Tire designing and manufacturing system and method based on AI large model

Abstract

The invention discloses a tire designing and manufacturing system and method based on an AI large model, and the system comprises a 5G communication system which is used for achieving the real-time communication between the AI model and tire manufacturing equipment; and a master control system based on an AI model, wherein the master control system comprises four subsystems including a PC terminal, an AI production system, an AI intelligent storage system and an operation and sales terminal. The application of the tire in the intelligent manufacturing technology is effectively improved by combining the artificial intelligence technology, so that the tire production is more efficient, accurate and intelligent, the tire product is finer, more intelligent and more personalized, the management level of the tire manufacturing equipment and the tire manufacturing process is improved, the outdated situation in the aspect of tire design software is improved, and the tire manufacturing cost is reduced. The condition that the automation degree of the whole tire manufacturing production line is relatively low is improved, so that tire enterprises can better adapt to social development requirements, the tire manufacturing management level is improved, the factory lean production level is improved, and energy consumption in the tire production process is reduced.

Description

Technical Field

[0001] The present invention belongs to the technical field of industrial design, and specifically relates to a tire design and manufacturing system and method based on an AI large model. Background Art

[0002] Insufficient management of tire manufacturing process results in frequent quality problems during the tire production process, low overall tire quality, poor product performance, and very difficult human resource management. Tire design software is relatively backward, which is mainly reflected in tire contour design, tire pattern design, tire material distribution design, tire construction design, tire semi-component design, tread and sidewall mouth design, etc.

[0003] Compared with foreign countries, the degree of automation of tire manufacturing equipment and systems is relatively low. The vast majority of tire companies still remain in labor-intensive production lines, and tire production, inspection, storage, and delivery are greatly affected by labor costs. Summary of the invention

[0004] The purpose of the present invention is to provide a tire design and manufacturing system and method based on an AI large model, effectively improving the application of tires in intelligent manufacturing technology, making tire production more efficient, accurate and intelligent, so as to solve the problems raised in the above background technology.

[0005] To achieve the above object, the present invention adopts the following technical solution: a tire design and manufacturing system based on AI large model, comprising:

[0006] An IoT-based 5G communication system to enable real-time communication between AI models and tire manufacturing equipment; and

[0007] The overall control system based on the AI ​​model includes four subsystems: PC terminal, AI production system, AI intelligent storage system, and operation and sales terminal, among which:

[0008] The PC terminal is used for information retrieval, deep learning CAD design, CAD automation, CAE simulation verification, tire knowledge rule transfer learning, visual verification analysis and result output;

[0009] The AI ​​production system includes semi-component process, molding, vulcanization, appearance inspection, dynamic average inspection, and sorting process;

[0010] The AI ​​intelligent storage system is used to manage the storage and unmanned transportation of raw materials, semi-components, embryos, and finished products;

[0011] The operation and sales terminals are used for new product development, production planning, sales, and after-sales tracking;

[0012] The deep learning CAD design generates a tire design by acquiring image data and referring to it for 2D pattern or sidewall scheme design, reducing data dimension to avoid dimensionality disaster and overfitting, and using dimension reduction features to access a three-dimensional design platform to generate a CAD model;

[0013] It also includes: a feedback module that uses knowledge rule transfer learning to improve simulation prediction performance, guides design decisions through visual verification results, and ultimately outputs reliable solutions to the storage server and feeds them back to the production process.

[0014] Preferably, the 5G communication system includes a computing server, a storage server, a mobile terminal information receiving device, and a PLC communication control unit module in the tire production equipment.

[0015] Preferably, the information retrieval in the PC terminal is used to provide the client with the ability to retrieve tire product information from the Internet and an internal database of an enterprise.

[0016] Preferably, the semi-component processes in the AI ​​production system include mixing, calendering, extrusion, cord and belt layer cutting, wire winding and bonding equipment.

[0017] Preferably, the AI ​​intelligent storage system dynamically adjusts the inventory of raw materials, semi-components and finished products to match production plan requirements through unmanned transport vehicles and intelligent management systems.

[0018] Preferably, the production plan in the operation and sales terminal is used to automatically adjust the production plan based on market forecasts and customer orders to achieve supply chain collaboration.

[0019] Preferably, the tire knowledge rule transfer learning adopts a convolutional neural network, which optimizes model parameters by learning from historical simulation data to improve the accuracy of future product performance prediction.

[0020] Preferably, during the visual verification analysis, engineers evaluate the physical characteristics and performance of the design through an interactive interface.

[0021] Preferably, the PC terminal result output system is used to upload design drawings to a storage server, and automatically generate various component dimension drawings and material configuration information required for production, directly connecting to the production process.

[0022] On the other hand, the present invention provides a tire design and manufacturing method based on an AI large model, comprising the following steps:

[0023] Establish a real-time communication network through the 5G communication system based on the Internet of Things to realize data exchange and command transmission between the AI ​​model and equipment in various links of tire manufacturing;

[0024] Establish a comprehensive AI model control system, which consists of four subsystems: PC terminal, AI production system, AI intelligent storage system, and operation and sales terminal, to build intelligent management of the entire chain from design to manufacturing to market circulation;

[0025] The following steps are implemented on the PC terminal: Information retrieval: Collect and analyze tire product information on the Internet and within the company to provide data support for new designs; Deep learning CAD design: Use image data to generate 2D tire patterns or sidewall designs; Data dimensionality reduction processing: Reduce the dimensionality of pattern design data; 3D model generation: Import the reduced feature data into the 3D design platform and create a CAD model through the secondary development interface; CAE simulation verification: Simulate and verify the generated 3D model, including but not limited to tire footprint, stiffness and drainage performance; Knowledge rule transfer learning: Use convolutional neural networks to predict simulation results, and continuously learn to optimize model prediction performance; Visual verification analysis: Design engineers evaluate the rationality of the design through graphical displays; Result output: Reliable design solutions are uploaded to the storage server and fed back to the production system;

[0026] In the AI ​​production system, the semi-component processing, tire molding, vulcanization, appearance and dynamic balance inspection, and sorting are subdivided;

[0027] The AI ​​intelligent storage system realizes intelligent management of raw materials, semi-components, embryos and finished products, uses unmanned transport vehicles for material dispatch, and automatically adjusts storage strategies according to production plans;

[0028] The operation and sales terminals cover new product planning, production scheduling, sales activity management and after-sales service tracking to ensure the consistency between market response speed and customer needs;

[0029] A feedback mechanism is established to continuously optimize the model prediction capability based on design verification results and knowledge rule transfer learning, and the confirmed design plan is fed back to the production process to guide actual manufacturing.

[0030] Technical effects and advantages of the present invention: Compared with the prior art, the tire design and manufacturing system and method based on AI large model proposed by the present invention has the following advantages:

[0031] The present invention combines artificial intelligence technology to effectively improve the application of intelligent manufacturing technology in tires, making tire production more efficient, accurate and intelligent, making tire products more refined, intelligent and personalized, improving the management level of tire manufacturing equipment and tire manufacturing processes, improving the backwardness of tire design software, and improving the relatively low degree of automation of the overall tire manufacturing production line, so that tire companies can better adapt to the needs of social development, improve the level of tire manufacturing management, improve the level of lean production in factories, and reduce energy consumption in the tire production process. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 is a block diagram of the overall control system of the present invention;

[0033] Figure 2 is a block diagram of a PC terminal of the present invention;

[0034] Figure 3 is a block diagram of the AI ​​production system of the present invention;

[0035] Figure 4 A block diagram of the AI ​​intelligent storage system of the present invention;

[0036] Figure 5 A block diagram of the operation and sales terminal of the present invention;

[0037] Figure 6 The block diagram of the tire design and manufacturing system based on the AI ​​big model of the present invention. DETAILED DESCRIPTION

[0038] The following will be combined with the accompanying drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments. The specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0039] The present invention provides a tire design and manufacturing system based on AI big model, such as Figure 6 As shown, it includes: 5G communication system, overall control system and feedback module. Among them,

[0040] The 5G communication system is based on the Internet of Things and is used to achieve real-time communication between the AI ​​model and various tire manufacturing equipment. It uses an advanced 5G communication system based on the Internet of Things technology to ensure high-speed, low-latency communication capabilities, providing a solid information transmission foundation for the entire tire design and manufacturing system. The system integrates computing servers, storage servers, mobile terminal information receiving devices, and PLC communication control unit modules in tire production equipment, achieving comprehensive coverage from design to production. Equipment such as semi-component unmanned transport vehicles, tire blank molding semi-component intelligent storage systems, intelligent sorting, machine vision inspection, and finished tire intelligent sorting and storage systems all achieve instant communication with the AI ​​model through the 5G network, ensuring real-time data transmission and instruction execution.

[0041] The overall control system based on the AI ​​model includes four subsystems: PC terminal, AI production system, AI intelligent storage system, and operation and sales terminal. Figure 1 shown.

[0042] The PC terminal is used for information retrieval, deep learning CAD design, CAD automation, CAE simulation verification, tire knowledge rule transfer learning, visual verification analysis and result output; information retrieval is used to provide the client with tire product information retrieval from the Internet and the internal database of the enterprise. As the human-computer interaction interface of the system, it integrates the information retrieval client, deep learning CAD design tool, CAD automation, CAE simulation verification module, tire product knowledge rule transfer learning system, visual verification analysis and result output system. The information retrieval client can quickly obtain internal and external tire design reference data, while the deep learning CAD design quickly iterates the tire design scheme through three stages (2D generation, dimensionality reduction processing, and three-dimensional model generation). Tire knowledge rule transfer learning uses convolutional neural networks to optimize model parameters and improve the accuracy of future product performance predictions by learning historical simulation data. During the visual verification analysis process, engineers evaluate the physical characteristics and performance of the design through an interactive interface. The PC terminal result output system is used to upload design drawings to the storage server, and automatically generate various component dimension drawings and material configuration information required for production, directly connecting to the production process.

[0043] The production capacity product information retrieval client mainly conducts information retrieval of tire products, including Internet product information retrieval, company-owned product information retrieval, and integrated CAD system, which mainly includes three stages:

[0044] First, 2D generative design, specifically including, obtaining information retrieval client product classification image data and using it as a reference, and then generating a series of two-dimensional tire patterns or sidewall solutions based on the information retrieval client image data combined with topology optimization and deep neural network learning.

[0045] In the second stage, the dimensionality reduction operation is performed on the tread pattern or sidewall solution data of the first stage to avoid the dimensionality disaster during the system learning process or to avoid overfitting of the model, which will lead to excessive complexity of the model and geometric growth of computing resources.

[0046] In the third stage, by connecting the computer-aided design platform with the AI ​​model, the reduced-dimensional tire pattern and sidewall feature data vectors are connected to the three-dimensional design platform through the secondary development interface to generate CAD data or three-dimensional models.

[0047] CAD automated design includes extended design based on existing product design factors, and also includes automatically generating 3DCAD data as CAE input and preprocessing the model, including meshing, 3D model coordinate conversion, and CAE simulation experiment verification, which specifically includes tire footprint simulation, stiffness simulation, drainage simulation, etc.

[0048] Tire product knowledge rule transfer learning uses convolutional neural grids with transfer learning to establish an agent model to predict simulation results, and through continuous data expansion and transfer learning, the system continuously reduces fitting and improves the prediction performance of the model; in the visual verification and analysis stage, product design engineers or simulation engineers use visual analysis results to judge the reliability of product design solutions, and submit reliable solutions to the result output system.

[0049] The result output system feeds back to the PC terminal system and uploads the corresponding 3D model and tire structure plan of the plan submitted in the visualization verification analysis phase to the storage server, including tire pattern design drawings, sidewall design drawings, 3D material distribution diagrams, etc.

[0050] The result output system also performs reverse process simulation on the three-dimensional pattern model and the corresponding three-dimensional material distribution diagram model according to the visualization submission plan to produce the corresponding tread component mouth shape size diagram, sidewall mouth shape diagram, belt layer material and size diagram, crown belt optimal winding method and pitch data, inner liner thickness and corresponding size data, No. 1 and No. 2 or No. 3 cord material and size data, etc. The above data is submitted to the PC end, and then submitted to the AI ​​model main server through the PC end, and the main server performs the next step according to the AI ​​model instructions.

[0051] The AI ​​production system is subdivided into semi-component process, molding process, vulcanization process, appearance inspection process, dynamic average inspection process and sorting process, such as Figure 3 As shown. It covers the entire production chain from raw material processing to the final product. Each process is equipped with specific equipment, such as mixing equipment, calender, extruder, etc., to ensure production efficiency and quality; the semi-component process includes mixing, calendering, extrusion, cord and belt cutting, wire winding and bonding equipment.

[0052] Through the AI ​​general service, the PC-side tire design and manufacturing 3D model and related semi-component model that have been confirmed and optimized by engineers are obtained, and the trial production plan of the product and the corresponding semi-component product design data are issued according to the new product development plan or existing product production plan of the operation and sales side, and the feedback is sent to the subsystem production system through the AI ​​service. The production system splits the corresponding semi-component process key parameters according to the total AI service design parameters, and sends them to the corresponding production process. After confirmation by the corresponding production process process technicians, they are transmitted to the corresponding production equipment.

[0053] The AI ​​intelligent storage system mainly includes unmanned transport vehicles, raw material storage, semi-component storage, tire embryo storage, finished tire storage, etc. Figure 4As shown in the figure, through the intelligent management of unmanned transport vehicles, raw materials and semi-components, the automated storage and scheduling of raw materials to finished products is realized. The system can independently determine the inventory situation and provide decision support for production planning. The AI ​​intelligent storage system dynamically adjusts the inventory of raw materials, semi-components and finished products to match the production plan requirements through unmanned transport vehicles and intelligent management systems.

[0054] The operation and sales terminal includes functional modules such as new product development submission, production plan formulation, sales management, and after-sales tracking. Figure 5 As shown in the figure, market feedback is closely connected with production plan. Production plan is used to automatically adjust production plan according to market forecast and customer orders to achieve supply chain collaboration.

[0055] The deep learning CAD design generates a tire design by acquiring image data and referring to it for 2D pattern or sidewall scheme design, reducing data dimension to avoid dimensionality disaster and overfitting, and using dimension reduction features to access a three-dimensional design platform to generate a CAD model;

[0056] When implementing this deep learning CAD design process, high-performance image acquisition equipment, such as Teledyne DALSA Gen ie Nano series industrial cameras, are first used with high-resolution lenses to accurately capture the classified image data of tire products. These image data are transmitted in real time via the 5G communication system to the computing server, which is equipped with Intel Xeon Scale processors and NVIDIA Tesla V100 GPUs to provide powerful AI computing support.

[0057] On the information retrieval client, the algorithm uses TensorFlow or PyTorch framework to build a deep learning model, combined with topology optimization algorithms such as genetic algorithm or particle swarm optimization, to preprocess and extract features from image data. The goal of this stage is to generate innovative two-dimensional tire patterns or sidewall designs. Through trained models, such as U-Net or GANs (generative adversarial networks), a variety of design patterns can be generated.

[0058] In order to overcome the curse of dimensionality and prevent model overfitting, dimensionality reduction techniques such as PCA (Principal Component Analysis) or t-SNE (t-Distributed Neighbor Embedding) are introduced to effectively reduce the dimension of the feature space, retaining key data information while reducing the computational burden. This step is performed on the server's GPU, using CUDA to accelerate the calculation to ensure efficient completion.

[0059] After that, the reduced feature vectors are integrated with mainstream computer-aided design software such as Autodesk Fusion 360 or Dassau lt Systèmes SOLIDWORKS through a custom API interface. Using the SDK of these software, the feature vectors are directly mapped to CAD parameters to drive the automatic generation of 3D models. Designers can make fine adjustments based on this and further improve the design details using the built-in advanced modeling tools of the software.

[0060] Finally, the generated 3D tire model is verified by CAE simulation, such as ANSYS or ABAQUS software, to conduct simulation tests on tire footprint, stiffness, and drainage performance. The simulation results are fed back to the knowledge rule transfer learning module, and pre-trained convolutional neural networks such as ResNet or Inception are used for transfer learning to continuously optimize simulation predictions and improve model prediction accuracy. Engineers use visualization tools such as Tableau or Power BI to analyze simulation results and make a comprehensive assessment of design reliability.

[0061] After the entire design process is completed, the 3D model and design drawings confirmed by the engineers are uploaded to the storage server through the company's internal network to ensure data security and efficient access. These design results become the basis for subsequent manufacturing instructions, promote the automated production process of the tire manufacturing system, and achieve seamless connection from design to manufacturing.

[0062] The feedback module is used to improve simulation prediction performance by using knowledge rule transfer learning, guide design decisions through visual verification results, and finally output reliable solutions to the storage server and feed them back to the production process.

[0063] As a core component in the tire design and manufacturing system based on AI big model, the feedback module’s workflow and technical details are crucial. The specific implementation is as follows:

[0064] First, the feedback module integrates knowledge rule transfer learning capabilities, using high-performance hardware such as NVIDIA RTXA6000 GPUs and workstations equipped with Intel Xeon W-3275M processors, running TensorFlow or PyTorch deep learning frameworks to achieve fast and efficient data processing. At this stage, the system builds a proxy model through a convolutional neural network (CNN), using transfer learning technology to learn patterns from previous simulation results and continuously optimize prediction performance. Specifically, it extracts a large amount of simulation data from a historical database, including performance indicators of different tire designs (such as wear resistance, grip, rolling resistance, etc.), and adjusts the pre-trained model weights through fine tuning technology, so that the model can more accurately predict the performance of new designs, thereby reducing the need for actual physical experiments and reducing R&D costs.

[0065] In order to improve the accuracy of simulation predictions, the feedback module will also integrate an automatic data expansion mechanism to increase the diversity of training samples through random rotation, flipping, scaling and other transformations, further enhancing the generalization ability of the model. This process is managed by a high-performance computing cluster, such as the HPE Apollo 6500 system, to ensure the efficiency and stability of large-scale data processing.

[0066] Next, the visual verification phase relies on advanced data visualization software, such as MATLAB or Tableau, to convert complex simulation results into intuitive charts and 3D images, allowing product design engineers to quickly understand the pros and cons of the design and make scientific decisions. Engineers can dynamically adjust design parameters through an interactive interface, instantly observe changes in simulation results, and quickly iterate to optimize the design. This step is performed on a high-resolution display to ensure that details are clearly visible.

[0067] Finally, the proven and reliable design solutions are transmitted to enterprise-level storage servers, such as IBM Storwi ze V7000, through high-speed networks, such as 5G-based IoT architectures, for secure storage and backup. These data not only include the tire's CAD drawings and structural design documents, but also all relevant parameters and simulation reports, providing accurate manufacturing instructions for subsequent production processes. The system also synchronizes the data to the production scheduling center, which may use the Simens Opcenter MES system to automatically trigger production plan adjustments to ensure that every link on the production line can be executed according to the latest optimized design, thereby improving overall manufacturing efficiency and product quality.

[0068] In summary, the feedback module integrates efficient hardware, advanced algorithms and software to achieve effective migration of knowledge rules, continuous optimization of simulation prediction performance and visual decision-making assistance, ensuring seamless connection and intelligent feedback loop from design to production.

[0069] On the other hand, the present invention provides a tire design and manufacturing method based on an AI large model, comprising the following steps:

[0070] Step 1: Establish a real-time communication network through the 5G communication system based on the Internet of Things to realize data exchange and command transmission between the AI ​​model and equipment in various links of tire manufacturing;

[0071] Step 2: Establish a comprehensive AI model control system, which consists of four subsystems: PC terminal, AI production system, AI intelligent storage system, and operation and sales terminal, to build intelligent management of the entire chain from design to manufacturing to market circulation;

[0072] Step 3: Implement the following steps on the PC terminal: Information retrieval: Collect and analyze tire product information on the Internet and within the company to provide data support for new designs; Deep learning CAD design: Use image data to generate 2D tire patterns or sidewall designs; Data dimensionality reduction processing: Reduce the dimensionality of pattern design data; 3D model generation: Import the reduced feature data into the 3D design platform and create a CAD model through the secondary development interface; CAE simulation verification: Simulate and verify the generated 3D model, including but not limited to tire footprint, stiffness and drainage performance; Knowledge rule transfer learning: Use convolutional neural networks to predict simulation results, and continuously learn to optimize model prediction performance; Visual verification analysis: Design engineers evaluate the rationality of the design through graphical displays; Result output: Reliable design solutions are uploaded to the storage server and fed back to the production system;

[0073] Step 4: In the AI ​​production system, subdivide the semi-component processing, tire molding, vulcanization, appearance and dynamic balance inspection, and sorting;

[0074] Step 5: The AI ​​intelligent storage system realizes the intelligent management of raw materials, semi-components, embryos and finished products, uses unmanned transport vehicles for material dispatch, and automatically adjusts the storage strategy according to the production plan;

[0075] Step 6: Operation and sales terminals cover new product planning, production scheduling, sales activity management and after-sales service tracking to ensure the consistency of market response speed and customer needs;

[0076] Step 7: Establish a feedback mechanism to continuously optimize the model prediction capability based on design verification results and knowledge rule transfer learning, and feed back the confirmed design plan to the production process to guide actual manufacturing.

[0077] To sum up, the present invention combines artificial intelligence technology to effectively improve the application of tires in intelligent manufacturing technology, making tire production more efficient, accurate and intelligent, making tire products more refined, intelligent and personalized, improving the management level of tire manufacturing equipment and tire manufacturing processes, improving the backwardness of tire design software, and improving the relatively low degree of automation of the overall tire manufacturing production line, so that tire companies can better adapt to the needs of social development, improve the level of tire manufacturing management, improve the level of lean production in factories, and reduce energy consumption in the tire production process.

[0078] In some other embodiments, a large artificial intelligence model with parameters exceeding 100 billion is also included, which is used for tire design, manufacturing, process management, warehousing, operation, logistics, etc. and for related artificial intelligence training.

[0079] In some other embodiments, a PLC communication control unit module based on the AI ​​large model of tire manufacturing equipment is also included, which mainly includes a tire mixer communication control module, a calender communication control module, a cutting machine communication control module, an extruder communication control module, a molding machine communication control module, a vulcanizer communication control module, an appearance inspection machine communication control module, a uniformity machine inspection communication control module, a dynamic balancing machine inspection communication control module, etc.

[0080] In some other embodiments, an online measurement system of the tire manufacturing equipment is also included, which is used to detect in real time whether the tire production semi-components meet the process requirements, specifically including an online size measurement system, an online temperature measurement system, an online pressure measurement system, and an online visual measurement system for automatically detecting whether the semi-components contain steam drums, impurities, color anomalies or other problems. The online measurement system mainly includes laser ranging sensors, temperature sensors, pressure sensors, etc., which are used to online measure and feedback data information such as the size, temperature, and pressure of each semi-component of the tire for AI model judgment and analysis.

[0081] In some other embodiments, a computer-aided design tool based on an AI large model is also included to assist in the design and calculation of the size information of each half component of the tire. The design tool includes a two-dimensional design platform and a three-dimensional design platform based on the AI ​​large model, and includes an application integration interface and secondary development plug-in for commonly used office software.

[0082] In some other embodiments, an automatic adjustment and compensation device for the tire manufacturing equipment is provided on the basis of the existing tire manufacturing equipment, which is used to adjust the deviation of the tire during the manufacturing process. According to different production equipment, it includes a pressure control and adjustment system, a temperature control system, a hydraulic control system, a stroke control and adjustment control system, a roller speed control system, a correction device or a robot, etc.

[0083] In some other embodiments, it also includes an APP based on a mobile phone or a mobile terminal, which is used to receive information from the tire design and manufacturing system based on the AI ​​large model, and is used for the system to provide real-time feedback to process managers on equipment problems and semi-component production problems in the tire production process. The feedback mechanism is divided into two parts. The first part is based on the semi-component production online detection system, which is used to provide real-time feedback to the system to adjust equipment parameters in real time to meet the production requirements of tire semi-components. The other part is used to provide feedback to relevant process managers on problems that the AI ​​model cannot handle or the system cannot handle, and to provide feedback on production equipment information, location, machine number, etc., so that process managers can quickly locate the specific location of production problems.

[0084] The tire design level has been improved. The tire design platform based on the AI ​​model is conducive to improving the problems in tire design and improving the overall performance level of tires through the design training of a large number of existing tire products. Compared with the traditional tire design and manufacturing system, the tire design system connected to the AI ​​model makes it possible for an experienced process personnel to monitor and perform duties in real time in each process of the tire production process. According to the effective data training of the process personnel for each process, the system can make corresponding processing methods for different types of problems. It specifically includes three aspects. First, the system can meet production needs by adjusting parameters. Second, according to the online measurement feedback data, based on the judgment of the AI ​​model, combined with the actual production needs, the tire production equipment is decided to stop or continue production, and the system message is sent to the process personnel's mobile terminal in time and reminded. Third, for the lack of corresponding raw materials for each production equipment, it is directly fed back to the superior raw material process process personnel or the intelligent delivery vehicle to add raw materials.

[0085] Improve the management level of tire manufacturing, improve the lean production level of the factory, reduce energy consumption in the tire production process, and enhance the competitiveness of tire production. Specifically, it includes the level of tire digital manufacturing technology, which specifically includes, first, tire design and manufacturing. Through the tire design and manufacturing system based on AI big model, in the design stage, by combining the data training of designers for AI big model, the AI ​​big model has the ability and experience of general designers, and can effectively carry out basic tire design, and output solutions, so that tire designers can optimize the design and improve the product more quickly according to the solution, and improve the product design level and efficiency. Second, realize the full life cycle management of tire product development and design based on AI model. On top of the traditional full life cycle management system, establish full life cycle automatic management and standardization based on AI system, so that planning, production, sales, process technicians, sorting, and warehousing personnel can work more efficiently, concisely, and intelligently.

[0086] Improve production quality. Compared with traditional equipment and systems, production quality can be improved from the following aspects. First, by connecting production equipment to the system and coordinating the corresponding self-correcting physical adjustment mechanism and system, the measurement system based on online real-time feedback can effectively solve the tolerance of production parts caused by other factors within a certain range by combining physical self-correcting equipment, so as to achieve self-adjustment and correction of equipment within a given tolerance range. Second, by adding an online intelligent visual inspection system, it not only saves labor costs, but also effectively reduces the appearance of waste and defective products in tire production. Third, through the tire design and manufacturing system based on AI large models, the design accuracy of semi-components in the tire production and design process can be effectively improved. By comparing the online feedback measurement data and design data, the errors in the tire design process can be effectively reduced, and the precision performance of product design can be improved.

[0087] Compared with traditional equipment and production systems, improving tire production efficiency mainly includes the following aspects: First, AI models based on artificial intelligence are effectively trained for tire design and manufacturing processes, including intelligent scheduling and use of semi-components, intelligent storage of semi-finished products for embryo molding, and intelligent sorting of vulcanized tires. Second, the appearance inspection system based on machine vision is combined with a small amount of manual inspection to effectively improve the efficiency of appearance inspection. Third, through communication and interconnection with the mobile terminals of process personnel, according to the timely push of AI models, timely and effective solutions are provided for process problems in the tire production process, reducing downtime problems caused by tire process production and improving tire production efficiency.

[0088] Finally, it should be noted that the above is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, it is still possible for those skilled in the art to modify the technical solutions described in the aforementioned embodiments or to make equivalent substitutions for some of the technical features therein. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A tire design and manufacturing system based on AI big model, characterized in that: include: An IoT-based 5G communication system to enable real-time communication between AI models and tire manufacturing equipment; as well as The overall control system based on the AI ​​model includes four subsystems: PC terminal, AI production system, AI intelligent storage system, and operation and sales terminal, among which: The PC terminal is used for information retrieval, deep learning CAD design, CAD automation, CAE simulation verification, tire knowledge rule transfer learning, visual verification analysis and result output; The AI ​​production system includes semi-component process, molding, vulcanization, appearance inspection, dynamic average inspection, and sorting process; The AI ​​intelligent storage system is used to manage the storage and unmanned transportation of raw materials, semi-components, embryos, and finished products; The operation and sales terminals are used for new product development, production planning, sales, and after-sales tracking; The deep learning CAD design generates a tire design by acquiring image data and referring to it for 2D pattern or sidewall scheme design, reducing data dimension to avoid dimensionality disaster and overfitting, and using dimension reduction features to access a three-dimensional design platform to generate a CAD model; It also includes: a feedback module that uses knowledge rule transfer learning to improve simulation prediction performance, guides design decisions through visual verification results, and ultimately outputs reliable solutions to the storage server and feeds them back to the production process.

2. The tire design and manufacturing system based on AI big model according to claim 1 is characterized in that: The 5G communication system includes a computing server, a storage server, a mobile terminal information receiving device, and a PLC communication control unit module in the tire production equipment.

3. The tire design and manufacturing system based on AI big model according to claim 1 is characterized in that: The information retrieval in the PC terminal is used to provide the client with the ability to retrieve tire product information from the Internet and the company's internal database.

4. The tire design and manufacturing system based on AI big model according to claim 1 is characterized in that: The semi-component processes in the AI ​​production system include mixing, calendering, extrusion, cord and belt cutting, wire winding and bonding equipment.

5. The tire design and manufacturing system based on AI big model according to claim 1, characterized in that: The AI ​​intelligent storage system dynamically adjusts the inventory of raw materials, semi-components and finished products to match production plan requirements through unmanned transport vehicles and intelligent management systems.

6. The tire design and manufacturing system based on AI big model according to claim 1, characterized in that: The production plan in the operation and sales terminal is used to automatically adjust the production plan based on market forecasts and customer orders to achieve supply chain collaboration.

7. The tire design and manufacturing system based on AI big model according to claim 1 is characterized in that: The tire knowledge rule transfer learning adopts a convolutional neural network, which optimizes model parameters by learning historical simulation data, thereby improving the accuracy of future product performance prediction.

8. The tire design and manufacturing system based on AI big model according to claim 1, characterized in that: During the visual verification analysis process, engineers evaluate the physical characteristics and performance of the design through an interactive interface.

9. The tire design and manufacturing system based on AI big model according to claim 1, characterized in that: The PC terminal result output system is used to upload design drawings to the storage server, and automatically generate various component dimension drawings and material configuration information required for production, directly connecting to the production process.

10. A tire design and manufacturing method based on AI large model, characterized in that: The following steps are involved: Establish a real-time communication network through the 5G communication system based on the Internet of Things to realize data exchange and command transmission between the AI ​​model and equipment in various links of tire manufacturing; Establish a comprehensive AI model control system, which consists of four subsystems: PC terminal, AI production system, AI intelligent storage system, and operation and sales terminal, to build intelligent management of the entire chain from design to manufacturing to market circulation; The following steps are implemented on the PC terminal: Information retrieval: Collect and analyze tire product information on the Internet and within the company to provide data support for new designs; Deep learning CAD design: Use image data to generate 2D tire patterns or sidewall designs; Data dimensionality reduction processing: Reduce the dimensionality of pattern design data; 3D model generation: Import the reduced feature data into the 3D design platform and create a CAD model through the secondary development interface; CAE simulation verification: Simulate and verify the generated 3D model, including but not limited to tire footprint, stiffness and drainage performance; Knowledge rule transfer learning: Use convolutional neural networks to predict simulation results, and continuously learn to optimize model prediction performance; Visual verification analysis: Design engineers evaluate the rationality of the design through graphical displays; Result output: Reliable design solutions are uploaded to the storage server and fed back to the production system; In the AI ​​production system, the semi-component processing, tire molding, vulcanization, appearance and dynamic balance inspection, and sorting are subdivided; The AI ​​intelligent storage system realizes intelligent management of raw materials, semi-components, embryos and finished products, uses unmanned transport vehicles for material dispatch, and automatically adjusts storage strategies according to production plans; The operation and sales terminals cover new product planning, production scheduling, sales activity management and after-sales service tracking to ensure the consistency between market response speed and customer needs; A feedback mechanism is established to continuously optimize the model prediction capability based on design verification results and knowledge rule transfer learning, and the confirmed design plan is fed back to the production process to guide actual manufacturing.

Images