. In the formula Partially allows fitting parameters from experimental data and k, thereby more accurately simulating the specific effect of corrosive gas concentration on resistance; Under the test environment, test the contact resistance of the wiring harness terminal under different corrosive gas concentrations (for example, from 0 to 400 ppm, ppm is the unit of corrosive gas concentration), and record the corresponding corrosion-resistance data; Set initial parameters and k, based on the corrosion-resistance data, the initial parameters Substitute k into the corrosion gas compensation formula to calculate the predicted final corrected resistance value under each corrosion gas concentration. Use the least squares method to calculate and minimize the sum of square errors between the predicted final corrected resistance value and the actual resistance value in the corrosion-resistance data, and obtain the optimal parameters that minimize the sum of square errors. and k; Ways to obtain the corrected physical state data include: According to Ohm's law, the ratio of the contact resistance at time t to the final corrected resistance is calculated, and the result is multiplied by the current intensity (the original current intensity directly measured by the sensor) to obtain the corrected current; Take the ratio of the final corrected resistance to the contact resistance, and multiply it by the voltage drop (the original voltage drop value directly measured by the corresponding sensor) to obtain the corrected voltage; The calculated final correction resistance is used to calibrate the current intensity and voltage drop collected by the sensor, which can ensure that the electrical parameter data is closer to the actual value; Calculate the change in temperature of the wiring harness terminals due to Joule heating based on the corrected resistance and current ,in, Indicates the temperature change of the wiring harness terminal. Indicates the correction current, represents the time interval from t-1 to time t, m is the mass of the wiring harness terminal, and c is the specific heat capacity of the wiring harness terminal; The sum of the outside temperature and the wiring harness terminal temperature is taken as the correction temperature of the wiring harness terminal; The product of the corrected temperature of the harness terminal and the elastic modulus and thermal expansion coefficient of the harness terminal is taken as the mechanical stress change caused by thermal expansion; The sum of the mechanical stress and the mechanical stress change is taken as the corrected mechanical stress of the wiring harness terminal; The training methods of the aging diagnosis model include: Step 71: Use the LSTM model to build an aging diagnosis model, set the aging diagnosis model architecture to a shared layer and a branch layer, use the LSTM layer as the shared layer, set the branch layer to the first branch layer and the second branch layer, use the actual state sample data set as input, input it into the shared layer, and the output of the shared layer is input into the first branch layer and the second branch layer respectively; The first branch layer extracts information specific to the aging trend from the output of the shared layer (the shared layer outputs a fixed-dimensional feature vector that represents important information extracted from the input sequence), and predicts it through the fully connected layer to output the aging trend of the wiring harness terminal; The second branch layer extracts information specific to the remaining useful life from the output of the shared layer, and predicts it through the fully connected layer to output the remaining useful life of the wiring harness terminal; Step 72: Collect a training sample set, which includes an actual state data set and a corresponding aging trend label and a remaining service life label; the aging trend label indicates the value of the corrected state data of the wiring harness terminal changing over time, and the remaining service life label indicates the remaining normal working time of the wiring harness terminal; It should be noted that predicting the rate of change of the harness terminal status data over time helps to understand which factors have the greatest impact on aging, including the rate of change of contact resistance, the rate of change of current intensity (although the current intensity itself does not usually "age" directly, its change may indicate changes in load conditions, which in turn affects the aging rate of other parameters), the rate of change of voltage drop, the rate of change of temperature (if the temperature rises too quickly, it may be a sign of poor heat dissipation or degradation of insulation materials) and the rate of change of mechanical stress (such as increased metal fatigue, which may lead to reduced reliability of connection points); Step 73: Divide the training sample set into a training set and a validation set in proportion; Step 74: Define the loss function of the first branch layer as: ; in, represents the true value, represents the predicted value, ag represents the aging trend prediction task, Represents the aging trend prediction task The true value of , Represents the aging trend prediction task The predicted value of Represents the threshold, which is used to distinguish "large error" from "small error". The loss function can be adaptively adjusted based on the error size, and the loss value is dynamically adjusted according to the size of the prediction error. When the error is less than or equal to When , the power function is used to calculate the loss. This mechanism allows the model to pay more attention to large errors, thereby reducing these large errors more effectively during training. pc represents the power parameter that controls the growth rate of the loss, and 0<pc<1 (such as 0.5). exp() represents the natural exponential function, which is used to give higher penalties to larger errors. Using the exponential function to punish larger errors ensures that the model can quickly adjust parameters when encountering large errors, improves the robustness of the model, and uses the power function for small errors, so that the loss growth under small errors is relatively gentle, avoids excessive punishment of small errors, and maintains the stability of the model. Compared with the basic loss function (such as mean square error MSE), MSE treats all errors equally and cannot distinguish between large errors and small errors. The adaptively adjusted loss function can give higher penalties to large errors, so that the model pays more attention to reducing large errors. The loss function of the second branch layer is defined as: ; Among them, ru represents the remaining useful life prediction task, Remaining useful life prediction task The predicted value of Remaining useful life prediction task The true value of , Represents the coefficient used when the predicted service life is greater than the actual service life, The factor used when the predicted useful life is less than or equal to the actual useful life, and > , to ensure that overly optimistic forecasts are penalized more, relatively small Makes the model more forgiving in this situation; It should be noted that this loss function pays special attention to the situation where the prediction results are too optimistic (that is, the predicted life expectancy is greater than the actual life expectancy). > , giving a greater penalty to such overly optimistic predictions, ensuring that the model does not overestimate the remaining service life, thereby improving the safety and reliability of the system. In many practical applications, overestimation of the remaining service life may lead to serious consequences (such as equipment failure, safety hazards, etc.). This loss function combines the specific needs of this field and provides a more targeted optimization goal. Compared with basic loss functions (such as mean square error MSE), MSE treats the error between the predicted value and the true value equally and cannot distinguish the impact of overestimation and underestimation. This loss function can impose a greater penalty on overestimation, ensuring that the prediction results are more conservative and reliable; Combine the loss functions of the first branch layer and the second branch layer to get the total loss function ,in, Represents a balance factor, which is used to control the relative importance between the aging trend prediction task and the remaining service life prediction task. It takes values ​​between 0 and 1. If the aging trend prediction task is more important, then can be set higher and vice versa; Set the initial loss parameter combination, including ,pc, , and , use network search to optimize parameters, and use cross-validation method to evaluate different loss function parameter combinations to obtain the optimal loss function parameter combination; Step 75: Use the training set to forward propagate the model to obtain the predicted values ​​of aging trend and remaining service life, use the corresponding loss functions to calculate the losses of the first branch layer and the second branch layer respectively, and then calculate the total loss function, and perform back propagation through the total loss function. In each iteration of the model, use the Adam optimization algorithm to update the model parameters, including the parameters of the shared layer and the two branch layers; Step 76: For each iteration, use AUC as the evaluation metric and calculate the AUC value on the validation set; According to the AUC value on the validation set, calculate the difference between the AUC value after the current iteration and the AUC value of the previous iteration, which is recorded as the iteration difference; Step 77: setting an iteration difference threshold, if the iteration difference is greater than the iteration difference threshold, it is determined that the performance of the model is improved; If the iteration difference is less than or equal to the iteration difference threshold, it is determined that the performance of the model has not been improved; If the performance of the model on the validation set does not improve in consecutive DD iterations, the training is stopped to obtain a trained aging diagnosis model; Ways to compare real-time data with predicted aging trends include: Acquire the state data of the wiring harness terminal at the current moment ut, and use the aging prediction model to calculate and output the aging trend of the corrected state data of the wiring harness terminal over time starting from the current moment; Set a time window uu (such as one week or one month), and use the current moment uu as the previous moment of the time window. Within the time window, continuously collect the status data of the wiring harness terminal. At the moment (ut+uu), obtain the latest status data for correction. Perform difference calculation between the corrected latest status data and the predicted value of the status data at the same moment in the aging trend to obtain the status deviation value. If the state deviation value is greater than the preset state deviation threshold (set by industry personnel based on experience), it is determined that the actual aging speed is greater than the predicted aging trend; If the state deviation value is less than or equal to the preset state deviation threshold, it is determined that the actual aging speed is less than or equal to the predicted aging trend; Ways to determine if there are abnormalities and potential failure risks in the wiring harness terminals include: If it is determined that the actual aging speed is greater than the predicted aging trend, the time period in which the actual aging speed begins to increase is identified and obtained, and is marked as an increasing time period, and a time period of equal length in which the actual aging speed does not begin to increase before the increasing time period is marked as a normal time period; Calculate the absolute difference between the average working environment data in the increased time period and the average working environment data in the normal time period. If the absolute difference is greater than the preset environment difference threshold (set by industry personnel based on experience), it is determined that the actual aging rate is greater than the predicted aging trend, which is caused by sudden environmental factors. If the absolute difference is less than or equal to the preset environmental difference threshold, it is determined that the actual aging speed is greater than the predicted aging trend and is not caused by sudden environmental factors; If it is determined that it is not caused by sudden environmental factors, use the latest status data at the time (ut+uu) to recalculate the latest predicted remaining service life through the aging prediction model, set the safety margin qp% (set by relevant personnel based on industry experience and the importance of the equipment, such as 80%), and if the latest predicted remaining service life is greater than or equal to qp% of the predicted remaining service life at the time ut, it is determined that there is no potential abnormality in the harness terminal; If the latest predicted remaining service life is less than qp% of the predicted remaining service life at time ut, it is determined that there is a potential abnormality in the wiring harness terminal.

[0018] In this embodiment, by collecting the electrical parameters (such as contact resistance, current intensity, voltage drop) and physical state data (such as temperature, mechanical stress) of the wiring harness terminal, as well as its working environment data (such as external temperature, humidity, and corrosive gas concentration), the actual health status of the wiring harness terminal is fully reflected, eliminating the limitation of single-dimensional data monitoring; secondly, a multi-level data correction formula (including humidity correction resistance formula, temperature correction resistance formula, and corrosive gas compensation formula) is adopted to reconstruct the original contact resistance data with errors caused by environmental influences, reduce the influence of environmental factors on the measurement results, ensure that the final resistance value is closer to the true value, and correct other electrical parameters and physical state data, thereby improving the accuracy and reliability of the data; then, the LSTM model is used to construct an aging diagnosis model, which includes two branch layers , focusing on aging trend prediction and remaining service life prediction respectively, and jointly optimizing through customized loss functions, significantly improving the accuracy and pertinence of the prediction; finally, based on the aging trend and remaining service life prediction values ​​output by the aging diagnosis model, the status data of the wiring harness terminal is monitored in real time and compared with the predicted trend. When the actual aging speed is greater than the predicted trend, it is analyzed whether it is caused by sudden environmental factors, helping to eliminate external interference factors, and realizing global optimization and flexible adjustment of the health status of the wiring harness terminal; avoiding the expansion of problems and reducing the impact of abnormal situations on the system, it can not only effectively prevent the occurrence of potential problems, but also maximize resource utilization efficiency, reduce operating costs, ensure the stable operation of the wiring harness terminal, improve the stability and reliability of the entire system, and ensure instant response and high-precision anomaly detection.

[0019] Embodiment 2

[0020] See also Figure 2 As shown, the part not described in detail in this embodiment is described in Example 1, which provides an auxiliary method for diagnosing aging of automobile wiring harness terminals, including: S1, real-time capture of wiring harness terminal status data, and simultaneously collection of wiring harness terminal operating scenario working environment data; S2, correcting electrical parameters according to the working environment data, and using the corrected electrical parameters to correct the physical state data to obtain an actual state data set; S3, constructing an aging diagnosis model, setting the aging diagnosis model architecture to a shared layer and a branch layer, setting the branch layer to the first branch layer and the second branch layer, and reconstructing the loss function for the first branch layer and the second branch layer, using the actual state data set as input to train the aging diagnosis model, and outputting the predicted aging trend and remaining service life of the wiring harness terminal based on the aging diagnosis model; S4. Analyze whether there are potential abnormalities in the wiring harness terminals based on the predicted aging trend and remaining service life of the wiring harness terminals, and issue preventive maintenance instructions to the user.

[0021] Embodiment 3

[0022] This embodiment discloses an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the operation mode of the above-mentioned automobile wiring harness terminal aging diagnosis auxiliary method is implemented.

[0023] Since the electronic device introduced in this embodiment is an electronic device used to implement an auxiliary method for diagnosing the aging of automobile wiring harness terminals in the embodiment of this application, based on the auxiliary method for diagnosing the aging of automobile wiring harness terminals introduced in the embodiment of this application, the technical personnel of this field can understand the specific implementation of the electronic device of this embodiment and its various variations, so how the electronic device implements the method in the embodiment of this application will not be described in detail here. As long as the technical personnel of this field implement the electronic device used in the auxiliary method for diagnosing the aging of automobile wiring harness terminals in the embodiment of this application, it belongs to the scope of protection of this application.

[0024] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters and thresholds in the formula are set by technicians in this field according to actual conditions.

[0025] The above is only a preferred embodiment of the present invention, and the protection scope of the present invention is not limited to the above embodiments. All technical solutions under the concept of the present invention belong to the protection scope of the present invention. It should be pointed out that for ordinary technical users in this technical field, some improvements and modifications without departing from the principle of the present invention should also be regarded as the protection scope of the present invention.

Claims

1. An automotive wiring harness terminal aging diagnosis auxiliary system, characterized in that: include: Multi-dimensional information perception unit: captures the status data of the wiring harness terminal in real time, and collects the working environment data of the wiring harness terminal operation scene; Among them, the status data includes electrical parameter data and physical status data; Environmental coupling compensation unit: used to correct the electrical parameter data according to the working environment data, including: correcting the contact resistance to obtain the final corrected resistance, and correcting the remaining electrical parameters by the final corrected resistance to obtain the corrected electrical parameter data; Then, the physical state data is corrected using the corrected electrical parameter data, and an actual state data set is obtained according to the corrected electrical parameter data and the physical state data; Aging intelligent assessment unit: construct an aging diagnosis model, set the aging diagnosis model architecture as a shared layer and a branch layer, set the branch layer to the first branch layer and the second branch layer, and reconstruct the loss function for the first branch layer and the second branch layer, use the actual state data set as input to train the aging diagnosis model, and output the predicted aging trend and remaining service life of the wiring harness terminal based on the aging diagnosis model; Real-time monitoring and early warning unit: Analyzes whether there are potential abnormalities in the wiring harness terminals based on the predicted aging trend and remaining service life of the wiring harness terminals, and issues preventive maintenance instructions to users.

2. The automotive wiring harness terminal aging diagnosis auxiliary system according to claim 1 is characterized in that: The electrical parameter data includes contact resistance, current intensity and voltage drop; the contact resistance represents the resistance of the connection point between the wiring harness terminal and its paired conductive component, the current intensity represents the current value passing through the cross section of the wiring harness terminal, and the voltage drop represents the voltage difference generated by the resistance of the wiring harness terminal itself and the contact resistance when the current passes through the wiring harness terminal; Physical status data includes harness terminal temperature and mechanical stress; Working environment data include external temperature, external humidity and corrosive gas concentration; Perform data cleaning, missing value filling and standardization on electrical parameter data, physical status data and working environment data.

3. The automotive wiring harness terminal aging diagnosis auxiliary system according to claim 2 is characterized in that: The actual status data set is obtained in the following manner: According to the processed state data and working environment data, the humidity correction resistance formula is used to correct the contact resistance by the external humidity to obtain the humidity correction resistance; Use the temperature correction resistance formula to correct the humidity correction resistance according to the external temperature to obtain the temperature and humidity correction resistance; The temperature and humidity correction resistor is corrected by the concentration of the corrosive gas to obtain the final correction resistor; Based on the final corrected resistance, the current intensity and the voltage drop are corrected to obtain a corrected current and a corrected voltage; The temperature and mechanical stress of the wiring harness terminal are corrected by using the final correction resistance, correction current and correction voltage combined with the external humidity to obtain the corrected temperature and corrected mechanical stress of the wiring harness terminal; The working environment data is integrated with the corrected electrical parameter data and physical state data into an actual state data set.

4. The automotive wiring harness terminal aging diagnosis auxiliary system according to claim 3 is characterized in that: The humidity correction resistor is obtained by: In the test environment, test the contact resistance of the wiring harness terminals under different humidity conditions and record the corresponding humidity-resistance data; Define the average outdoor humidity of the area as the initial reference humidity. According to the humidity-resistance data, take the contact resistance value under the initial reference humidity as the initial contact resistance, and take the contact resistance value corresponding to the maximum humidity as the initial maximum contact resistance value. At the same time, set the initial control parameters. The set initial reference humidity, initial contact resistance value, initial maximum contact resistance value and initial control parameters are used as the initial parameter combination of the humidity correction resistance formula, and the nonlinear regression tool is used to iterate the parameters of the humidity correction resistance formula starting from the initial parameter combination according to the humidity-resistance data, and the humidity correction resistance estimation value under the current parameter combination is calculated in each iteration; The least square method is used to calculate the sum of square errors between the estimated value of humidity-corrected resistance under the current parameter combination and the actual contact resistance value, and then the parameter combination is adjusted by the gradient descent method to reduce the error; When the error between the humidity-corrected resistance estimate and the actual contact resistance is less than a preset resistance error threshold or the maximum number of iterations is reached, the iteration stops and the optimal parameter combination is obtained; The optimal parameter combination is substituted into the humidity correction resistance formula, and the humidity correction resistance of the wiring harness terminal at time t is calculated according to the external humidity at time t using the humidity correction resistance formula.

5. The automotive wiring harness terminal aging diagnosis auxiliary system according to claim 4, characterized in that: The final corrected resistance is obtained by: Based on the humidity correction resistance, the temperature and humidity correction resistance is calculated by using the temperature correction resistance formula through the external temperature at time t; Define room temperature as the reference temperature, test the contact resistance of the wiring harness terminal at different temperatures under the test environment, and record the corresponding temperature-resistance data; The initial temperature sensitivity coefficient is set. According to the temperature-resistance data, the initial temperature sensitivity coefficient and the humidity correction resistance are substituted into the temperature correction resistance formula to calculate the predicted temperature and humidity correction resistance value at each temperature point. The least square method is used to calculate and minimize the sum of square errors between the predicted temperature and humidity correction resistance value and the actual resistance value in the temperature-resistance data, and the optimal temperature sensitivity coefficient that minimizes the sum of square errors is obtained. Substitute the optimal temperature sensitivity coefficient into the temperature correction resistance formula, and use the temperature correction resistance formula to calculate the temperature and humidity correction resistance through the external temperature and humidity correction resistance at time t; Under the test environment, test the contact resistance of the wiring harness terminals under different corrosive gas concentrations and record the corresponding corrosion-resistance data; The initial parameters of the corrosion gas compensation formula are set, and according to the corrosion-resistance data, the initial parameters are substituted into the corrosion gas compensation formula to calculate the predicted final corrected resistance value under each corrosion concentration gas, and the least square method is used to calculate and minimize the sum of square errors between the predicted final corrected resistance value and the actual resistance value in the corrosion-resistance data, so as to obtain the optimal parameters that minimize the sum of square errors; Substitute the optimal parameters of the corrosion gas compensation formula into the corrosion gas compensation formula, correct the resistance based on temperature and humidity, and use the corrosion gas compensation formula to calculate the final corrected resistance through the corrosion gas concentration at time t.

6. The automotive wiring harness terminal aging diagnosis auxiliary system according to claim 5, characterized in that: The corrected temperature and corrected mechanical stress of the wiring harness terminal are obtained by: According to Ohm's law, the ratio of the contact resistance at time t to the final corrected resistance is calculated, and the result is multiplied by the current intensity to obtain the corrected current; Take the ratio of the final corrected resistance to the contact resistance, multiply it by the voltage drop, and get the corrected voltage; Based on the corrected resistance and current, the temperature change of the wiring harness terminal due to the Joule heating effect is calculated; The sum of the outside temperature and the temperature change of the wiring harness terminal is taken as the correction temperature of the wiring harness terminal; The product of the corrected temperature of the harness terminal and the elastic modulus and thermal expansion coefficient of the harness terminal is taken as the mechanical stress change caused by thermal expansion; The sum of the mechanical stress and the mechanical stress change is taken as the corrected mechanical stress of the wiring harness terminal.

7. The automotive wiring harness terminal aging diagnosis auxiliary system according to claim 6, characterized in that: The training method of the aging diagnosis model includes: Step 71: Use the LSTM model to build an aging diagnosis model, set the aging diagnosis model architecture to a shared layer and a branch layer, use the LSTM layer as the shared layer, set the branch layer to the first branch layer and the second branch layer, use the actual state sample data set as input, input it into the shared layer, and the output of the shared layer is input into the first branch layer and the second branch layer respectively; The first branch layer extracts information specific to the aging trend from the output of the shared layer, and predicts it through the fully connected layer to output the aging trend of the wiring harness terminal; The second branch layer extracts information specific to the remaining useful life from the output of the shared layer, and predicts it through the fully connected layer to output the remaining useful life of the wiring harness terminal; Step 72: Collect a training sample set, which includes an actual state data set and a corresponding aging trend label and a remaining service life label; the aging trend label indicates the value of the corrected state data of the wiring harness terminal changing over time, and the remaining service life label indicates the remaining normal working time of the wiring harness terminal; Step 73: Divide the training sample set into a training set and a validation set in proportion; Step 74: define a loss function of the first branch layer and a loss function of the second branch layer; Combine the loss functions of the first branch layer and the second branch layer to obtain the total loss function; Step 75: Use the training set to forward propagate the model to obtain the predicted values ​​of aging trend and remaining service life, use the corresponding loss functions to calculate the losses of the first branch layer and the second branch layer respectively, and then calculate the total loss function, and perform back propagation through the total loss function. In each iteration of the model, use the Adam optimization algorithm to update the model parameters, including the parameters of the shared layer and the two branch layers; Step 76: For each iteration, use AUC as the evaluation metric and calculate the AUC value on the validation set; According to the AUC value on the validation set, calculate the difference between the AUC value after the current iteration and the AUC value of the previous iteration, which is recorded as the iteration difference; Step 77: setting an iteration difference threshold, if the iteration difference is greater than the iteration difference threshold, it is determined that the performance of the model is improved; If the iteration difference is less than or equal to the iteration difference threshold, it is determined that the performance of the model has not been improved; If the performance of the model on the validation set does not improve in consecutive DD iterations, the training is stopped to obtain a trained aging diagnosis model.

8. The automotive wiring harness terminal aging diagnosis auxiliary system according to claim 7, characterized in that: The method of analyzing whether there is a potential abnormality in the wiring harness terminal includes: Acquire the state data of the wiring harness terminal at the current moment ut, and use the aging prediction model to calculate and output the aging trend of the corrected state data of the wiring harness terminal over time starting from the current moment; Set the time window uu, and take the current moment uu as the moment before the time window. In the time window, continuously collect the status data of the wiring harness terminal. At the moment (ut+uu), obtain the latest status data for correction. Perform difference calculation between the corrected latest status data and the predicted value of the status data at the same moment in the aging trend to obtain the status deviation value. If the state deviation value is greater than the preset state deviation threshold, it is determined that the actual aging speed is greater than the predicted aging trend; If the state deviation value is less than or equal to the preset state deviation threshold, it is determined that the actual aging speed is less than or equal to the predicted aging trend; If it is determined that the actual aging rate is greater than the predicted aging trend, further analysis is conducted to determine whether it is caused by sudden environmental factors. If not, analysis is conducted to determine whether there are potential abnormalities in the wiring harness terminals.

9. The automotive wiring harness terminal aging diagnosis auxiliary system according to claim 8, characterized in that: The analysis is whether it is caused by sudden environmental factors. If not, the methods for analyzing whether there are potential abnormalities in the wiring harness terminals include: If it is determined that the actual aging speed is greater than the predicted aging trend, the time period in which the actual aging speed begins to increase is identified and obtained, and is marked as an increasing time period, and a time period of equal length in which the actual aging speed does not begin to increase before the increasing time period is marked as a normal time period; Calculate the absolute difference between the average working environment data in the increased time period and the average working environment data in the normal time period. If the absolute difference is greater than the preset environment difference threshold, it is determined that the actual aging speed is greater than the predicted aging trend, which is caused by sudden environmental factors. If the absolute difference is less than or equal to the preset environmental difference threshold, it is determined that the actual aging speed is greater than the predicted aging trend and is not caused by sudden environmental factors; If it is determined that it is not caused by sudden environmental factors, use the latest status data at time (ut+uu) to recalculate the latest predicted remaining service life through the aging prediction model, set the safety margin qp%, and if the latest predicted remaining service life is greater than or equal to qp% of the predicted remaining service life at time ut, it is determined that there is no potential abnormality in the harness terminal; If the latest predicted remaining service life is less than qp% of the predicted remaining service life at time ut, it is determined that there is a potential abnormality in the wiring harness terminal.

10. An auxiliary method for diagnosing the aging of automobile wiring harness terminals, applied to the auxiliary system for diagnosing the aging of automobile wiring harness terminals as claimed in any one of claims 1 to 9, characterized in that: include: S1, real-time capture of wiring harness terminal status data, and simultaneously collection of wiring harness terminal operating scenario working environment data; S2, correcting electrical parameters according to the working environment data, and using the corrected electrical parameters to correct the physical state data to obtain an actual state data set; S3, constructing an aging diagnosis model, setting the aging diagnosis model architecture to a shared layer and a branch layer, setting the branch layer to the first branch layer and the second branch layer, and reconstructing the loss function for the first branch layer and the second branch layer, using the actual state data set as input to train the aging diagnosis model, and outputting the predicted aging trend and remaining service life of the wiring harness terminal based on the aging diagnosis model; S4. Analyze whether there are potential abnormalities in the wiring harness terminals based on the predicted aging trend and remaining service life of the wiring harness terminals, and issue preventive maintenance instructions to the user.

Images