Intelligent monitoring management system and method for wire harness production line

Abstract

The application discloses a wire harness production line intelligent monitoring management system and method, relates to the technical field of wire harness manufacturing, and comprises the following steps: extracting crimping force and die temperature data from historical wire harness production data, encoding the data into real number encoding gene chains, combining material properties to generate an initial gene library, constructing a digital twin model, simulating stress field distribution and heat conduction effects in the wire harness crimping process through finite element simulation, obtaining twin prediction values, pre-verifying process parameters in the initial gene library based on the digital twin model, screening process parameter combinations with deformation errors less than an error threshold, and issuing the process parameter combinations to a crimping machine through a production line control hub to obtain real-time wire harness deformation data, comparing the real-time wire harness deformation data with the twin prediction values frame by frame, generating a deformation error rate, and generating an abnormality traceability report based on the deformation error rate. The application pre-verifies by constructing an initial gene library and combining a digital twin model, so that the optimization of process parameters is more accurate and effective.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of wire harness manufacturing, in particular to a wire harness production line intelligent monitoring management system and method. BACKGROUND

[0002] With the iterative development of industrial technology, the intelligentization and automation level of wire harness production lines are continuously improving. In traditional technology, wire harness production mainly relies on manual operation and simple mechanical auxiliary equipment, which not only has low efficiency, but also has great difficulty in quality control. In recent years, with the development of computer-aided design, finite element analysis, and digital twin technology, the precision and efficiency of wire harness production have been significantly improved.

[0003] In the existing wire harness production monitoring and management technology, there are still some deficiencies. On the one hand, there is a lack of intelligent method for optimizing key process parameters such as crimping force and mold temperature. Usually, it relies on experience or simple mathematical model for rough estimation, which leads to the problem of excessive deformation error in the production process. On the other hand, the existing technology lacks efficient and automated abnormal detection and processing mechanism. When the production line appears abnormal, the existing monitoring and management mechanism cannot quickly locate the cause and provide effective solutions, usually requiring manual intervention, which not only consumes time but also is prone to errors. SUMMARY

[0004] In view of the above existing problems, the present application is proposed.

[0005] Therefore, the present application provides a wire harness production line intelligent monitoring management method to solve the problems of insufficient process parameter optimization and untimely abnormal processing.

[0006] To solve the above technical problems, the present application provides the following technical solutions:

[0007] In a first aspect, the present application provides a wire harness production line intelligent monitoring management method, which comprises,

[0008] Extracting the data of crimping force and mold temperature from historical wire harness production data, encoding it into a real number encoding gene chain, and generating an initial gene library combined with material properties;

[0009] Building a digital twin model, simulating the stress field distribution and heat conduction effect in the wire harness crimping process through finite element simulation, and obtaining the twin prediction value;

[0010] Based on the digital twin model, pre-validate the process parameters in the initial gene library, select the process parameter combination with deformation error less than the error threshold, and issue it to the crimping machine through the production line control hub to obtain the real-time wire harness deformation data;

[0011] The data of the real-time wire harness deformation variable is compared with the twin prediction value frame by frame to generate a deformation error rate, and an abnormality traceability report is generated based on the deformation error rate;

[0012] Abnormal parameters are obtained based on the abnormality traceability report, and mutation operations of bit flipping and block swapping are performed, while combining with workshop environment data to dynamically compensate for the deviation of the abnormal parameters, to generate primary optimization parameters;

[0013] The primary optimization parameters are input into the production line control center to monitor and dynamically adjust the crimping force to obtain final optimization parameters, while integrating wire harness production data to generate a monitoring management report.

[0014] As a preferred scheme of the wire harness production line intelligent monitoring management method, the initial gene library is obtained as follows,

[0015] PCA is used to extract key thermal distribution characteristics from the mold temperature, and the process characteristics of the dynamic change rate and overshoot of the crimping force are combined to construct a real number coding gene chain.

[0016] The real number coding gene chain is fused with the properties of the yield strength and thermal expansion coefficient of the material, and constraint rules are obtained based on historical work order data, combined with an environmental dynamic compensation coefficient, and an evolutionary algorithm is used to generate an initial gene library.

[0017] As a preferred scheme of the wire harness production line intelligent monitoring management method, the twin prediction value is obtained as follows,

[0018] Based on industrial CT scanning and reverse engineering, the crimping mold and the wire harness terminal are three-dimensionally geometrically reconstructed, and the spatial position and posture of the mold are calibrated by a laser tracker;

[0019] According to the spatial position and posture of the mold, a multi-level parameterized model integrating material properties is established, and real-time control signals of the production line control center are integrated as dynamic boundary conditions to construct a digital twin model;

[0020] Based on the digital twin model, a nonlinear finite element solver is used for multi-physical field coupling transient simulation to simulate the stress field distribution in the crimping process;

[0021] According to the simulated stress field distribution, the thermal conduction effect between the mold and the wire harness is simulated by combining a thermal-mechanical coupling algorithm, and an adaptive mesh refinement method is used to dynamically track the plastic deformation characteristics of the high strain area to generate the twin prediction value.

[0022] As a preferred scheme of the wire harness production line intelligent monitoring management method, the data of the real-time wire harness deformation variable is obtained as follows,

[0023] The digital twin model extracts process parameters in the initial gene pool through an industrial communication protocol, maps to a nonlinear finite element solver, and performs simulation verification in combination with a Johnson-Cook plasticity model and a thermal force coupling algorithm.

[0024] After the simulation verification is completed, the deformation error is obtained by using a DTW algorithm, the deformation error is compared with an error threshold, and the process parameter combination that meets the deformation error is selected based on a dynamic programming algorithm.

[0025] The process parameter combination is transmitted to a production line control center, the control parameters of the servo crimping equipment are dynamically updated by the production line control center, and the data of the real-time wire harness deformation variable collected by the laser displacement meter is synchronously triggered.

[0026] As a preferred scheme of the wire harness production line intelligent monitoring management method, the abnormality traceability report is obtained as follows,

[0027] The real-time wire harness deformation data and the time sequence of the twin prediction value are aligned, the deformation absolute error and the deformation relative error rate are calculated frame by frame, and the error rate is generated by weighted fusion;

[0028] Based on a random forest classification model, the error rate is analyzed, the abnormal type is identified, and an abnormality processing suggestion is generated through EHM current fluctuation analysis and thermal expansion compensation analysis;

[0029] The abnormal type and the abnormality processing suggestion are integrated to generate an abnormality traceability report.

[0030] As a preferred scheme of the wire harness production line intelligent monitoring management method, the initial optimization parameter is obtained as follows,

[0031] The abnormal type in the abnormality traceability report is analyzed to obtain an abnormal parameter;

[0032] Based on the abnormal parameter, bit flip mutation is performed, a flip position is randomly selected in the abnormal parameter sequence, and a probability flip value state is preset;

[0033] After the bit flip mutation is completed, block exchange mutation is performed, the abnormal parameter is divided into functional blocks according to the process logic, and the functional blocks are exchanged and recombined between the same type of functional blocks in different production lines;

[0034] The abnormal parameter subjected to bit flip and block exchange is input into a dynamic compensation mechanism, the workshop temperature fluctuation is monitored in real time, the crimping force value is dynamically adjusted according to the temperature fluctuation through thermal expansion compensation, the deviation of the abnormal parameter is compensated, and the initial optimization parameter is generated.

[0035] As a preferred scheme of the wire harness production line intelligent monitoring management method, the final optimization parameter is obtained as follows,

[0036] The initial optimization parameter is input into the production line control center, collaborative control is carried out based on the reinforcement learning model, deformation deviation of the crimping force is dynamically compensated, and after SPC monitoring and anti-saturation safety protection, the final optimization parameter is generated through wavelet denoising.

[0037] In a second aspect, the present application provides an intelligent monitoring and management system for a wiring harness production line, comprising a database construction module, a model simulation module, a verification module, an error analysis module, an optimization module, and a monitoring and adjustment module,

[0038] The database construction module is used to extract data of crimping force and mold temperature from historical wiring harness production data, encode the data into real number encoding gene chains, and generate an initial gene library in combination with material properties;

[0039] The model simulation module is used to construct a digital twin model, simulate stress field distribution and heat conduction effects in the wiring harness crimping process through finite element simulation, and obtain twin prediction values;

[0040] The verification module is used to pre-verify process parameters in the initial gene library based on the digital twin model, screen process parameter combinations with deformation errors less than an error threshold, and issue the process parameter combinations to a crimping machine through a production line control hub to obtain real-time wiring harness deformation data;

[0041] The error analysis module is used to compare the real-time wiring harness deformation data with the twin prediction values frame by frame, generate a deformation error rate, and generate an abnormality traceability report based on the deformation error rate;

[0042] The optimization module is used to obtain abnormal parameters based on the abnormality traceability report, perform mutation operations of bit flipping and block swapping, dynamically compensate deviations of the abnormal parameters in combination with workshop environment data, and generate initial optimization parameters;

[0043] The monitoring and adjustment module is used to input the initial optimization parameters into the production line control hub, monitor and dynamically adjust the crimping force to obtain final optimization parameters, and integrate wiring harness production data to generate a monitoring and management report.

[0044] In a third aspect, the present application provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, and wherein the computer program, when executed by the processor, implements any step of the intelligent monitoring and management method for a wiring harness production line according to the first aspect of the present application.

[0045] In a fourth aspect, the present application provides a computer readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements any step of the intelligent monitoring and management method for a wiring harness production line according to the first aspect of the present application.

[0046] The application has the advantages that: by constructing an initial gene library and combining with digital twin model for pre-validation, the optimization of key process parameters such as crimping force and mold temperature is more accurate and effective, and various factors in the production process, including material properties and environmental changes, can be more comprehensively considered, thereby reducing deformation errors and improving product consistency and reliability. Secondly, the random forest classification model is used to analyze the real-time wire harness deformation data and twin prediction values, automatically generate abnormal traceability reports and provide processing suggestions, which not only speeds up the identification speed of abnormal conditions, but also realizes intelligent management, reduces the dependence on manual intervention, and improves the response speed and stability of the production line. BRIEF DESCRIPTION OF DRAWINGS

[0047] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings needed in the embodiment description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0048] Fig. 1 The flowchart of the wire harness production line intelligent monitoring management method in embodiment 1.

[0049] Fig. 2 The schematic diagram of the wire harness production line intelligent monitoring management system in embodiment 1.

[0050] Fig. 3 The schematic diagram of the initial optimization parameter acquisition process in embodiment 1.

[0051] Fig. 4 The schematic diagram of the wire harness production line intelligent monitoring management method in embodiment 1. DETAILED DESCRIPTION

[0052] In order to make the above-mentioned purposes, features and advantages of the present application more obvious and easy to understand, the specific embodiments of the present application will be described in detail below with reference to the drawings.

[0053] In the following description, many specific details are set forth in order to provide a thorough understanding of the present application, but the present application can also be implemented in other ways different from those described herein, and those skilled in the art can make similar generalizations without departing from the connotation of the present application, therefore the present application is not limited to the specific embodiments disclosed below.

[0054] Secondly, the "one embodiment" or "embodiment" referred to herein means that the specific features, structures or characteristics can be included in at least one implementation of the present application. "In one embodiment" appearing in different places in the specification does not mean the same embodiment, nor is it an independent or alternative embodiment that excludes other embodiments.

[0055] Embodiment 1, Reference Figs. 1-4 As a first embodiment of the present application, the embodiment provides an intelligent monitoring and management method for a wiring harness production line, comprising the following steps:

[0056] S1, extract the crimping force and mold temperature data from the historical wiring harness production data, encode it into a real number encoding gene chain, and generate an initial gene library by combining material properties.

[0057] Specifically, the following steps are included,

[0058] Based on historical wiring harness production data, precise positioning is achieved through SQL query, and time series data containing crimping force and mold temperature is selected. Key sample points are extracted based on a set time interval, such as the pressure value change of each crimping operation and the key node temperature of the heating and cooling stages of the mold temperature. Pressure values and mold temperature data usually exist in the form of time series, recording the changes of process parameters in different production stages;

[0059] The extracted crimping force and mold temperature data are preprocessed. First, data cleaning is performed, and missing values and outliers are identified and processed using statistical methods or machine learning algorithms such as Z-score and Isolation Forest. Then, the data is normalized by min-max scaling or standardization method to ensure the consistency of the scales between different features. Finally, the continuous time series data is segmented according to the complete production cycle of the wiring harness to better capture the factors affecting the production quality of the wiring harness;

[0060] Before constructing the real number encoding gene chain, PCA is used to extract key thermal distribution features from the mold temperature data. PCA is a dimensionality reduction technique that can identify several dimensions that have a significant impact on wiring harness production, thereby reducing the complexity of subsequent calculations. At the same time, considering the dynamic change rate and overshoot of the crimping force and other process characteristics, the problems that may occur during the crimping process can be better understood. According to the key thermal distribution of the mold, the dynamic change rate and overshoot of the crimping force, a real number encoding gene chain is constructed, containing multi-dimensional features, as the basis for further optimization;

[0061] After the construction of the real number encoding gene chain is completed, the yield strength and thermal expansion coefficient of the material are determined. Material samples are selected and tensile testing is performed using a material testing machine. The stress-strain curve is recorded, and the stress value at which the material begins to deform plastically is determined as the yield strength. At the same time, a thermal mechanical analyzer (TMA) is used to measure the dimensional change of the material sample under controlled temperature changes. The thermal expansion coefficient is determined by calculating the length change rate of the material at different temperatures;

[0062] The real number coding gene chain is combined with the yield strength and thermal expansion coefficient of the material, which ensures that the selected material meets the requirements of the production process under specific conditions, and then constraint rules are obtained based on historical work order data, which not only includes specific operation parameters of the production line, but also includes problems encountered in previous production and solutions, and reasonable constraint conditions are formulated based on this, in addition, considering the changes of workshop environment temperature and humidity, the real number coding gene chain is adjusted by using environmental dynamic compensation coefficient, the environmental dynamic compensation coefficient is an adjustment factor calculated according to the actual temperature and humidity changes of the workshop, in the specific operation, based on the historical temperature, humidity data and process parameters of the workshop, an environmental adaptive dynamic compensation model is constructed to describe the influence of temperature and humidity changes on the behavior of materials, then the environmental adaptive dynamic compensation model is calibrated using historical wire harness production data to ensure its accuracy and reliability, then neural network is used to identify the most relevant features and quantify the influence of these features on process parameters, and specific compensation coefficient values are output to correct the process parameters in the real number coding gene chain to adapt to different environmental conditions, ensuring that the production process parameters can be dynamically adjusted according to the real-time environmental changes, thereby ensuring the stability of the production process and the consistency of the product quality, finally, evolutionary algorithm is used to simulate the natural selection process, and the real number coding gene chain is gradually optimized through iterative selection, crossover and mutation operations, and finally the initial gene library is formed;

[0063] Through the above operations, the initial gene library of wire harness production data is constructed, realizing the deep integration of modern manufacturing technology and information technology, and greatly improving the intelligent level of wire harness production.

[0064] S2, construct a digital twin model, simulate the stress field distribution and heat conduction effect in the wire harness crimping process through finite element simulation, and obtain the twin prediction value.

[0065] Specifically includes the following operations,

[0066] In order to reconstruct the three-dimensional geometry of the crimping die and the wire harness terminal, obtain the three-dimensional model, adopt the industrial CT scanning technology to obtain the high-precision geometric shape, density distribution and material characteristics of the crimping die and the wire harness terminal, and convert them into computer aided design (CAD) model through reverse engineering method, this process not only requires high-resolution scanning equipment, but also needs professional software tools to process scanning data, to ensure that the generated three-dimensional model can truly reflect the geometric characteristics of the crimping die and the wire harness terminal, then use laser tracker to accurately calibrate the spatial position and attitude of the die, this step is crucial for subsequent construction of digital twin model, because any slight positional deviation may lead to inaccurate simulation results, laser tracker determines the position of the crimping die by emitting laser beams to the target point and measuring the time difference of reflection, its accuracy can reach sub-millimeter level, which can effectively avoid the generation of error data;

[0067] After obtaining the three-dimensional model of the crimping die and the wire harness terminal, according to the spatial position and attitude of the die, a multi-level parameterized model integrating material properties is established, multi-level means not only considering the physical properties of the die itself, such as stiffness, density, etc., but also considering the interaction between different materials and the response of the entire crimping production equipment under different working conditions, for example, in the wire harness production line, there may be different thermal expansion coefficients between metal wires and plastic sheath, these differences need to be fully considered in the digital twin model, in addition, integrating the real-time control signals of the production line control center as dynamic boundary conditions is also one of the key steps in building the digital twin model, real-time control signals are instructions or data continuously issued by the production line control center during the operation of the crimping production equipment, such as pressure, speed, etc. Parameters are used to dynamically adjust the state of the crimping production equipment and synchronize to the digital twin model for simulation, finally realize data interaction and real-time update function through MATLAB, complete the construction of digital twin model;

[0068] Based on the digital twin model, use nonlinear finite element solver (such as ANSYS or Abaqus) for multi-physical field coupling transient simulation, nonlinear finite element solver can consider the interaction between force, heat and other physical phenomena at the same time, so as to more accurately simulate the stress field distribution in the crimping process, in specific operation, first set the simulation parameters, including material properties, boundary conditions, initial conditions, etc., then run the nonlinear finite element solver to perform calculation, since the crimping process is a highly nonlinear process involving plastic deformation, contact friction and other complex mechanisms, therefore, it is particularly important to choose appropriate solution strategy, for example, implicit integration scheme can be selected to improve stability, or explicit integration scheme can be used to speed up calculation;

[0069] After the simulation of stress field distribution is completed, further combined with the thermal coupling algorithm to simulate the heat conduction effect between the mold and the wire harness, the heat conduction effect can analyze how heat is transferred between different components, and how this transfer affects the overall performance of the crimping production equipment, in order to better capture the behavior characteristics of local high temperature areas or high strain areas, the adaptive mesh refinement technique is used to dynamically adjust the grid density, the adaptive mesh refinement method can significantly improve the resolution of local details while maintaining computational efficiency, for example, automatically encrypting the grid in the high strain area, so that the physical phenomena in this area can be described more finely;

[0070] Integrate the simulated stress field distribution data and the heat conduction effect data between the mold and the wire harness, the integrated data not only contains the pressure values and temperature changes at different positions, but also reflects the specific mode of material deformation and heat transfer, at the same time, in order to ensure accuracy, the integrated data is error corrected using the root mean square error, to ensure that the integrated data is as close to the actual situation as possible, the error corrected data is comprehensively analyzed to extract key performance indicators, which not only include stress distribution and temperature change trend, but also may cover other important information such as plastic deformation degree, contact area change, etc., by analyzing the key performance indicators, you can fully understand the various physical phenomena and their interaction mechanisms in the crimping process, at the same time, all key performance indicators are summarized to form twin prediction values, which not only provide a scientific basis for subsequent quality control, but also can be used to optimize production process parameters, improve product quality and production efficiency, for example, by analyzing potential problem points shown in the twin prediction values, measures can be taken in advance to avoid defects, thereby reducing scrap rate and shortening development cycle;

[0071] Through the above operation, not only can greatly improve the accuracy and efficiency of wire harness production, but also can effectively reduce the trial and error cost, and lay a solid foundation for realizing intelligent wire harness manufacturing.

[0072] S3, based on the digital twin model, pre-verify the process parameters in the initial gene library, screen the process parameter combination with deformation error less than the error threshold value, and issue to the crimping machine through the production line control center to obtain the real-time wire harness deformation data.

[0073] Specifically includes the following steps,

[0074] The digital twin model extracts process parameters from the initial gene library, including crimping force, mold temperature and other key performance indicators, in order to apply process parameters to actual simulation, it needs to be mapped to the nonlinear finite element solver through industrial communication protocol (such as PROFINET or Modbus);

[0075] After completing the process parameter mapping, start the nonlinear finite element solver to perform simulation verification, here Johnson-Cook plasticity model is used in combination with thermal-mechanical coupling algorithm for simulation verification, Johnson-Cook model is a widely used constitutive equation for metal plasticity deformation analysis, which considers the influence of strain rate, temperature and other factors on material behavior, in terms of operation, first define material properties and boundary conditions in finite element software, then set up the simulation environment according to the extracted process parameters, for example, in Abaqus, you can import process parameters and configure simulation scenarios to ensure that each simulation accurately reflects the state under actual production conditions, the core of simulation verification is to use advanced numerical methods to simulate various physical phenomena that may occur during wire harness crimping process, and evaluate their impact on the final wire harness performance;

[0076] After simulation verification is completed, DTW (Dynamic Time Warping) algorithm is used to calculate deformation error, DTW is a technique for measuring the similarity between two time series, especially when the length of the time series of the simulation verification results and the real-time wire harness deformation data is not consistent, DTW can effectively align the two sequences and calculate their similarity, thus quantifying the deformation error, at the same time, based on the maximum deformation error value allowed by production quality standards and process requirements, define the error threshold, and compare the deformation error with the error threshold, filter out the process parameter combination whose deformation error is less than the error threshold, that is, the process parameter combination that meets the standard, the filtering process usually uses dynamic programming algorithm to optimize the selection process, to ensure that the selected process parameter combination can maximize production efficiency while ensuring quality;

[0077] After obtaining the process parameter combination that meets the requirements, it is transmitted to the production line control center through the industrial communication protocol to ensure the safety and real-time performance of data transmission, after the production line control center receives the process parameters, it dynamically updates the control parameters of the servo crimping equipment, adjusts the operating conditions on the actual production line to match the simulation verification results, at the same time, trigger the laser displacement meter to collect real-time wire harness deformation data, laser displacement meter is a high-precision sensor that can provide sub-millimeter level measurement accuracy, which is crucial for capturing subtle deformation changes, for example, multiple laser displacement meters can be installed at the crimping head, mold contact and wire harness entry and exit area of the crimping machine to monitor the deformation of the wire harness in real time, and feed back the real-time wire harness deformation data to the production line control center;

[0078] The acquired real-time wire bundle deformation variable data is analyzed to further verify the accuracy of the digital twin model and identify potential improvements. For example, if a significant deviation is found between the simulation verification results and the real-time wire bundle deformation variable data under certain conditions, the digital twin model parameters can be adjusted for optimization. In addition, machine learning algorithms can be used to deeply mine historical wire bundle deformation variable data to find potential patterns and rules, providing references for future process optimization. For example, a neural network model can be trained on a large amount of historical wire bundle deformation variable data to identify the main factors affecting deformation error, and process parameters can be adjusted accordingly.

[0079] Through the above operations, not only can the quality and efficiency of wire bundle production be significantly improved, but also a solid foundation can be provided for subsequent quality control and process optimization.

[0080] S4, compare the real-time wire bundle deformation variable data with the twin prediction value frame by frame, generate a deformation error rate, and generate an abnormality traceability report based on the deformation error rate.

[0081] Specifically includes the following steps,

[0082] Due to interference factors in the actual production environment, the DTW algorithm is used to align the real-time wire bundle deformation data with the twin prediction value. After alignment, the deformation absolute error and deformation relative error rate are calculated frame by frame. The deformation absolute error refers to the difference between the real-time wire bundle deformation data and the twin prediction value, while the deformation relative error rate is the proportion of the deformation absolute error relative to the twin prediction value. In order to more comprehensively evaluate the error situation, a weighted fusion method is adopted to assign different weights to the importance of different positions. Finally, a comprehensive error rate is generated. For example, errors in critical stress concentration areas or temperature sensitive areas may be assigned higher weights to highlight the impact of these areas on overall quality. The specific formula is as follows:

[0083]

[0084] Where E represents the comprehensive error rate, N represents the total number of wire bundle deformation data, i represents the index of wire bundle deformation data, w(i) represents the weight of the i-th wire bundle deformation data, a represents the adjustment coefficient of absolute error and relative error rate, E1(i) represents the absolute error of the i-th wire bundle deformation data, and E2(i) represents the relative error rate of the i-th wire bundle deformation data.

[0085] The comprehensive error rate is analyzed using a random forest classification model based on machine learning algorithms. In the specific operation, first define the training data set, including historical error rates and their corresponding labels, such as normal, slight abnormal, severe abnormal, etc., then use the Scikit-learn library in Python to train the random forest model, after training, use the random forest classification model to classify the new error rate data, identify whether there is an abnormal type in the current production process, the random forest classification model is a powerful ensemble learning method, suitable for processing high-dimensional data and has good generalization ability, by building multiple decision trees and voting on the classification results to make predictions, thereby improving the stability and accuracy of the random forest classification model, in order to improve the accuracy of classification, feature engineering is also needed to extract more features that contribute to classification, such as statistical features of mean and variance, frequency domain features after Fourier transform, and current fluctuation and temperature change features, which can help the random forest classification model better understand the data pattern and improve the classification effect;

[0086] In addition to the random forest model, other analysis methods can be combined to further refine the identification of abnormal types, such as EHM (Electrical Health Monitoring) current fluctuation analysis, which can help detect whether the crimping production equipment is running normally, by monitoring the trend of current changes to find potential electrical faults, thermal expansion compensation analysis is used to evaluate the impact of temperature changes on the behavior of the wire harness material, especially when working in high temperature environments, EHM current fluctuation analysis and thermal expansion compensation analysis not only provide more diagnostic information, but also provide a basis for subsequent abnormal processing;

[0087] Based on the identified abnormal types and diagnostic information provided by EHM current fluctuation analysis and thermal expansion compensation analysis, targeted processing schemes can be developed, such as when a large deformation error is found in a section of the wire harness accompanied by current fluctuation abnormalities, it may indicate that the crimping force setting is improper or the die is worn out, at this time it is recommended to adjust the crimping force parameters or replace the die parts, similarly, for deformation errors caused by temperature changes, thermal expansion compensation measures can be taken, such as optimizing cooling or adjusting process temperature, in order to ensure the effectiveness and operability of the recommendations, usually need to be customized according to the specific conditions and limitations of the production line, in addition, automated tools can be used to help quickly generate and implement these processing recommendations, improve the response speed and efficiency of the production line control center, for example, Python can be used to automatically generate optimization recommendations, and directly issued to the production line control center through industrial communication protocols to realize automatic adjustment;

[0088] Integrate the relevant information of abnormal types and processing suggestions to generate a detailed abnormal traceability report, which not only contains basic information such as abnormal type, location, timestamp, but also lists the processing suggestions and their expected effects for each abnormality. To enhance readability and interactivity, the abnormal traceability report is presented in a structured format with tables and charts, and interactive dashboards are created using Tableau and Power BI, allowing users to dynamically view the abnormal traceability report, helping management quickly locate abnormalities, promoting cross-departmental collaboration, and improving overall work efficiency;

[0089] Through the above operations, the efficiency of abnormal handling in the wire harness production process is significantly improved, and cross-departmental collaboration is promoted, improving overall work efficiency and product consistency.

[0090] S5, based on the abnormal traceability report, obtain the abnormal parameters, and perform bit flipping and block swapping mutation operations, and combine with the workshop environment data to dynamically compensate the deviation of the abnormal parameters to generate the initial optimization parameters.

[0091] Specifically includes the following steps,

[0092] To optimize the abnormal parameters in the wire harness crimping process, first analyze the generated abnormal traceability report in depth to identify the specific abnormal types and their related abnormal parameters. For example, if the abnormal traceability report indicates that a certain section of wire harness has a large deformation error at a specific time point, the corresponding crimping force or mold temperature may be the abnormal parameter that needs to be adjusted. By analyzing the detailed information in the report, such as the timestamp and location of the abnormality, the parameter causing the abnormality can be accurately located. Once the abnormal parameter is determined, it is extracted and cleaned as necessary to ensure data consistency and comparability.

[0093] After obtaining the pre-processed abnormal parameters, perform bit flipping mutation operations. The core of the bit flipping mutation operation is to randomly select one or more positions in the abnormal parameter sequence and flip the value state at these positions with a certain preset probability. For example, for binary-coded abnormal parameters, 0 becomes 1 and 1 becomes 0. For real number coding, a small perturbation value can be added or subtracted to simulate the flipping effect. In specific operations, Python can be used to determine the flipping position and probability using a random number generator (such as random), and the new parameters after bit flipping mutation are verified for their reasonableness. For example, the digital twin model can be called again for simulation verification to evaluate the impact of the new parameters on the production process, and finite element analysis software (such as Abaqus or ANSYS) can be used to simulate the deformation of the wire harness under the new crimping force setting to confirm whether the expected effect has been achieved.

[0094] On the basis of bit flip mutation, further perform block exchange mutation operation, first define process logic based on different process flow of production line, divide abnormal parameters into different functional blocks according to process logic, for example, in the crimping process, functional blocks can be defined according to different areas of the mold and different stages of the process, the purpose of this is to better simulate the change mode in actual production, make the mutation operation more in line with the actual situation, complete the functional block division, and exchange and recombine between the same functional blocks of different production lines, this step aims to increase the exploration space through more diversified parameter combinations, which helps to find better solutions, for example, a set of process parameters can be selected from the functional blocks of one production line to replace the process parameters in the same functional blocks of another production line, in the specific operation, block exchange operation can be automatically performed by using Python, and the results of each exchange are recorded for subsequent analysis;

[0095] The abnormal parameters after bit flip and block exchange are input into the dynamic compensation mechanism, and through the industrial internet of things (IIoT) technology, the temperature sensor is connected to the production line control hub to monitor the temperature data in the workshop in real time and transmit it to the production line control hub to obtain temperature fluctuation data;

[0096] Based on the temperature fluctuation data, a thermal expansion compensation algorithm is used to dynamically adjust the crimping force value. Thermal expansion effect will cause the size of the material to change, thereby affecting the crimping quality. By analyzing the temperature fluctuation, the expansion and contraction degree of the material is determined, and the crimping force is adjusted accordingly to compensate for the deviation of abnormal parameters. For example, based on the production process requirements and material characteristics, a standard temperature value is defined. When the temperature in the workshop is higher than the standard temperature value, the crimping force is reduced to offset the excessive deformation of the material. When the temperature is lower than the standard temperature value, the crimping force is increased to compensate for the material shrinkage effect;

[0097] The parameters processed by the dynamic compensation mechanism are input into the crimping production equipment for verification. By comparing the wire harness production data before and after input, the effect of parameter optimization is evaluated. If the optimization effect meets the production expectation, the initial optimization is considered successful. Otherwise, the bit flip, block exchange and temperature compensation operations need to be repeated until the best effect is achieved to generate the initial optimization parameters;

[0098] Through the above operations, not only the quality and efficiency of wire harness production are significantly improved, but also a solid foundation is provided for subsequent quality control and process optimization.

[0099] S6, input the initial optimization parameters into the production line control hub, monitor and dynamically adjust the crimping force to obtain the final optimization parameters, and integrate the wire harness production data to generate a monitoring and management report.

[0100] Specifically includes the following steps,

[0101] First, the initial optimization parameters are input into the production line control center. In specific operations, the initial optimization parameters can be transmitted to the production line control center through the PROFINET industrial communication protocol. This automatic transmission process not only improves work efficiency, but also reduces the possibility of human error;

[0102] Once the initial optimization parameters are input into the production line control center, the next step is to monitor and dynamically adjust the crimping force in real time. Continuous collection of crimping force and temperature data provided by pressure sensors and temperature sensors, and dynamic adjustment of crimping force based on crimping force and temperature data ensure that high-quality wire harness crimping effects can be maintained even in the case of changing environmental conditions. The dynamic adjustment step usually uses a PID controller to achieve closed-loop control, and the PID controller calculates the amount of adjustment needed to maintain the stability and accuracy of wire harness production;

[0103] In order to further optimize the initial optimization parameters, a reinforcement learning model is used for collaborative control. First, a reinforcement learning model is constructed, which usually consists of a state space, an action space, and a reward function. The state space includes current process parameters such as crimping force and temperature, and the action space is the adjustable control variable such as increasing or decreasing the crimping force. The reward function defines the goal, such as minimizing the deformation error. The reinforcement learning algorithm used is Q-learning, which can use the TensorFlow framework in Python to implement reinforcement learning model construction and training. During training, the reinforcement learning model simulates different scenarios and tries various adjustment strategies to gradually learn how to make the best decisions to achieve the goal. The trained reinforcement learning model is deployed in the production line control center, which receives real-time crimping force and temperature data from pressure sensors and temperature sensors as state input and outputs corresponding action suggestions such as adjusting the crimping force;

[0104] In specific operations, the reinforcement learning model will predict the best action based on real-time wire harness production data. For example, when detecting an increase in deformation error, the reinforcement learning model will recommend reducing the crimping force to reduce the deformation error. This dynamic adjustment mechanism improves the stability and efficiency of the production process. In addition, through continuous trial and error and learning, the reinforcement learning model can gradually find the optimal control strategy, making the production process more stable and efficient. For example, if the reinforcement learning model predicts that a certain adjustment may reduce the deformation error, it will recommend that the production line control center make the corresponding adjustment, which not only improves production efficiency, but also enhances adaptability and flexibility;

[0105] After the dynamic compensation of the deformation deviation of the crimping force by the reinforcement learning model, the statistical process control (SPC) technology is used to monitor the process parameters. SPC tracks the trend of process parameters through control charts to detect abnormal fluctuations in a timely manner. For example, control charts can be drawn using relevant libraries in Python to analyze whether the process parameters exceed the control limits. If abnormalities are found, the production line control hub will issue an alert to take timely measures. Through real-time monitoring and early warning, potential problems can be detected and corrected quickly to prevent further expansion of the problem. In addition, early warning thresholds can be set based on the safe operating range and quality standards of production to provide early warnings when process parameters approach the control limits, allowing for early preparation. To avoid process parameters exceeding their physical and design limits, an anti-saturation safety protection mechanism is set, which typically involves setting upper and lower threshold values based on the quality standards and safe operating range of production equipment and processes. When process parameters approach or exceed the upper and lower threshold values, the production line is automatically suspended to prevent equipment damage or product quality degradation.

[0106] After SPC monitoring and anti-saturation safety protection, the process parameters are preprocessed to remove noise interference. Wavelet transform is an effective signal processing technology that can be used for noise reduction without losing important information. In specific operations, wavelet decomposition and reconstruction can be performed using Python. First, select an appropriate wavelet basis function, then perform multi-layer decomposition on the process parameters, extract the main features of the low-frequency components, and remove the noise of the high-frequency components. Finally, reconstruction is performed to improve the accuracy of subsequent analysis. The parameters after wavelet noise reduction are considered as final optimized parameters with higher precision and stability, suitable for actual production environment. These final optimized parameters can be saved as a JSON configuration file for reference in subsequent batch production to guide subsequent production activities.

[0107] During the acquisition of final optimized parameters, the production line control hub continuously collects various harness production data and integrates the data of final optimized parameters to generate detailed monitoring management reports, including not only final optimized parameters but also abnormal information in the production process, such as abnormal type, occurrence location, timestamp, etc. In addition, PDF or HTML format reports can be automatically generated for easy reference.

[0108] Through the above operations, monitoring management reports are generated to provide a comprehensive perspective for harness production, helping production and management personnel better understand and optimize the harness production process to ensure that each link is in the best state.

[0109] The embodiment also provides a harness production line intelligent monitoring management system, which includes a database construction module, a model simulation module, a verification module, an error analysis module, an optimization module, a monitoring and adjustment module,

[0110] A database construction module is configured to extract data of crimping force and mold temperature from historical harness production data, encode the data into real number encoding gene chains, and generate an initial gene library in combination with material properties;

[0111] A model simulation module is configured to construct a digital twin model, simulate stress field distribution and heat conduction effects in the crimping process of the harness through finite element simulation, and obtain twin prediction values;

[0112] A verification module is configured to perform pre-verification on process parameters in the initial gene library based on the digital twin model, screen process parameter combinations with deformation error less than an error threshold, and issue the process parameter combinations to a crimping machine through a production line control hub to obtain real-time harness deformation data;

[0113] An error analysis module is configured to compare the real-time harness deformation data with the twin prediction values frame by frame, generate a deformation error rate, and generate an abnormality traceability report based on the deformation error rate;

[0114] An optimization module is configured to obtain abnormal parameters based on the abnormality traceability report, perform mutation operations of bit flipping and block swapping, dynamically compensate deviations of the abnormal parameters in combination with workshop environment data, and generate primary optimization parameters;

[0115] A monitoring and adjustment module is configured to input the primary optimization parameters into the production line control hub, monitor and dynamically adjust the crimping force to obtain final optimization parameters, and generate a monitoring and management report in combination with the harness production data.

[0116] The embodiment also provides a computer device suitable for the harness production line intelligent monitoring and management method, which includes a memory and a processor.

[0117] The computer device can be a terminal, and the computer device includes a processor, a memory, a communication interface, a display screen and an input device connected by a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with external terminals in a wired or wireless manner. The wireless manner can be achieved by WIFI, an operator network, NFC (near field communication) or other technologies. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen. The input device of the computer device can be a touch layer overlaid on the display screen, or a key, trackball or touchpad arranged on the shell of the computer device, or an external keyboard, touchpad or mouse, etc.

[0118] The embodiment also provides a storage medium having a computer program stored thereon, the program being executed by a processor to implement the method for intelligently monitoring and managing a wire harness production line as described above. The storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as a static random access memory (SRAM), an electrically erasable programmable read-only memory (EEPROM), an erasable programmable read-only memory (EPROM), a programmable read-only memory (PROM), a read-only memory (ROM), a magnetic storage device, a flash memory, a magnetic disk or an optical disk.

[0119] To sum up, the present application is more accurate and effective in optimizing key process parameters such as crimping force and mold temperature by constructing an initial gene library and conducting pre-validation in combination with a digital twin model, can more comprehensively consider various factors in the production process, including material properties and environmental changes, thereby reducing deformation errors and improving product consistency and reliability. Secondly, a random forest classification model is used to analyze real-time wire harness deformation data and twin prediction values, automatically generate an abnormality traceability report and provide processing suggestions, not only speeding up the identification speed of abnormal conditions, but also realizing intelligent management, reducing the dependence on manual intervention, and improving the response speed and stability of the production line.

[0120] It should be noted that the above examples are only used to illustrate the technical solutions of the present application but not limit the present application. Although the present application is described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present application can be modified or equivalently replaced, without departing from the spirit and scope of the technical solutions of the present application, which should be covered in the scope of the claims of the present application.

Claims

1. A method for intelligent monitoring and management of a wiring harness production line, characterized in that: Comprising, Extracting the crimping force and die temperature data from historical harness production data, encoding them into real number encoding gene chains and generating an initial gene pool by combining material properties; Building a digital twin model to simulate the stress field distribution and heat conduction effects during the crimping process through finite element simulation, obtaining twin prediction values, as follows, Based on industrial CT scanning and reverse engineering, the crimping die and harness terminals are reconstructed in three-dimensional geometry, and the spatial position and attitude of the die are calibrated by a laser tracker; According to the spatial position and attitude of the die, a multi-level parameterized model integrating material properties is established, and real-time control signals from the production line control center are integrated as dynamic boundary conditions to build a digital twin model; Based on the digital twin model, a nonlinear finite element solver is used for multi-physical field coupling transient simulation to simulate the stress field distribution during the crimping process; According to the simulated stress field distribution, the heat conduction effect between the die and the harness is simulated by combining the thermal-mechanical coupling algorithm, and the adaptive mesh refinement method is used to dynamically track the plastic deformation characteristics in the high strain area to generate twin prediction values; Based on the digital twin model, the process parameters in the initial gene pool are pre-verified, and process parameter combinations with deformation error less than the error threshold are selected, and then transmitted to the crimping machine through the production line control center to obtain real-time harness deformation data; Compare the real-time harness deformation data with the twin prediction values frame by frame to generate a deformation error rate, and generate an abnormality traceability report based on the deformation error rate; Based on the abnormality traceability report, abnormal parameters are obtained, and mutation operations such as bit flipping and block swapping are performed, while combining with workshop environment data to dynamically compensate the deviation of abnormal parameters, generating initial optimization parameters; Input the initial optimization parameters into the production line control center to monitor and dynamically adjust the crimping force to obtain the final optimization parameters, and integrate the harness production data to generate a monitoring and management report.

2. The harness production line intelligent monitoring management method of claim 1, wherein: The initial gene pool is obtained as follows, Extract key thermal distribution features from die temperature using PCA, combine with process features such as dynamic change rate and overshoot of crimping force, and construct real number encoding gene chains; Fuse the real number encoding gene chains with the yield strength and thermal expansion coefficient of the material, obtain constraint rules based on historical work order data, combine with environmental dynamic compensation coefficient, and generate an initial gene pool using evolutionary algorithms.

3. The harness production line intelligent monitoring management method of claim 1, wherein: The real-time harness deformation data is obtained as follows, The digital twin model extracts process parameters from the initial gene pool through industrial communication protocols, maps them to a nonlinear finite element solver, and performs simulation verification using the Johnson-Cook plasticity model and thermal-mechanical coupling algorithm; After simulation verification, use the DTW algorithm to obtain the deformation error, compare it with the error threshold, and select process parameter combinations that meet the deformation error based on the dynamic programming algorithm; The process parameter combinations are transmitted to the production line control center, which dynamically updates the control parameters of the servo crimping equipment, and synchronously triggers the laser displacement meter to collect real-time harness deformation data.

4. The harness production line intelligent monitoring management method of claim 1, wherein: The abnormality traceability report is obtained as follows, Aligning the time series of real-time wire harness deformation data and twin prediction values, calculating deformation absolute error and deformation relative error rate frame by frame, and generating error rate by weighted fusion; Based on the random forest classification model, the error rate is analyzed to identify the type of anomaly, and the EHM current fluctuation analysis and thermal expansion compensation analysis are used to generate abnormal processing suggestions; Integrate the abnormal type and abnormal processing suggestions to generate the abnormal traceability report.

5. The harness production line intelligent monitoring management method of claim 4, wherein: The acquisition process of the initial optimization parameter is as follows, Analyze the abnormal type in the abnormal traceability report to obtain the abnormal parameter; Based on the abnormal parameter, perform bit flip mutation, randomly select the flip position in the abnormal parameter sequence, and preset the probability of flipping the value state; After completing the bit flip mutation, perform block exchange mutation, divide the abnormal parameter into functional blocks according to the process logic, and exchange and recombine between the same type of functional blocks in different production lines; The abnormal parameters after bit flip and block exchange are input into the dynamic compensation mechanism, and the temperature fluctuation in the workshop is monitored in real time. The thermal expansion compensation dynamically adjusts the crimping force value according to the temperature fluctuation to compensate for the deviation of the abnormal parameter, and generates the initial optimization parameter.

6. The harness production line intelligent monitoring management method of claim 1, wherein: The acquisition steps of the final optimization parameter are as follows, The initial optimization parameter is input into the production line control center, the reinforcement learning model is used for collaborative control, the crimping force deformation deviation is dynamically compensated, and the final optimization parameter is generated through wavelet denoising after SPC monitoring and anti-saturation safety protection.

7. A harness production line intelligent monitoring management system based on the harness production line intelligent monitoring management method according to any one of claims 1 to 6, characterized in that: It includes a database construction module, a model simulation module, a verification module, an error analysis module, an optimization module, and a monitoring adjustment module, The database construction module is used to extract the data of crimping force and mold temperature from historical wire harness production data, encode it into a real number coding gene chain, and generate an initial gene library combined with material properties; The model simulation module is used to build a digital twin model to simulate the stress field distribution and heat conduction effect during wire harness crimping through finite element simulation to obtain twin prediction values; The verification module is used to pre-verify the process parameters in the initial gene library based on the digital twin model, filter the process parameter combinations with deformation error less than the error threshold, and send them to the crimping machine through the production line control center to obtain real-time wire harness deformation data; The error analysis module is used to compare the real-time wire harness deformation data with the twin prediction values frame by frame to generate deformation error rate and generate an abnormal traceability report based on the deformation error rate; The optimization module is used to obtain abnormal parameters based on the abnormal traceability report and perform bit flip and block exchange mutation operations, while combining with workshop environment data to dynamically compensate for the deviation of abnormal parameters and generate initial optimization parameters; The monitoring adjustment module is used to input the initial optimization parameter into the production line control center to monitor and dynamically adjust the crimping force to obtain the final optimization parameter, and integrate the wire harness production data to generate a monitoring management report.

8. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that: The processor executes the computer program to realize the steps of the wire harness production line intelligent monitoring management method of any one of claims 1-6.

9. A computer readable storage medium having stored thereon a computer program, characterized in that: The computer program is executed by the processor to realize the steps of the wire harness production line intelligent monitoring management method of any one of claims 1-6.

Images