Automobile precision shaft structure design method based on AI optimization

Abstract

The invention relates to the technical field of automobile manufacturing, in particular to an automobile precision shaft structure design method based on AI optimization, which comprises the steps of data collection and preprocessing, AI model construction and training, design optimization and simulation analysis, design output and manufacturing preparation and continuous optimization. Compared with the prior art which only depends on simulation software to carry out simulation analysis and possibly has the defects of insufficient model accuracy and difficulty in comprehensively reflecting actual conditions, the scheme adopts a method of combining simulation and experiments, and the precision shaft structure simulation model is established while correct parameters are set. An experimental model is manufactured, a sensor is installed to collect data, and the process of simulation and experiment, data comparison and analysis and model optimization and adjustment is repeatedly operated until the simulation result is highly matched with the experimental data, so that the accuracy and reliability of the simulation model are effectively improved; and a firmer theoretical and experimental foundation is provided for the optimization design of the automobile precision shaft structure.

Description

Technical Field

[0001] The present invention relates to the technical field of automobile manufacturing, and particularly to a design method for the precise shaft structure of an automobile optimized based on AI. Background Art

[0002] The existing design of the precise shaft structure of an automobile mainly relies on engineers' experience and traditional analysis methods. The design process is cumbersome, the optimization efficiency is low, and it is difficult to meet the development needs of automobile lightweight and high efficiency. Traditional mechanical design methods often have difficulty accurately capturing the interaction relationship between complex and variable structural parameters and material properties, resulting in design schemes that may have problems such as insufficient performance, shortened lifespan, and high manufacturing costs.

[0003] To solve the above problems, people have begun to explore using artificial intelligence (AI) technology to optimize the design of the precise shaft structure of an automobile. AI algorithms have powerful data processing and recognition capabilities, and can extract key features from a large amount of structural design cases, material performance data, manufacturing process data, and test results, and establish more accurate modeling relationships. Using the AI model, the structural parameters can be optimized automatically, and the performance of different design schemes can be evaluated quickly to obtain better design schemes.

[0004] However, the existing AI-based design methods for the precise shaft structure of an automobile still have some limitations. For example, data collection and preprocessing are relatively difficult, the training data of the model is insufficient, the generalization ability of the model is limited, and the actual manufacturing process is not considered. In addition, traditional simulation analysis methods often rely on means such as finite element analysis, with a large amount of calculation and long time consumption. Therefore, there is an urgent need to develop an efficient, accurate, and scalable design method for the precise shaft structure of an automobile optimized based on AI to meet the rapid development needs of the automobile industry. Summary of the Invention

[0005] To overcome the problems raised in the above background art, the present invention proposes a design method for the precise shaft structure of an automobile optimized based on AI.

[0006] The technical solution of the present invention is as follows: A design method for the precise shaft structure of an automobile optimized based on AI, comprising the following steps:

[0007] S11: Data collection and preprocessing, collecting data related to the design of the precise shaft structure of a vehicle from various sources, and preprocessing the collected data;

[0008] S12: AI model construction and training, extracting the required features from the preprocessed data, constructing a model framework, and training the constructed model using the extracted features;

[0009] S13: Design optimization and simulation analysis, using the AI model to optimize the design parameters of the shaft structure and perform simulation analysis on the optimized precision shaft structure;

[0010] S14: Design output and manufacturing preparation, generating the design drawings and detailed specifications of the optimized shaft structure and generating the manufacturing process flow according to the structural requirements;

[0011] S15: Continuous optimization, after the shaft structure is put into use, collecting its performance monitoring data, optimizing the precision shaft structure, collecting the specific problems of the precision shaft structure, and optimizing and iterating the model according to the problems.

[0012] Preferably, when collecting data related to the design of the precision shaft structure for the vehicle from multiple sources, the specific data content includes historical design cases, material property data, manufacturing process data, performance test results, and actual application data. The data sources include:

[0013] A11: Enterprise internal database, containing the enterprise's past design, manufacturing, and test data;

[0014] A12: Industry database, containing the design, manufacturing, and test data of other enterprises in the industry;

[0015] A13: Professional websites, containing the latest research and applications in materials, manufacturing processes, and performance testing;

[0016] A14: Academic papers, containing in-depth research and experimental data in materials, manufacturing processes, and performance testing;

[0017] A15: Government agencies, containing standards and specifications in materials, manufacturing processes, and performance testing;

[0018] A16: Customer feedback, containing the actual needs of customers and the problems that occur during the actual use of the product.

[0019] Preferably, when collecting data related to the design of the precision shaft structure for the vehicle from multiple sources and preprocessing the collected data, the following steps are included:

[0020] S21: Data collection and integration, collecting the required data through multiple channels, integrating the collected data to form a unified data format and standard, and inspecting the data to correct incorrect data;

[0021] S22: Data cleaning, including removing duplicate data, identifying and handling missing values, and correcting incorrect data;

[0022] S23: Data optimization, including using data smoothing techniques to eliminate data noise, data augmentation, optimizing data formats, and data normalization processing;

[0023] S24: Data arrangement, arranging and integrating the data, and establishing a version library according to the design scheme of the precision shaft structure of the vehicle, and dividing the data into different version libraries.

[0024] Preferably, when preprocessing the collected data, it also includes synthesizing the data, that is, generating simulated data through computer algorithms to simulate the data distribution and characteristics of the real world. Specifically, it includes the following steps:

[0025] S31: Prepare the dataset, collect a real dataset related to the target task for training the GAN model;

[0026] S32: Build the GAN model, build a GAN model including a generator and a discriminator. Among them, the generator is responsible for generating synthetic data, and the discriminator is responsible for distinguishing real data and synthetic data;

[0027] S33: Define the loss function, define the loss function for the GAN model, including the loss of the generator and the loss of the discriminator;

[0028] S34: Train the model, use the real dataset to train the GAN model;

[0029] S35: Generate synthetic data. When the GAN model is trained, use the generator to generate synthetic data and verify and evaluate the generated synthetic data.

[0030] Preferably, when extracting the required features from the preprocessed data, building the model architecture, and training the constructed model using the extracted features, it includes the following steps:

[0031] S41: Feature extraction and selection, extract material properties, geometric dimensions, load conditions, fault types, and other preset feature data from the preprocessed data, and select the extracted feature data through a filter-based feature selection algorithm;

[0032] S42: Build the model, build a deep learning network model, including defining the input layer, hidden layer, and output layer, and determining the number of network layers and the number of nodes in each layer;

[0033] S43: Model training, use the extracted feature data to train the model.

[0034] Preferably, when building the deep learning network model, the loss function of the deep learning network model is:

[0035]

[0036] Among them, y i is the actual result, is the predicted result, n is the number of samples, and n is also the number of input features.

[0037] Preferably, when using the AI model to optimize the design parameters of the shaft structure and performing simulation analysis on the optimized precision shaft structure, the following steps are included:

[0038] S51: Enter the design scheme, and enter the design scheme of the precision shaft structure into the established deep learning network model;

[0039] S52: Optimize the design parameters, and use the trained AI model to optimize the design parameters of the shaft structure, including adjusting the material selection and geometric dimensions;

[0040] S53: Perform simulation analysis, establish a simulation model of the optimized precision shaft structure, and perform operation analysis on the established simulation model;

[0041] S54: Verify the results, design corresponding experimental schemes according to the simulation analysis results to verify the accuracy of the simulation results, and adjust and optimize the model.

[0042] Preferably, when establishing a simulation model of the optimized precision shaft structure and performing operation analysis on the established simulation model, the following steps are included:

[0043] S61: Use simulation software to establish a simulation model of the precision shaft structure, and set correct material properties and boundary conditions for the simulation model;

[0044] S62: At the same time, according to the design of the simulation model, fabricate a corresponding experimental model, and install sensors for collecting key parameter data on the experimental model;

[0045] S63: Run the established precision shaft structure simulation model in the simulation software to obtain corresponding simulation results and data during the operation of the simulation model;

[0046] S64: At the same time, run the experimental model under the same conditions, compare the data of the simulation model with the data of the experimental model, and optimize and adjust the simulation model according to the results of the comparison analysis;

[0047] S65: Repeat the process of running the simulation model, collecting experimental data, comparative analysis, and optimization and adjustment until the simulation results and experimental data reach sufficient consistency.

[0048] Preferably, when generating the design drawings and detailed specifications of the optimized shaft structure and generating the manufacturing process flow according to the structural requirements, the following steps are included:

[0049] S71: Parameter generation, list the specific parameters of the precision shaft structure and convert them into data in a standard format;

[0050] S72: Convert the optimized precision shaft structure into a design drawing in a two-dimensional plane and generate corresponding specifications according to the design specifications;

[0051] S73: Develop a detailed manufacturing process flow based on the design drawing and specifications. The manufacturing process flow includes the processing, heat treatment, surface treatment of the shaft, and the specific operations and process parameters of other key steps.

[0052] Preferably, when developing a detailed manufacturing process flow based on the design drawing and specifications, the following steps are included:

[0053] S81: Process route planning, plan the entire process route from raw materials to finished products according to the design drawing and specifications;

[0054] S82: Process parameter determination, for each process step, determine the specific operating parameters, and the determined parameters are considered based on material properties, production equipment capabilities, and quality control standards;

[0055] S83: Simulation and modeling, conduct manufacturing simulation and modeling, perform simulation according to the existing equipment, and compare the simulated manufactured product with the design drawing of the precision shaft structure;

[0056] S84: According to the comparison results, repeat steps S81 - S83 for iterative optimization until the simulated manufactured product and the design drawing of the precision shaft structure reach a sufficient degree of coincidence.

[0057] Preferably, after the shaft structure is put into use, collect its performance monitoring data, optimize the precision shaft structure, and collect the specific problems of the precision shaft structure. When optimizing and iterating the model according to the problems, the following steps are included:

[0058] S91: Performance monitoring data collection, after the shaft structure is put into use, collect its performance monitoring data, including the life, failure rate, vibration and noise of the shaft, and customer feedback;

[0059] S92: Problem analysis, analyze the collected performance monitoring data and conduct diagnosis to identify the root cause of the problem;

[0060] S93: Model optimization and iteration: According to the results of the problem analysis, optimize and iterate the AI model and optimize the design of the precision shaft structure.

[0061] Advantages of the present invention:

[0062] 1. Compared with the prior art that only relies on simulation software for simulation analysis, which may have the drawbacks of insufficient model accuracy and difficulty in comprehensively reflecting the actual situation, this solution adopts a method combining simulation and experiment. While establishing a simulation model of the precision shaft structure and setting correct parameters, an experimental model is made and sensors are installed to collect data. Through the process of repeatedly running simulation and experiment, comparing and analyzing data, and optimizing and adjusting the model until the simulation results are highly consistent with the experimental data, this solution effectively improves the accuracy and reliability of the simulation model, providing a more solid theoretical and experimental basis for the optimal design of the automotive precision shaft structure;

[0063] 2. Compared with the prior art that relies on manual experience for drawing and process planning, which may have the drawbacks of large design errors, low manufacturing efficiency, and inaccurate process flows, this solution adopts an automated parameter generation and conversion technology to accurately list the optimized precision shaft structure parameters and convert them into a standard data format, and then automatically generate two-dimensional design drawings and detailed specifications. At the same time, according to the design drawings and specification requirements, a manufacturing process flow including key steps such as shaft machining, heat treatment, and surface treatment is accurately formulated. This solution not only greatly improves the design efficiency and accuracy but also ensures the refinement and standardization of the manufacturing process, significantly enhancing the manufacturing quality and production efficiency of the automotive precision shaft structure;

[0064] 3. Compared with the prior art that formulates manufacturing process flows solely based on experience, which may lead to unreasonable process routes, inaccurate parameter settings, and difficulty in guaranteeing product quality, this solution adopts a process route planning based on design drawings and specifications, accurately determines process parameters in combination with material properties, production equipment capabilities, and quality control standards, and through the comparison of manufacturing simulation and design drawings, conducts iterative optimization until the simulated manufactured product is highly consistent with the precision shaft structure design drawings. This solution significantly improves the accuracy and reliability of the manufacturing process, ensures product quality, optimizes the process route, and improves production efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0065] Figure 1 The flowchart showing the design method of the automotive precision shaft structure optimized based on AI of the present invention is presented;

[0066] Figure 2 The flowchart showing the work process of design optimization and simulation analysis in the design method of the automotive precision shaft structure optimized based on AI of the present invention is presented. DETAILED DESCRIPTION OF THE INVENTION

[0067] The present invention will be further described below with reference to the drawings and embodiments.

[0068] Please refer to Figure 1-2, the present invention provides an embodiment: a method for designing the precise shaft structure of an automobile based on AI optimization, including the following steps:

[0069] S11: Data collection and preprocessing, collecting data related to the design of the precise shaft structure of the vehicle from various sources and preprocessing the collected data;

[0070] S12: AI model construction and training, extracting the required features from the preprocessed data, constructing a model framework, and training the constructed model using the extracted features;

[0071] S13: Design optimization and simulation analysis, using the AI model to optimize the design parameters of the shaft structure and performing simulation analysis on the optimized precise shaft structure;

[0072] S14: Design output and manufacturing preparation, generating the design drawings and detailed specifications of the optimized shaft structure and generating the manufacturing process flow according to the structural requirements;

[0073] S15: Continuous optimization, after the shaft structure is put into use, collecting its performance monitoring data, optimizing the precise shaft structure, collecting the specific problems of the precise shaft structure, and optimizing and iterating the model according to the problems.

[0074] As described above, through comprehensive collection and preprocessing of relevant data, precise construction and training of the AI model, intelligent optimization and simulation analysis of design parameters, preparation of detailed design drawings and manufacturing processes, as well as continuous performance monitoring and optimization iteration, the present invention significantly improves the efficiency and accuracy of the design of the precise shaft structure of the automobile, ensures the excellent performance and manufacturing feasibility of the product, and realizes the full-chain intelligent upgrade from design to manufacturing and then to optimization.

[0075] Preferably, when collecting data related to the design of the precise shaft structure of the vehicle from various sources, the specific data content includes historical design cases, material performance data, manufacturing process data, performance test results, and actual application data, and the data sources include:

[0076] A11: The enterprise internal database, containing the enterprise's past design, manufacturing, and test data;

[0077] A12: The industry database, containing the design, manufacturing, and test data of other enterprises in the industry;

[0078] A13: Professional websites, containing the latest research and applications in materials, manufacturing processes, and performance testing;

[0079] A14: Academic papers, containing in-depth research and experimental data in materials, manufacturing processes, and performance testing;

[0080] A15: Government agencies, including standards and specifications regarding materials, manufacturing processes, and performance testing;

[0081] A16: Customer feedback, including the actual needs of customers and the problems that occur during the actual use of the product.

[0082] As described above, the present invention widely collects historical design cases, material property data, manufacturing process data, performance test results, and actual application data from various sources such as the enterprise internal database, industry database, professional websites, academic papers, government agencies, and customer feedback. This not only enriches the data content and diversity but also ensures the comprehensiveness and accuracy of the data, providing a solid data foundation for the subsequent construction and training of the AI model. Thereby, it can more effectively optimize the design of automotive precision shaft structures, improve the performance and quality of products, meet the actual needs of customers, and comply with industry standards and specifications.

[0083] Preferably, when collecting data related to the design of automotive precision shaft structures from various sources and preprocessing the collected data, the following steps are included:

[0084] S21: Data collection and integration. Collect the required data through multiple channels, integrate the collected data to form a unified data format and standard, and inspect the data to correct incorrect data;

[0085] S22: Data cleaning, including removing duplicate data, identifying and handling missing values, and correcting incorrect data;

[0086] S23: Data optimization, including using data smoothing techniques to eliminate data noise, data augmentation, optimizing the data format, and data normalization processing;

[0087] S24: Data arrangement. Arrange and fuse the data, establish a version library according to the design scheme of automotive precision shaft structures, and divide the data into different version libraries.

[0088] As described above, the present invention collects and integrates data related to the design of automotive precision shaft structures through multiple channels to ensure the comprehensiveness and consistency of the data; subsequently, a strict data cleaning process is implemented to effectively remove redundancy, fill in missing values, and correct incorrect data, improving the data quality; further, data optimization means such as smoothing, augmentation, and normalization are adopted to reduce noise and enhance the usability of the data; finally, the data is carefully arranged and incorporated into the version library for easy management and traceability. This series of preprocessing steps lays a solid data foundation for the subsequent construction of the AI model, significantly improving the accuracy and efficiency of the design of automotive precision shaft structures.

[0089] Preferably, when preprocessing the collected data, it also includes synthesizing the data, that is, generating simulated data through computer algorithms to simulate the data distribution and characteristics of the real world. Specifically, it includes the following steps:

[0090] S31: Prepare a dataset and collect a real dataset related to the target task for training the GAN model;

[0091] S32: Build a GAN model, build a GAN model including a generator and a discriminator. Among them, the generator is responsible for generating synthetic data, and the discriminator is responsible for distinguishing real data and synthetic data;

[0092] S33: Define a loss function, define a loss function for the GAN model, including the loss of the generator and the loss of the discriminator;

[0093] S34: Train the model, use the real dataset to train the GAN model;

[0094] S35: Generate synthetic data. After the GAN model is trained, use the generator to generate synthetic data and verify and evaluate the generated synthetic data.

[0095] As described above, the present invention synthesizes data by introducing a GAN model, trains the generator and the discriminator using a real dataset, and guides the model training through a carefully defined loss function, and finally generates synthetic data that highly simulates the data distribution and characteristics of the real world. This process not only enriches the data resources, but also enhances the diversity and representativeness of the data, provides more comprehensive and reliable data support for the AI model of the automotive precision shaft structure design, and significantly improves the learning effect and prediction accuracy of the model.

[0096] Preferably, when extracting the required features from the preprocessed data, constructing a model architecture, and training the constructed model using the extracted features, it includes the following steps:

[0097] S41: Feature extraction and selection, extract material properties, geometric dimensions, load conditions, fault types, and other preset feature data from the preprocessed data, and select the extracted feature data through a filter-based feature selection algorithm;

[0098] S42: Build a model, build a deep learning network model, including defining an input layer, a hidden layer, and an output layer, and determining the number of layers of the network and the number of nodes in each layer;

[0099] S43: Model training, train the model using the extracted feature data.

[0100] Preferably, when building a deep learning network model, the loss function of the deep learning network model is:

[0101]

[0102] Among them, y i is the actual result, is the predicted result, n is the number of samples, and n is also the number of input features.

[0103] As described above, through precise feature extraction and selection, the present invention screens out key features such as material properties, geometric dimensions, and load conditions that are crucial for the design of automotive precision shaft structures from the preprocessed data, and uses a deep learning network model to construct an efficient learning architecture. By reasonably configuring the number of layers and nodes, and training the model with feature data, this process effectively improves the learning efficiency and prediction accuracy of the model, providing strong technical support for the optimal design of automotive precision shaft structures.

[0104] Preferably, when using the AI model to optimize the design parameters of the shaft structure and perform simulation analysis on the optimized precision shaft structure, the following steps are included:

[0105] S51: Enter the design scheme, and enter the design scheme of the precision shaft structure into the established deep learning network model;

[0106] S52: Optimize the design parameters, use the trained AI model to optimize the design parameters of the shaft structure, including adjusting the material selection and geometric dimensions;

[0107] S53: Perform simulation analysis, establish a simulation model of the optimized precision shaft structure, and perform running analysis on the established simulation model;

[0108] S54: Verify the results, according to the simulation analysis results, design corresponding experimental schemes to verify the accuracy of the simulation results, and adjust and optimize the model.

[0109] As described above, the present invention enters the design scheme into the deep learning network model, uses the AI model to intelligently optimize the design parameters of the shaft structure, including precisely adjusting the material selection and geometric dimensions, then constructs a simulation model for detailed running analysis, and verifies the accuracy of the simulation results through experiments, and iteratively optimizes the model accordingly. These steps not only significantly improve the efficiency and accuracy of shaft structure design, but also ensure the feasibility and reliability of the design scheme, providing strong technical support for the innovative design of automotive precision shaft structures.

[0110] Preferably, when establishing a simulation model of the optimized precision shaft structure and performing running analysis on the established simulation model, the following steps are included:

[0111] S61: Establish a simulation model of the precision shaft structure using simulation software, and set the correct material properties and boundary conditions for the simulation model;

[0112] S62: Meanwhile, according to the design of the simulation model, fabricate a corresponding experimental model, and install sensors on the experimental model for collecting data of key parameters;

[0113] S63: Run the established simulation model of the precision shaft structure in the simulation software to obtain the corresponding simulation results and data during the operation of the simulation model;

[0114] S64: Meanwhile, run the experimental model under the same conditions, compare the data of the simulation model with the data of the experimental model, and optimize and adjust the simulation model according to the results of the comparison and analysis;

[0115] S65: Repeat the processes of running the simulation model, collecting experimental data, comparative analysis, and optimization and adjustment until the simulation results and the experimental data reach a sufficient degree of coincidence.

[0116] As described above, the present invention only relies on simulation software for simulation analysis compared with the prior art, which may have the disadvantages of insufficient model accuracy and difficulty in comprehensively reflecting the actual situation. This solution adopts a method combining simulation and experiment. While establishing a simulation model of the precision shaft structure and setting correct parameters, fabricate an experimental model and install sensors to collect data. Through the processes of repeatedly running simulation and experiment, comparing and analyzing data, and optimizing and adjusting the model until the simulation results and the experimental data are highly coincident. This solution effectively improves the accuracy and reliability of the simulation model, providing a more solid theoretical and experimental basis for the optimized design of the automotive precision shaft structure.

[0117] Preferably, when generating the optimized shaft structure design drawings and detailed specifications, and generating the manufacturing process flow according to the structural requirements, the following steps are included:

[0118] S71: Parameter generation, list the specific parameters of the precision shaft structure and convert them into data in a standard format;

[0119] S72: Convert the optimized precision shaft structure into two-dimensional plane design drawings, and generate corresponding specifications according to the design specifications;

[0120] S73: According to the design drawings and specifications, formulate a detailed manufacturing process flow, and the manufacturing process flow includes the processing, heat treatment, surface treatment of the shaft, and the specific operations and process parameters of other key steps.

[0121] As described above, compared with the prior art that relies on manual experience for drawing and process planning, the present invention may have the disadvantages of large design errors, low manufacturing efficiency, and inaccurate process flows. This solution uses automated parameter generation and conversion technology to accurately list and convert the optimized precision shaft structure parameters into a standard data format, and then automatically generate two-dimensional design drawings and detailed specifications. At the same time, according to the design drawings and specification requirements, a manufacturing process flow including key steps such as shaft machining, heat treatment, and surface treatment is accurately formulated. This solution not only greatly improves the design efficiency and accuracy but also ensures the refinement and standardization of the manufacturing process, significantly enhancing the manufacturing quality and production efficiency of the automotive precision shaft structure.

[0122] Preferably, when formulating a detailed manufacturing process flow according to the design drawings and specifications, the following steps are included:

[0123] S81: Process route planning. According to the design drawings and specifications, plan the entire process route from raw materials to finished products;

[0124] S82: Process parameter determination. For each process step, determine specific operating parameters, and the determined parameters are considered based on material properties, production equipment capabilities, and quality control standards;

[0125] S83: Simulation. Conduct manufacturing simulation, perform simulation according to the existing equipment, and compare the simulated manufactured product with the precision shaft structure design drawings;

[0126] S84: According to the comparison results, repeat steps S81 - S83 for iterative optimization until the simulated manufactured product and the precision shaft structure design drawings reach a sufficient degree of coincidence.

[0127] As described above, compared with the prior art that formulates manufacturing process flows solely based on experience, which may lead to unreasonable process routes, inaccurate parameter settings, and difficult-to-guarantee product quality, this solution uses process route planning based on design drawings and specifications, accurately determines process parameters in combination with material properties, production equipment capabilities, and quality control standards, and conducts iterative optimization through the comparison of manufacturing simulation with design drawings until the simulated manufactured product and the precision shaft structure design drawings are highly coincident. This solution significantly improves the accuracy and reliability of the manufacturing process, ensures product quality, optimizes the process route at the same time, and improves production efficiency.

[0128] Preferably, after the shaft structure is put into use, when collecting its performance monitoring data, optimizing the precision shaft structure, and collecting specific problems of the precision shaft structure, and optimizing and iterating the model according to the problems, the following steps are included:

[0129] S91: Performance monitoring data collection. After the shaft structure is put into use, collect its performance monitoring data, including the shaft's lifespan, failure rate, vibration noise, and customer feedback;

[0130] S92: Problem analysis. Analyze the collected performance monitoring data and conduct diagnosis to identify the root cause of the problem;

[0131] S93: Model optimization and iteration: According to the results of problem analysis, optimize and iterate the AI model, and optimize the design of the precision shaft structure.

[0132] As described above, compared with the prior art, which lacks systematic data collection and analysis after the shaft structure is put into use, resulting in a lag in problem discovery and a long optimization and iteration cycle, this solution adopts a comprehensive performance monitoring data collection strategy, including multi-dimensional information such as the shaft's lifespan, failure rate, vibration noise, and customer feedback. Subsequently, through in-depth problem analysis and diagnosis, the root cause of the problem is accurately identified. Based on this, this solution conducts targeted optimization and iteration on the AI model and the design of the precision shaft structure. This process not only ensures the continuous optimization of the shaft structure's performance but also significantly shortens the cycle of problem discovery and solution, improving the overall reliability of the product and customer satisfaction. At the same time, through continuous model iteration, the prediction accuracy and design optimization ability of the AI model are also significantly enhanced.

[0133] The embodiments of the present invention have been described in detail above with reference to the accompanying drawings. However, the present invention is not limited to the above embodiments. Within the scope of knowledge possessed by those skilled in the art, various changes can be made without departing from the gist of the present invention.

Claims

1. A method for designing automotive precision shaft structure based on AI optimization; characterized by: The following steps are included: S11: Data collection and preprocessing: collect data related to the design of vehicle precision axle structure from various sources and preprocess the collected data; S12: AI model construction and training: extract the required features from the preprocessed data, build the model framework, and use the extracted features to train the constructed model; S13: Design optimization and simulation analysis: using AI models to optimize the design parameters of the shaft structure and perform simulation analysis on the optimized precision shaft structure; S14: Design output and manufacturing preparation, generate optimized shaft structure design drawings and detailed specifications, and generate manufacturing process flow according to structural requirements; S15: Continuous optimization. After the shaft structure is put into use, collect its performance monitoring data, optimize the precision shaft structure, collect specific problems of the precision shaft structure, and optimize and iterate the model according to the problems.

2. The method for designing an automotive precision shaft structure based on AI optimization according to claim 1, characterized in that: When collecting data related to automotive precision shaft structure design from a variety of sources, the specific data content includes historical design cases, material performance data, manufacturing process data, performance test results, and actual application data. The data sources include: A11: The internal database of the enterprise contains the enterprise’s past design, manufacturing and testing data; A12: Industry database, which contains design, manufacturing and testing data of other companies in the industry; A13: Professional website, including the latest research and applications on materials, manufacturing processes and performance testing; A14: Academic papers, including in-depth research and experimental data on materials, manufacturing processes and performance testing; A15: Government agencies, including standards and specifications for materials, manufacturing processes, and performance testing; A16: Customer feedback includes customers’ actual needs and problems encountered during actual use of the product.

3. The method for designing an automotive precision shaft structure based on AI optimization according to claim 2, characterized in that: When collecting data related to the design of automotive precision shaft structures from various sources and preprocessing the collected data, the following steps are included: S21: Data collection and integration: collect the required data through multiple channels, integrate the collected data to form a unified data format and standard, and verify the data to correct erroneous data; S22: Data cleaning, including removing duplicate data, identifying and handling missing values, and correcting erroneous data; S23: Data optimization, including data smoothing technology to eliminate data noise, data enhancement, data format optimization and data normalization; S24: Data collation: sort and merge the data, establish a version library according to the design plan of the vehicle precision axle structure, and divide the data into different version libraries.

4. The method for designing an automotive precision shaft structure based on AI optimization according to claim 3 is characterized in that: When preprocessing the collected data, it also includes synthesizing the data, that is, generating simulated data through computer algorithms to simulate the data distribution and characteristics of the real world. Specifically, it includes the following steps: S31: Prepare the dataset, collect a real dataset related to the target task for training the GAN model; S32: Build a GAN model, which includes a generator and a discriminator. The generator is responsible for generating synthetic data, while the discriminator is responsible for distinguishing between real data and synthetic data. S33: Define the loss function for the GAN model, including the loss of the generator and the loss of the discriminator; S34: Training model, using real data sets to train the GAN model; S35: Generate synthetic data. After the GAN model training is completed, use the generator to generate synthetic data, and verify and evaluate the generated synthetic data.

5. The method for designing an automotive precision shaft structure based on AI optimization according to claim 4 is characterized in that: When extracting the required features from the preprocessed data and building a model framework, and using the extracted features to train the constructed model, the following steps are included: S41: Feature extraction and selection, extracting material properties, set size, load conditions, fault type and other preset feature data from the pre-processed data, and selecting the extracted feature data through a filtering feature selection algorithm; S42: Build a model, build a deep learning network model, including defining the input layer, hidden layer and output layer, and determining the number of layers of the network and the number of nodes in each layer; S43: Model training: using the extracted feature data to train the model.

6. The method for designing an automotive precision shaft structure based on AI optimization according to claim 5, characterized in that: When using the AI ​​model to optimize the design parameters of the shaft structure and perform simulation analysis on the optimized precision shaft structure, the following steps are included: S51: input the design scheme, and input the design scheme of the precision shaft structure into the established deep learning network model; S52: Design parameter optimization, using the trained AI model to optimize the design parameters of the shaft structure, including adjusting material selection and geometric dimensions; S53: simulation analysis, establishing a simulation model of the optimized precision shaft structure, and performing operation analysis on the established simulation model; S54: Result verification: Based on the simulation analysis results, design corresponding experimental schemes to verify the accuracy of the simulation results, and adjust and optimize the model.

7. The method for designing an automotive precision shaft structure based on AI optimization according to claim 6 is characterized in that: When establishing a simulation model of the optimized precision shaft structure and performing operation analysis on the established simulation model, the following steps are included: S61: Use simulation software to establish a simulation model of the precision shaft structure and set the correct material properties and boundary conditions for the simulation model; S62: At the same time, according to the design of the simulation model, a corresponding experimental model is manufactured, and sensors for collecting data of key parameters are installed on the experimental model; S63: running the established precision shaft structure simulation model in the simulation software to obtain corresponding simulation results and data during the simulation model running process; S64: At the same time, the experimental model is run under the same conditions, and the data of the simulation model is compared with the data of the experimental model, and the simulation model is optimized and adjusted according to the results of the comparison analysis; S65: Repeat the process of running the simulation model, collecting experimental data, comparative analysis, and optimization and adjustment until the simulation results are sufficiently consistent with the experimental data.

8. The method for designing an automotive precision shaft structure based on AI optimization according to claim 7 is characterized in that: When generating optimized shaft structure design drawings and detailed specifications and generating manufacturing process flow according to structural requirements, the following steps are included: S71: Parameter generation, listing the specific parameters of the precision shaft structure and converting them into data in a standard format; S72: converting the optimized precision shaft structure into a two-dimensional design drawing, and generating corresponding specification descriptions according to the design specifications; S73: Based on the design drawings and specifications, develop a detailed manufacturing process flow, which includes the specific operations and process parameters of shaft machining, heat treatment, surface treatment and other key steps.

9. The method for designing an automotive precision shaft structure based on AI optimization according to claim 8, characterized in that: When developing a detailed manufacturing process flow based on design drawings and specifications, the following steps are included: S81: Process route planning, planning the entire process route from raw materials to finished products according to design drawings and specifications; S82: Determination of process parameters: for each process step, specific operating parameters are determined, and the determined parameters are based on material properties, production equipment capabilities and quality control standards; S83: Simulation, performing manufacturing simulation, performing simulation according to existing equipment, and comparing the simulated manufacturing product with the precision shaft structure design drawing; S84: According to the comparison results, steps S81-S83 are repeated for iterative optimization until the simulated manufacturing product and the precision shaft structure design drawing have a sufficient degree of consistency.

10. The method for designing an automotive precision shaft structure based on AI optimization according to claim 9, characterized in that: After the shaft structure is put into use, its performance monitoring data is collected, and the precision shaft structure is optimized. The specific problems of the precision shaft structure are collected, and the model is optimized and iterated according to the problems, including the following steps: S91: Performance monitoring data collection: After the shaft structure is put into use, its performance monitoring data is collected, including the shaft life, failure rate, vibration and noise, and customer feedback; S92: Problem analysis: analyze the collected performance monitoring data and perform diagnosis to identify the root cause of the problem; S93: Model optimization and iteration: Based on the results of problem analysis, the AI ​​model is optimized and iterated, and the design of the precision shaft structure is optimized.

Images