An AI large model-based high-precision industrial quality control system and an application method thereof

Abstract

The application discloses a kind of high-precision industrial quality control system and application method based on AI big model, it is characterized in that, applied to the detection and measurement of industrial product, including product data acquisition module in industrial production process, data preprocessing module, AI big model suitable for industrial quality control, quality detection module, result feedback and decision support module, control module and system management module, application AI big model suitable for industrial quality control includes constructing AI big model suitable for industrial quality control and calling AI big model suitable for industrial quality control, executes the above-mentioned control system, multimodal data fusion, the product data acquisition module of the industrial production process of the application acquires the data of multiple modalities for quality detection and measurement, AI big model suitable for industrial quality control uses data fusion algorithm, the feedback module provides intuitive visual detection result, and the detected quality problem is intuitively shown to user in the form of image, chart.

Description

TECHNICAL FIELD

[0001] The present application relates to the use of AI large models for automated detection and quality control of industrial products, including but not limited to equipment failure prediction maintenance, product quality control, production planning and scheduling optimization, industrial data analysis and mining, etc. fields, especially a high-precision industrial quality control system based on AI large model. BACKGROUND

[0002] In the field of industrial production, ensuring product quality has always been the core concern of manufacturers and consumers. Traditional quality control methods mainly rely on manual inspection and experience-based judgment. Although this method can find obvious defects to some extent, it has many limitations and challenges:

[0003] Subjectivity: Relying on personnel inspection, the state, experience, subjective judgment of the inspector will greatly affect the accuracy and consistency of the test results. Different inspectors may have different judgments on the quality of the same product, resulting in inconsistent and inaccurate quality standards.

[0004] Low efficiency: Manual inspection is relatively slow, especially in large-scale production, it is difficult to conduct detailed and comprehensive inspection on each product, and it is easy to miss the inspection, which cannot meet the needs of efficient production.

[0005] Poor adaptability: Traditional automated detection equipment is not capable of dealing with complex, irregular product features or minor defects. Its fixed algorithm and preset rule working mode cannot adapt to product diversification and production changes, and cannot accurately identify atypical quality problems.

[0006] Lack of learning ability: Traditional equipment cannot learn and improve from a large amount of data like AI large models. As production processes improve and products update, the detection accuracy and effectiveness will decrease, and manual adjustment and upgrading are required.

[0007] Data mining deficiency: Although there is a large amount of data in industrial production, traditional quality control methods lack effective mining and analysis capabilities, and a large amount of valuable information is idle, making it difficult to provide deep decision support for quality control and find potential quality problems and production optimization space.

[0008] Data island problem: The data of different production links and detection equipment are independent of each other, lacking effective integration and correlation analysis, making it difficult for quality control personnel to control product quality from a global perspective and to fully understand the impact of various factors on quality.

[0009] With the rapid development of artificial intelligence and machine learning technology, high-precision industrial quality control systems based on AI large models have emerged. This system integrates advanced AI technology and machine learning algorithms to achieve automatic, accurate, and efficient detection and control of industrial product quality. SUMMARY

[0010] To solve the above-mentioned prior art problems, the present application provides a high-precision industrial quality control system based on AI large model, characterized by being applied to the detection and measurement of industrial products, including product data acquisition module in industrial production process, data preprocessing module, AI large model suitable for industrial quality control, quality detection module, result feedback and decision support module, control module and system management module. The product data acquisition module in the industrial production process is used to comprehensively collect product data in the industrial production process. The data preprocessing module performs data cleaning, data labeling, data conversion and normalization preprocessing operations on the collected data to prepare for subsequent analysis and judgment. The AI large model suitable for industrial quality control is applied to analyze and train the product data in the industrial production process to generate high-precision industrial quality standards for products in the industrial production process. The AI large model suitable for industrial quality control includes building an AI large model suitable for industrial quality control and calling an AI large model suitable for industrial quality control. The quality detection module calls the AI large model suitable for industrial quality control to detect and measure the user's industrial products. The result feedback and decision support module feeds back the detection results to the user in real time so that they can understand the quality status of the products. Based on the detection results and historical data, the user is provided with decision support. The control module automatically adjusts the production line parameters or issues instructions based on the detection results to correct deviations and ensure product quality. The system management module stores and manages all related data, including historical data, detection results, model parameters, etc., to facilitate analysis and backtracking, monitor the running status of the system, and timely discover and solve system failures and problems. Regular maintenance and upgrading of the system are carried out to ensure the stability and reliability of the system.

[0011] As an improvement of the high-precision industrial quality control system based on AI large model of the present application, the product data acquisition module in the industrial production process is connected with various types of sensors (such as optical sensors, pressure sensors, displacement sensors, etc.) to collect various physical parameters of industrial products in the production process in real time, such as size, shape, weight, temperature, pressure, etc. High-resolution image acquisition equipment such as industrial cameras, scanners, etc. are used to collect images of the product appearance from all directions to detect surface defects, scratches, color differences, and other appearance defects of the product.

[0012] As an improvement of the high-precision industrial quality control system based on AI large model, in the electronic chip manufacturing, the product data acquisition module in the industrial production process acquires the circuit information on the chip surface by collecting the appearance image of the chip through the industrial camera.

[0013] As an improvement of the high-precision industrial quality control system based on AI large model, the data preprocessing module preprocesses the collected data, including the following three steps:

[0014] The first step is cleaning, which removes noise, outliers and duplicate data, etc., to ensure the accuracy and reliability of the data.

[0015] The second step is data labeling, which labels the cleaned data to provide accurate label information for model training. For image data, labelers need to mark the defect location, type, etc. of the product in the image; for numerical data, the corresponding product quality grade, qualified range, etc. information needs to be labeled.

[0016] The third step is to convert the collected data of different formats and units into a unified format and unit, and perform normalization processing, so that the data can be effectively calculated and analyzed in the model.

[0017] As an improvement of the high-precision industrial quality control system based on AI large model, the AI large model suitable for industrial quality control includes the following four steps:

[0018] The first step is to select a deep learning model suitable for industrial quality control, which is a convolutional neural network (CNN) for image recognition, a recurrent neural network (RNN) for processing time series data, and a deep learning model.

[0019] The second step is to combine machine learning algorithms such as support vector machines (SVM), decision trees, etc. to improve the performance and accuracy of the model.

[0020] The third step is to train and fine-tune the model using a large amount of labeled data, and continuously adjust the parameters of the model to enable it to learn the complex relationship between product quality features and qualification standards.

[0021] The fourth step is to evaluate the model, and during the training process, the model is evaluated regularly, and the validation set data is used to test the accuracy, recall rate, precision rate, etc. of the model. According to the evaluation results, the model is optimized, such as adjusting the structure of the model, increasing the amount of training data, adjusting the training parameters, etc. to improve the performance of the model.

[0022] As an improvement of the high-precision industrial quality control system based on AI large model of the present application, the AI large model suitable for industrial quality control is constructed, including selecting a large-scale language model based on Transformer architecture suitable for industrial quality control, which is adjusted and optimized for image and data analysis.

[0023] As an improvement of the high-precision industrial quality control system based on AI large model of the present application, the AI large model suitable for industrial quality control is constructed, including training the AI large model using a large amount of labeled data, which comes from historical production data, laboratory test data and artificially labeled samples, by continuously adjusting the parameters of the model to accurately identify the quality problems and defects of the products, using distributed training and optimization algorithm to improve the training efficiency and shorten the model training time.

[0024] As an improvement of the high-precision industrial quality control system based on AI large model of the present application, the AI large model suitable for industrial quality control is called, including deploying the trained AI large model to the production line for real-time detection and analysis of the products being produced:

[0025] For image data, image recognition technology is used to detect the appearance defects of the products, such as scratches, dents and color differences, and image processing technology is used to accurately measure the size, angle and other parameters of the products to ensure that they meet the design requirements;

[0026] For sensor data, the performance parameters of the products are analyzed to see if they are within the normal range, and dynamic measurements are made using sensor data, such as vibration frequency and temperature changes, to detect potential quality problems in a timely manner.

[0027] Intelligent judgment and classification, according to the detection results, the products are intelligently judged and classified into different categories, such as qualified products, unqualified products and products that need further re-inspection, and the corresponding detection data and results are recorded for subsequent tracing and analysis.

[0028] As an improvement of the high-precision industrial quality control system based on AI large model of the present application, an abnormality detection and alarm module is also provided, which can automatically or prompt the operator to handle when an abnormality or system failure is detected, to minimize production interruption;

[0029] A report and statistics module is also provided, which generates quality control reports and statistical analysis to help management understand the production quality status and system performance;

[0030] A safety and compliance module is also provided, which ensures that the system meets industry safety and compliance standards, and protects data security and privacy;

[0031] There is also an integration and compatibility module that ensures the system can be seamlessly integrated with existing production line equipment and software, supporting multiple data formats and communication protocols.

[0032] The application method of the high-precision industrial quality control system based on AI large model provided by the application has the characteristics that the high-precision industrial quality control system based on AI large model is executed, multi-modal data fusion is performed, a product data acquisition module in an industrial production process acquires data of multiple modes for quality detection and measurement, an AI large model suitable for industrial quality control adopts a data fusion algorithm, and a feedback module provides intuitive visual detection results and intuitively displays detected quality problems in the form of images and charts to users, thereby facilitating users to quickly understand the quality status of products.

[0033] As an improvement of the application method of the high-precision industrial quality control system based on AI large model, the result feedback and decision support module workflow includes:

[0034] Firstly, feedback information is sent to the user production system in a timely manner according to the detection results, an alarm device is automatically triggered to notify the user operator to process if a quality problem is found, or the parameters of the production equipment are directly adjusted to realize automatic control and avoid the continuous production of problem products.

[0035] Secondly, decision support is provided based on the detection results and historical data to provide decision support for production personnel and management personnel, the system can analyze possible causes and propose corresponding improvement suggestions such as adjusting production process parameters and replacing raw materials.

[0036] Thirdly, a detailed quality report can be automatically generated, including the number of detected products, the number of qualified products, the number of unqualified products, defect types and distribution information, and these reports can provide data support for the quality management of enterprises and help enterprises to develop quality improvement plans.

[0037] As an improvement of the application method of the high-precision industrial quality control system based on AI large model, the control module performs statistics and analysis on the detection results to provide data support for the optimization of the production process, analyze the distribution and frequency of quality problems, find out the weak links in the production process, and propose improvement measures.

[0038] As an improvement of the application method of the high-precision industrial quality control system based on the AI large model of the application, the system management module manages the users of the system, including user registration, login, permission setting, etc., manages the data in the system, including data storage, backup, recovery, etc., ensures the security and integrity of the data, prevents data loss and leakage, monitors the running state of the system, discovers and solves the faults and problems of the system in time, and regularly maintains and upgrades the system to ensure the stability and reliability of the system.

[0039] The high-precision industrial quality control system based on the AI large model and the application method have the following beneficial effects: the precision and efficiency of quality detection are greatly improved, whether it is the complex appearance defects of small household appliances such as electric kettles, toasters or coffee machines, or the slight abnormalities of parameters such as temperature, pressure and flow in the production process, can be accurately and quickly identified, greatly reducing the missed and false detection, realizing real-time multi-modal detection and intelligent judgment classification on the production line, ensuring that qualified products can be smoothly transferred, repairable products can be properly handled, and unqualified products can be timely isolated, effectively reducing the rate of defective products and improving the overall quality of products. At the same time, the result feedback and decision support function enables the enterprise management to quickly grasp the production quality status and trend, and optimizes the production process and adjusts the supplier strategy according to the data-driven decision suggestions, ensuring the continuous improvement of production. In addition, the system management function ensures the reasonable rights of different users, the safe storage and efficient use of data, and the stable operation of the system, which helps manufacturing enterprises to achieve excellent improvement in quality control, production efficiency, cost control and sustainable development from multiple aspects. BRIEF DESCRIPTION OF DRAWINGS

[0040] Figure 1 The flowchart of the preferred embodiment of the high-precision industrial quality control system based on the AI large model and the application method of the application.

[0041] Figure 2 The flowchart of the data preprocessing module of the preferred embodiment of the high-precision industrial quality control system based on the AI large model and the application method of the application.

[0042] Figure 3 The flowchart of the AI large model suitable for industrial quality control of the preferred embodiment of the high-precision industrial quality control system based on the AI large model and the application method of the application.

[0043] Figure 4 The flowchart of the application method of the preferred embodiment of the high-precision industrial quality control system based on the AI large model and the application method of the application. DETAILED DESCRIPTION

[0044] In the following, the application will be described in detail in conjunction with the drawingsFigures 1-4 and the specific embodiments, it should be noted that the technical features described below can be combined in any manner to form new embodiments, without conflict.

[0045] Reference is made to the accompanying drawings Figures 1-4 In a preferred embodiment, the present application is based on an AI large model-based high-precision industrial quality control system, which is characterized by being applied to the detection and measurement of industrial products 1, including a product data acquisition module 2 in the industrial production process, a data preprocessing module 3, an AI large model suitable for industrial quality control 4, a quality detection module 5, a result feedback and decision support module 6, a control module 7 and a system management module 8. The product data acquisition module 2 in the industrial production process is used to comprehensively collect product data in the industrial production process. The data preprocessing module 3 performs data cleaning, data labeling, data conversion and normalization preprocessing operations on the collected data to prepare for subsequent analysis and judgment. The AI large model suitable for industrial quality control 4 is applied to analyze and train the product data in the industrial production process to generate high-precision industrial quality standards for products in the industrial production process. The application of the AI large model suitable for industrial quality control 4 includes constructing and calling the AI large model suitable for industrial quality control 4. The quality detection module 5 calls the AI large model suitable for industrial quality control 4 to detect and measure the industrial products 1 of the user 9. The result feedback and decision support module 6 feeds back the detection results to the user 9 in real time, so that they can understand the quality status of the products. Based on the detection results and historical data, the user 9 is provided with decision support. The control module 7 automatically adjusts the production line parameters or issues instructions based on the detection results to correct deviations and ensure product quality. The system management module 8 stores and manages all related data, including historical data, detection results, model parameters, etc., to facilitate analysis and traceability, monitor the running status of the system, timely discover and solve faults and problems in the system, and regularly maintain and upgrade the system to ensure the stability and reliability of the system.

[0046] In a preferred embodiment, the product data acquisition module 2 in the industrial production process is connected with various types of sensors (such as optical sensors, pressure sensors, displacement sensors, etc.) to collect various physical parameters of the industrial products 1 in the production process in real time, such as size, shape, weight, temperature, pressure, etc. High-resolution image acquisition equipment such as industrial cameras, scanners, etc. are used to comprehensively collect images of the appearance of the products, detect surface defects, scratches, color differences and other appearance defects of the products.

[0047] In the preferred embodiment, in the electronic chip manufacturing, the product data acquisition module 2 in the industrial production process acquires the circuit information on the chip surface by acquiring the appearance image of the chip through an industrial camera.

[0048] Reference is made to the accompanying drawings Figure 2 In the preferred embodiment, the data preprocessing module 3 pre-processes the collected data, including the following three steps:

[0049] 301. Cleaning, removing noise, outliers, and duplicate data, etc., to ensure the accuracy and reliability of the data.

[0050] 302. Data labeling, labeling the cleaned data to provide accurate label information for model training. For image data, the labeling personnel need to mark the defect position, type, etc. of the product in the image; for numerical data, the corresponding product quality grade, qualified range, etc. information needs to be labeled.

[0051] 303. Data transformation and normalization, converting the collected data of different formats and units into a unified format and unit, and performing normalization processing to enable effective calculation and analysis of the data in the model.

[0052] The data preprocessing module 3 performs data cleaning, data labeling, data conversion and normalization processing on real-time data. The data cleaning technology can use missing value processing (such as deletion method, filling method), outlier processing (based on statistical method or clustering method), and duplicate data processing (exact matching or fuzzy matching deduplication) technology; data labeling has manual labeling assisted by professional labeling personnel using labeling tools and semi-automatic labeling based on rules or model assistance; data conversion and normalization includes data type conversion (such as date and time to timestamp, categorical variable numericalization), data format conversion (unifying image data format), and linear normalization (mapping data to the [0, 1] interval) and standard deviation normalization (making data mean 0, standard deviation 1) and other technologies.

[0053] Reference is made to the accompanying drawings Figure 3 In the preferred embodiment, the AI large model suitable for industrial quality control 4 includes the following four steps:

[0054] 401. Select a deep learning model suitable for industrial quality control, the deep learning model is a convolutional neural network (CNN) for image recognition, a recurrent neural network (RNN) for processing time series data, and a deep learning model is constructed.

[0055] The AI large model 4 suitable for industrial quality control selects a deep learning model convolutional neural network (CNN) suitable for industrial quality control for image recognition, a recurrent neural network (RNN) for processing time series data, combines machine learning algorithms such as support vector machines (SVM), decision trees, etc., to improve the performance and accuracy of the model, and uses a large amount of labeled data to train and fine-tune the model, constantly adjusting the parameters of the model to enable it to learn the complex relationship between product quality characteristics and qualification standards. During the training process, the model is periodically evaluated and optimized.

[0056] 402. Combine machine learning algorithms such as support vector machines (SVM), decision trees, etc. to improve the performance and accuracy of the model.

[0057] 403. Use a large amount of labeled data to train and fine-tune the model, constantly adjust the parameters of the model, so that it can learn the complex relationship between product quality characteristics and qualification standards.

[0058] 404. Evaluate the model, periodically evaluate the model during the training process, use validation set data to test the accuracy, recall rate, precision rate, etc. of the model, and according to the evaluation results, optimize the model, such as adjusting the structure of the model, increasing the amount of training data, adjusting the training parameters, etc. to improve the performance of the model.

[0059] In a preferred embodiment, the AI large model 4 suitable for industrial quality control includes selecting a large-scale language model based on the Transformer architecture suitable for industrial quality control, which is adjusted and optimized for image and data analysis.

[0060] In a preferred embodiment, the AI large model 4 suitable for industrial quality control includes training the AI large model using a large amount of labeled data, which comes from historical production data, laboratory test data, and manually labeled samples, by constantly adjusting the parameters of the model, it can accurately identify the quality problems and defects of the product, using distributed training and optimization algorithms to improve training efficiency and shorten model training time.

[0061] In a preferred embodiment, the AI large model 4 suitable for industrial quality control includes deploying the trained AI large model 4 suitable for industrial quality control to the production line to perform real-time detection and analysis on the products being produced:

[0062] For image data, image recognition technology is used to detect product appearance defects such as scratches, dents, and color differences, and image processing technology is used to accurately measure product dimensions, angles, and other parameters to ensure that the product meets design requirements.

[0063] For sensor data, analyze whether the performance parameters of the product are within the normal range, use sensor data for dynamic measurement such as vibration frequency, temperature change, etc., and find potential quality problems in time.

[0064] Intelligent judgment and classification, according to the detection results, the product is intelligently judged and classified into qualified products, unqualified products and products that need to be further rechecked, etc. Different categories, and record the corresponding detection data and results for subsequent tracing and analysis.

[0065] The quality detection module 5 deploys the trained AI large model 4 suitable for industrial quality control to the production line to detect and analyze the products being produced in real time, uses its fast processing capability to quickly identify product defects, abnormalities or non-compliance with quality standards, and timely alarm, such as real-time monitoring of products on a high-speed production line, multi-modal detection technology can combine appearance, physical parameters and sound, vibration and other modal data to comprehensively evaluate quality, such as analyzing running sound and vibration to judge the internal component condition in mechanical equipment detection, intelligent judgment and classification technology according to the detection results The product is accurately divided into qualified, unqualified and rechecked categories, while recording data for tracing and analysis, so as to realize comprehensive, efficient and intelligent detection of the quality of industrial products 1.

[0066] In other embodiments, an abnormality detection and alarm module is also provided, which can automatically or prompt the operator to handle when an abnormality or system failure is detected to minimize production interruption. The technical scheme includes: abnormality detection technology identifies abnormal data and potential failure signs that deviate from normal production patterns through monitoring and analyzing a large amount of real-time data using statistical analysis methods, machine learning algorithms or rule-based logical judgment, data fusion technology integrates and correlates multi-source data from different sensors and different production stages to improve the accuracy and comprehensiveness of abnormality detection, and alarm triggering technology generates an alarm signal when an abnormality is detected according to a pre-set threshold, rule or model evaluation result, and through various ways such as audible and visual alarm devices, SMS notification, email reminder or system interface pop-up window, timely informs the operator. In addition, intelligent diagnosis technology can further analyze and diagnose abnormal conditions to provide detailed fault information and treatment suggestions for the operator to assist them in taking quick and accurate measures to effectively reduce production interruption time and ensure the continuity and stability of the production process.

[0067] In other embodiments, there are also reporting and statistical modules to generate quality control reports and statistical analysis to help management understand production quality and system performance, the technical scheme is to use data analysis technology, use statistical analysis methods such as mean, standard deviation, variance, etc. Calculation, and data mining algorithm, deep analysis of data, mining out important information such as product quality indicators, defect distribution law, quality trend, etc. In addition, the report generation technology, with the help of automatic templates and programming languages, according to the preset format and content requirements, the key data and conclusions analyzed are efficiently generated into clear, intuitive charts, tables and text descriptions, etc. Form, generate detailed quality control reports, and finally the visualization display technology presents the report content in an intuitive and easy-to-understand way through a graphical interface, such as bar charts showing the number of qualified products in different time periods, line charts showing the change trend of quality indicators, etc. It is convenient for management to quickly and accurately grasp the production quality and system performance.

[0068] In other embodiments, there are also security and compliance modules to ensure that the system meets industry safety and compliance standards, protect data security and privacy, the technical scheme is to use encryption technology, through encryption processing of data, such as using symmetric encryption or asymmetric encryption algorithm, to ensure the confidentiality of data in storage and transmission process, prevent data from being stolen or tampered with, access control technology through strict user authentication, authorization and access permission management, limit unauthorized user access to system resources and data, protect the security and integrity of data, data desensitization technology can process sensitive data, hide or replace key information without affecting data usability, reduce data leakage risk, network security protection technology such as firewall, intrusion detection system, etc. Can monitor and prevent external network attacks in real time, prevent malicious intrusion, ensure the network security of the system, in addition, there is also a security audit technology, which records and audits the operation and data access of the system, so as to discover and trace potential security problems in time, and ensure that the system always meets the industry safety and compliance standards.

[0069] In other embodiments, there are also integration and compatibility modules to ensure that the system can be seamlessly integrated with existing production line equipment and software, supporting multiple data formats and communication protocols. The technology includes device interface technology, which develops special interfaces for different production line equipment to achieve physical connection and data interaction between the system and the equipment; data format conversion technology, which uses data analysis, mapping, and other methods to standardize the conversion of data in different formats such as text, images, and numerical values, so that they can be uniformly processed in the system; communication protocol adaptation technology, which supports and converts multiple common communication protocols such as TCP / IP, OPC, Profibus, etc., to ensure accurate and stable communication between the system and various devices and software; the application of middleware technology plays the role of a bridge and a link, which can shield the differences between the underlying devices and software, provide a unified service interface, and simplify the complexity of system integration, thereby achieving high integration and good compatibility of the entire system, and ensuring the coordinated operation of each link in the industrial production process.

[0070] The application provides an application method of an AI large model-based high-precision industrial quality control system, characterized in that the AI large model-based high-precision industrial quality control system of any one of claims 1-9 is executed, multi-modal data fusion, the product data acquisition module 2 in the industrial production process acquires data of multiple modalities for quality detection and measurement, the AI large model 4 suitable for industrial quality control adopts a data fusion algorithm, and the feedback module provides intuitive visual detection results, visually displays the detected quality problems to the user 9 in the form of images and charts, and facilitates the user 9 to quickly understand the quality status of the product.

[0071] In a preferred embodiment, the result feedback and decision support module 6 workflow includes:

[0072] According to the detection result, feedback information is sent to the user 9 production system in a timely manner, if a quality problem is found, an alarm device can be automatically triggered to notify the user 9 operator to handle, or the parameters of the production equipment can be directly adjusted to realize automatic control and avoid the continuous production of problem products.

[0073] Decision support, based on detection results and historical data, provides decision support for production personnel and management personnel, the system can analyze possible causes and propose corresponding improvement suggestions such as adjusting production process parameters and replacing raw materials.

[0074] Detailed quality reports can be automatically generated, including the number of detected products, the number of qualified products, the number of unqualified products, defect types and distribution, etc. These reports can provide data support for the quality management of enterprises and help enterprises to develop quality improvement plans.

[0075] The result feedback and decision support module 6 feeds back the detection results to production personnel and management personnel in real time through result feedback technology in various forms such as display screen display, mobile phone short message notification, and email notification; uses data mining and analysis technology to mine the potential relationship between quality data and production factors based on detection results and historical data, and provides basis for decision support, such as analyzing the reason for the increase in defect rate and proposing improvement suggestions; and uses automatic report generation technology to automatically generate quality reports containing detailed information such as the number of detected products, the number of qualified products, the number of unqualified products, defect types, and distribution, helping enterprises to develop quality improvement plans and realize intelligent quality control and management.

[0076] In the preferred embodiment, the control module 7 performs statistics and analysis on the detection results to provide data support for production process optimization, analyzes the distribution and frequency of quality problems, finds out the weak links in the production process, and proposes improvement measures.

[0077] The control module 7 automatically adjusts production line parameters or issues instructions based on detection results to correct deviations and ensure product quality. The technical solution is data communication technology, which can realize fast and accurate transmission of detection result data between the quality control system and the production line equipment, ensuring timely information exchange. Secondly, data analysis and processing technology is used to deeply analyze the detection results, determine the deviation situation and the corresponding adjustment strategy according to the preset quality standards and rules. Thirdly, automatic control technology is used to send accurate instructions to the controller or actuator of the production line through programming logic and interface protocol, automatically adjust key production parameters such as temperature, pressure, and speed, or trigger alarms, shutdown, and other operations, so as to timely correct deviations and ensure the stability and consistency of product quality.

[0078] In the preferred embodiment, the system management module 8 manages the users 9 of the system, including user 9 registration, login, permission setting, etc., manages the data in the system, including data storage, backup, recovery, etc., ensures the security and integrity of the data, prevents data loss and leakage, monitors the running state of the system, discovers and solves system faults and problems in time, and regularly maintains and upgrades the system to ensure the stability and reliability of the system.

[0079] The system management module 8 stores and manages all relevant data, including historical data, detection results, model parameters, etc., to facilitate analysis and traceability, monitor the running state of the system, timely discover and solve faults and problems of the system, and regularly maintain and upgrade the system. The technical scheme is related to user 9 management technology, which ensures that different users 9 have corresponding operation permissions through the management of user 9 registration, login and permission setting, and guarantees the safe and orderly operation of the system; using data management technology, responsible for data storage, backup and recovery, effectively guaranteeing the integrity and security of data, preventing data loss and leakage; using system monitoring and maintenance technology, real-time monitoring of system running state, timely discovery and solution of fault problems, and regular maintenance and upgrading of the system to maintain the stability and reliability of the system, thereby providing strong support for the efficient operation of the entire high-precision industrial quality control system.

[0080] The application is a high-precision industrial quality control system and application method based on an AI large model, which is exemplified by a production workshop of a small household appliance manufacturing enterprise. Sensors are arranged at key positions in the production process of small household appliances, such as the inner liner stretching of an electric kettle, the shell assembly, the heating wire installation of a toaster, the baking groove assembly, the water pump installation of a coffee machine, and the vicinity of the grinder, to obtain data. High-definition industrial cameras are arranged along the production line to collect appearance data, such as taking pictures of the appearance of the electric kettle from different angles, including the body, lid, handle, and other parts, to capture surface defects such as scratches, pits, and uneven color, and to check whether the connection between the spout and the body is smooth. For the toaster, the camera takes pictures of the flatness of the shell, the smoothness of the baking groove, and the installation of the buttons and indicator lights to ensure that the appearance is defect-free, the operation buttons are flexible, and the appearance is delicate. In the data preprocessing stage, the collected data is cleaned to remove sensor outliers and image artifacts, and the data is labeled according to quality standards. For example, for the inner liner pressure data of the electric kettle, the pressure range value that meets the quality requirements is labeled. For the appearance image, the positions and types of defects such as scratches and pits are accurately labeled. For the temperature data of the toaster, the appropriate baking temperature interval is labeled. For the coffee machine, the normal flow range of the water pump and the normal vibration mode of the grinder are labeled as the quality pass range. The data collected by different sensors is uniformly converted into a format suitable for model processing, such as converting temperature data from Celsius to Kelvin and converting pressure data from different units to international standard units. In the model training stage, a convolutional neural network (CNN) is selected to process appearance image data, which has strong image feature extraction capability to identify small defects on the product surface. For time series data collected by sensors (such as temperature and pressure changes over time), a recurrent neural network structure such as long short-term memory (LSTM) or gated recurrent unit (GRU) is used, combined with a support vector machine (SVM) to optimize the model for distinguishing between quality pass and fail boundary conditions, improving the accuracy of the model. For example, in judging the quality of the inner liner of the electric kettle, the results of CNN judging the appearance and SVM analyzing the pressure data are combined. A large amount of labeled quality detection data of kitchen small household appliances from different batches and different production periods, including normal products and products with various quality problems, is used to train the model. By adjusting the weights and parameters of the model, the model can learn the complex relationship between product quality features and quality standards, such as the correlation between different styles of electric kettle appearance defects and quality problems, and the relationship between toaster temperature abnormalities and baking effects. The reserved validation set data is used to evaluate the trained model, calculating the accuracy, recall rate, F1 value, and other indicators of the model. If the model's accuracy in identifying coffee grinder faults is found to be low, it may be due to insufficient training data or unreasonable model structure. The model can be optimized by increasing similar fault training data, adjusting the number of layers or neurons of the neural network, optimizing the kernel function of the SVM, and other methods.The quality detection stage detects in real time, the electric kettle, toaster, coffee machine in production Model real-time analysis sensor and image data, if there is a problem, alarm, also through multi-modal detection Comprehensive multi-class data evaluation quality, according to the results of the product Intelligent classification; The result feedback and decision support stage displays through the display screen, notifies the relevant personnel in real time Feedback results, provides decision basis according to detection and historical data and generates quality report; The system management stage does a good job in user 9 Management, sets different permissions for different positions, carries out data storage, backup, recovery and access audit Data management, real-time monitoring of system state, timely maintenance and upgrade when problems occur.

[0081] The AI large model-based high-precision industrial quality control system and application method have the following advantages: greatly improving the precision and efficiency of quality detection, whether it is the complex appearance defects of small household appliances such as electric kettles, toasters, and coffee machines, or the slight abnormalities of parameters such as temperature, pressure, and flow in the production process, can be accurately and quickly identified, greatly reducing the missed and false detection situations; Real-time multi-modal detection and intelligent judgment classification on the production line ensure that qualified products flow smoothly, repairable products are properly handled, and unqualified products are timely isolated, effectively reducing the rate of defective products and improving the overall quality of products; At the same time, the result feedback and decision support function allows enterprise management to quickly grasp the production quality status and trend, optimize production processes and adjust supplier strategies based on data-driven decision recommendations, ensuring continuous improvement of production, in addition, the system management function ensures reasonable permissions for different users, safe storage and efficient use of data, and stable operation of the system, helping manufacturing enterprises to achieve outstanding improvement in quality control, production efficiency, cost control, and sustainable development.

[0082] The above has made a detailed description of the present application, the above is only a preferred embodiment of the present application, which cannot limit the scope of the present application, that is, any equivalent changes and modifications made within the scope of the present application shall still fall within the scope of the present application.

Claims

1. A high-precision industrial quality control system based on an AI large model, characterized in that, This system is applied to the detection and measurement of industrial products. It includes a product data acquisition module, a data preprocessing module, an AI model suitable for industrial quality control, a quality inspection module, a result feedback and decision support module, a control module, and a system management module. The product data acquisition module comprehensively collects product data during industrial production. The data preprocessing module performs data cleaning, labeling, transformation, and normalization preprocessing on the collected data to prepare for subsequent analysis and judgment. The AI ​​model suitable for industrial quality control is used to analyze and train the preprocessed product data to generate high-precision industrial quality standards for the products. The application of the AI ​​model includes constructing the AI ​​model and calling AI tools suitable for industrial quality control. The system employs a large-scale AI model. The quality inspection module utilizes a suitable AI model for industrial quality control to inspect and measure the user's industrial products. The result feedback and decision support module provides real-time feedback of the inspection results to the user, enabling them to understand the product's quality status. Based on the inspection results and historical data, it provides decision support for the user. The control module automatically adjusts production line parameters or issues commands based on the inspection results to correct deviations and ensure product quality. The system management module stores and manages all relevant data, including historical data, inspection results, and model parameters, for analysis and retrospective analysis. It monitors the system's operational status, promptly identifies and resolves system faults and problems, and regularly maintains and upgrades the system to ensure its stability and reliability. The large-scale model involves deploying a trained AI model onto the production line to perform real-time detection and analysis of products in production: For image data, image recognition technology is used to detect product appearance defects, including scratches, dents, and color differences; image processing technology is used to accurately measure the product's dimensions and angle parameters to ensure that the product meets design requirements; for sensor data, the performance parameters of the product are analyzed to ensure they are within normal ranges; sensor data is used for dynamic measurement, including vibration frequency and temperature changes, to promptly identify potential quality problems; intelligent judgment and classification are performed based on the detection results, classifying products into different categories such as qualified products, unqualified products, and products requiring further re-inspection, and recording the corresponding detection data and results for subsequent traceability and analysis.

2. The high-precision industrial quality control system based on an AI large model according to claim 1, characterized in that, The product data acquisition module in the industrial production process is connected to various types of sensors, including optical sensors, pressure sensors, and displacement sensors, to collect various physical parameters of industrial products in real time during the production process, including size, shape, weight, temperature, and pressure data. High-resolution image acquisition equipment, including industrial cameras and scanners, is used to collect images of the product's appearance from all angles, and to detect defects such as blemishes, scratches, and color differences on the product's surface.

3. The high-precision industrial quality control system based on an AI large model according to claim 2, characterized in that, In electronic chip manufacturing, the product data acquisition module in the industrial production process acquires images of the chip's appearance using an industrial camera to obtain circuit information on the chip's surface.

4. The high-precision industrial quality control system based on an AI large model according to claim 1, characterized in that, The data preprocessing module preprocesses the collected data in the following three steps: The first step is data cleaning, which removes noise, outliers, and duplicate data to ensure the accuracy and reliability of the data. The second step is data annotation. The cleaned data is annotated to provide accurate label information for model training. For image data, the annotators need to mark the location and type of defects in the product in the image; for numerical data, the corresponding product quality grade and acceptable range information need to be annotated. The third step is to convert the collected data in different formats and units into a unified format and units, and then perform normalization processing so that the data can be effectively calculated and analyzed in the model.

5. A high-precision industrial quality control system based on an AI large model according to claim 1, characterized in that, The construction of a large-scale AI model suitable for industrial quality control includes: The first step is to select a suitable deep learning model for industrial quality control. The deep learning model is a convolutional neural network (CNN) for image recognition and a recurrent neural network (RNN) for processing time series data, and then construct the deep learning model. The second step is to combine machine learning algorithms, such as support vector machines (SVM) and decision trees, to improve the performance and accuracy of the model. The third step is to train and fine-tune the model using a large amount of labeled data, continuously adjusting the model's parameters so that it can learn the complex relationship between product quality characteristics and qualification standards. The fourth step is to evaluate the model. During the training process, the model is evaluated regularly using validation set data to test the model's accuracy, recall, and precision. Based on the evaluation results, the model is optimized, including adjusting the model's structure, increasing the amount of training data, and adjusting the training parameters to improve the model's performance.

6. A high-precision industrial quality control system based on an AI large model according to claim 1, characterized in that, The construction of a large-scale AI model suitable for industrial quality control includes selecting a large-scale language model based on the Transformer architecture suitable for industrial quality control. This large-scale language model based on the Transformer architecture is then tuned and optimized for image and data analysis.

7. A high-precision industrial quality control system based on an AI large model according to claim 5 or 6, characterized in that, The construction of a large AI model suitable for industrial quality control involves training the AI ​​model with a large amount of labeled data, which comes from historical production data, laboratory test data, and manually labeled samples. By continuously adjusting the model's parameters, it can accurately identify product quality problems and defects. Distributed training and optimization algorithms are used to improve training efficiency and shorten model training time.

8. A high-precision industrial quality control system based on an AI large model according to claim 1, characterized in that: It is also equipped with an anomaly detection and alarm module. When an anomaly or system failure is detected, the anomaly detection and alarm module can automatically or prompt the operator to handle the situation in order to minimize production interruption. It also includes a reporting and statistics module, which generates quality control reports and statistical analyses to help management understand the production quality status and system performance; It also includes a security and compliance module, which ensures that the system complies with industry security and compliance standards, protecting data security and privacy; It also includes an integration and compatibility module that ensures seamless integration with existing production line equipment and software, and supports multiple data formats and communication protocols.

9. An application method for a high-precision industrial quality control system based on an AI large model, characterized in that, The high-precision industrial quality control system based on an AI large model as described in any one of claims 1-8 features multimodal data fusion. The product data acquisition module in the industrial production process collects data from multiple modalities for quality detection and measurement. The AI ​​large model suitable for industrial quality control adopts a data fusion algorithm. The feedback module provides intuitive and visual detection results, and displays the detected quality problems to the user in the form of images and charts, making it convenient for the user to quickly understand the quality status of the product. The workflow of the results feedback and decision support module includes: The first step is to promptly send feedback information to the user's production system based on the test results. If a quality problem is found, the alarm device can be automatically triggered to notify the user's operators to handle it, or the parameters of the production equipment can be directly adjusted to achieve automatic control and prevent the continued production of defective products. The second step is decision support. Based on the test results and historical data, the system provides decision support for production and management personnel. It can analyze possible causes and propose corresponding improvement suggestions, including adjusting production process parameters and replacing raw materials. The third step is to automatically generate detailed quality reports, including the number of products inspected, the number of qualified products, the number of unqualified products, the types of defects and their distribution. These reports can provide data support for the company's quality management and help the company develop quality improvement plans. The control module performs statistical analysis on the detection results, provides data support for optimizing the production process, analyzes the distribution and frequency of quality problems, identifies weak links in the production process, and proposes improvement measures. The system management module manages system users, including user registration, login, and permission settings; manages system data, including data storage, backup, and recovery, ensuring data security and integrity and preventing data loss and leakage; monitors system operation status, promptly detects and resolves system faults and problems; and regularly maintains and upgrades the system to ensure its stability and reliability.

Images