An AI vision-based production process quality real-time tracing system

Abstract

The application provides a production process quality real-time tracing system based on AI vision, which comprises a data acquisition and processing module, a quality fluctuation intelligent prediction module, a parameter adjustment strategy intelligent optimization module, a production equipment intelligent dynamic adjustment module and a quality monitoring report intelligent generation module; the data acquisition and processing module comprises a data access adaptation module, a data intelligent acquisition and processing module, a data intelligent storage and verification module, a data management operation module and a data feature intelligent extraction module connected in sequence.The application combines AI vision technology with the production process quality real-time tracing system, cooperates relevant modules in the production process with an AI vision processing module, an AI vision branch and an AI vision feedback quality result module, processes data, associates production data, generates a full-process quality tracing report and makes quality tracing more intelligent.

Description

Technical Field

[0001] This invention belongs to the field of traceability system technology, and in particular relates to a real-time traceability system for production process quality based on AI vision. Background Technology

[0002] A real-time traceability system for production process quality is an information system that uses technologies such as the Internet of Things and big data to collect, track, and monitor real-time data throughout the entire production lifecycle of a product, from raw material procurement to finished product delivery. This ensures the controllability, compliance, and safety of product quality and enables rapid location and response to quality issues by establishing a complete product quality data chain.

[0003] Current intelligent engineering automation production process quality traceability systems include modules such as data acquisition and preprocessing, data storage and verification, data feature extraction, quality fluctuation prediction, and parameter adjustment strategy optimization. These existing real-time production process quality traceability systems do not have AI visual inspection capabilities, and their data traceability and integration have a low level of intelligence. Summary of the Invention

[0004] To address the aforementioned technical issues, this invention provides a real-time traceability system for production process quality based on AI vision. In this system, relevant modules in the production process work in conjunction with the AI ​​vision processing module to process data, correlate it with production data, and generate a full-process quality traceability report.

[0005] The present invention is achieved through the following technical solutions: A real-time traceability system for production process quality based on AI vision includes a data acquisition and processing module, a quality fluctuation intelligent prediction module, a parameter adjustment strategy intelligent optimization module, a production equipment intelligent dynamic adjustment module, and a quality monitoring report intelligent generation module. The data acquisition and processing module includes a data access adaptation module, a data intelligent acquisition and processing module, a data intelligent storage and verification module, a data management operation module, and a data feature intelligent extraction module connected in sequence.

[0006] Preferably, the data acquisition and processing module, the intelligent prediction module for quality fluctuations, the intelligent optimization module for parameter adjustment strategies, the intelligent dynamic adjustment module for production equipment, and the intelligent generation module for quality monitoring reports are connected in sequence.

[0007] Preferably, the data acquisition and processing module further includes a data backup module.

[0008] Preferably, the data backup module is connected to the data intelligent acquisition and processing module.

[0009] Preferably, the data intelligent acquisition and processing module is equipped with an AI vision processing module, which is used to denoise and enhance images and clean image data.

[0010] Preferably, the data feature intelligent extraction module embeds a deep learning model, and the deep learning model has an AI vision branch for simultaneously extracting parameter trend features and image visual features.

[0011] Preferably, the parameter adjustment strategy intelligent optimization module includes an AI visual feedback quality result module.

[0012] Compared with the prior art, the beneficial effects of the present invention are as follows: This invention combines AI vision technology with a real-time quality traceability system for the production process. The relevant modules in the production process work in conjunction with the AI ​​vision processing module, AI vision branch, and AI vision feedback quality result module to process the data and correlate it with the production data, generating a full-process quality traceability report, making quality traceability more intelligent. Attached Figure Description

[0013] Figure 1 This is a schematic diagram of the structure of the present invention. Detailed Implementation

[0014] The present invention will now be described in detail with reference to the accompanying drawings. As shown in Figure 1, a real-time traceability system for production process quality based on AI vision includes a data acquisition and processing module, a quality fluctuation intelligent prediction module, a parameter adjustment strategy intelligent optimization module, a production equipment intelligent dynamic adjustment module, and a quality monitoring report intelligent generation module. The data acquisition and processing module is used to collect and process various parameters such as temperature and pressure during the production process; The intelligent prediction module for quality fluctuations analyzes data containing images and numerical values. AI vision analyzes the trend of product status changes through inter-frame comparison. The intelligent prediction module for quality fluctuations uses the ARIMA time series algorithm to analyze feature data and predict quality fluctuations. When the fluctuations exceed the set threshold, anomaly marking and warnings are triggered, which can predict quality problems in advance and avoid batch defects. The parameter adjustment strategy intelligent optimization module is equipped with an AI visual feedback quality result module, which incorporates image quality scores into the reward function. Based on the early warning data from the quality fluctuation intelligent prediction module and the training data of the reinforcement learning model, the module iteratively optimizes production parameters and visual analysis results adjustment strategies through reinforcement learning model, generating real-time updated optimization schemes and providing a scientific basis for equipment adjustment. The intelligent dynamic adjustment module for production equipment uses the AI ​​vision processing module to collect images of the adjusted product and equipment operating conditions in real time, and verifies the adjustment effect through visual measurement. It uses support vector machine and K-means algorithm to analyze the stability of the adjustment, and analyzes the equipment status data and visual verification results to determine whether the adjustment is stable. If the image display quality is not improved, it triggers secondary parameter loading, loads optimized parameters, and dynamically adjusts the equipment through the industrial control system to ensure that the production process runs according to the optimal parameters. The intelligent quality monitoring report generation module has an embedded image display module. The image display module presents the quality status intuitively through comparison charts and inter-frame change animations, visualizes the data, and generates intuitive reports. It marks anomalies and judges whether production meets standards, providing a clear basis for production decisions, forming a traceability loop, and generating a full-process quality traceability report.

[0015] In this implementation scheme, specifically, the data acquisition and processing module includes a data access adaptation module, a data intelligent acquisition and processing module, a data intelligent storage and verification module, a data management and operation module, and a data feature intelligent extraction module connected in sequence. The data access adapter module connects to various sensors, industrial cameras, PLCs and communication protocols in the production system, and converts heterogeneous data such as analog and digital quantities into standard formats. The processed data is then transmitted to the intelligent data acquisition and processing module for acquisition and processing, avoiding data acquisition failures due to inconsistencies between equipment and data structures. The intelligent data acquisition and processing module timestamps the data collected by sensors and industrial cameras from the production process, along with parameters such as temperature and pressure, and collects the data in both numerical and image formats. The data intelligent acquisition and processing module is equipped with an AI vision processing module. When processing data, the AI ​​vision processing module is used to denoise and enhance the image, clean the image data, and use edge computing nodes to classify and integrate image features and parameters. The intelligent data storage and verification module segments the cleaned image data into frames, associates them with parameters, and uploads them to the production system. When marking each frame of the image with a digital signature, it embeds the image feature hash value extracted by AI vision to enhance the anti-tampering dimension. When verifying the smart contract, it simultaneously verifies the data block signature and compares the consistency of the image feature hash value to ensure that the image has not been tampered with. At the same time, it binds the image storage location with the data block address of the corresponding parameter to achieve image and text linkage traceability. The data management operation module is used to implement access control and operation logs. It assigns data access permissions based on the roles of the users, restricts unauthorized personnel from viewing or modifying key production data, and automatically records all operation behaviors, such as parameter modification, equipment adjustment, and report export, generating tamper-proof operation logs that are linked to production data in the production system to trace the impact of human operations on quality, while also meeting compliance audit requirements. The data feature intelligent extraction module embeds a deep learning model, which has an AI vision branch that simultaneously extracts parameter trend features and image visual features. By combining the two types of features through multimodal fusion technology, it can uncover the correlation patterns between parameter anomalies and image defects, making the data more complete.

[0016] In this implementation scheme, specifically, the data acquisition and processing module, the intelligent prediction module for quality fluctuations, the intelligent optimization module for parameter adjustment strategies, the intelligent dynamic adjustment module for production equipment, and the intelligent generation module for quality monitoring reports are connected in sequence.

[0017] In this implementation scheme, specifically, the data acquisition and processing module also includes a data backup module, which implements the functions of scheduled backup, fault recovery and version management; By performing scheduled backups, the data from the intelligent data acquisition and processing module is backed up off-site according to a preset cycle and stored on the offline server of the production system or the cloud of the production system. When the production system experiences network failures, data block corruption, or loss, the backup data from the data backup module is restored to ensure uninterrupted production traceability. It retains historical versions of data and supports rolling back to a dataset at a specified point in time, making it easier to troubleshoot historical quality issues.

[0018] In this implementation scheme, specifically, the data backup module is connected to the data intelligent acquisition and processing module.

[0019] Any technical solution that achieves the above-mentioned technical effects by utilizing the technical solutions described in this invention, or by designing similar technical solutions by those skilled in the art under the inspiration of the technical solutions described in this invention, falls within the protection scope of this invention.

Claims

1. A real-time quality traceability system for production processes based on AI vision, characterized in that, The AI ​​vision-based real-time traceability system for production process quality includes a data acquisition and processing module, a quality fluctuation intelligent prediction module, a parameter adjustment strategy intelligent optimization module, a production equipment intelligent dynamic adjustment module, and a quality monitoring report intelligent generation module. The data acquisition and processing module includes a data access adaptation module, a data intelligent acquisition and processing module, a data intelligent storage and verification module, a data management operation module, and a data feature intelligent extraction module, which are connected in sequence.

2. The real-time traceability system for production process quality based on AI vision as described in claim 1, characterized in that, The data acquisition and processing module, the intelligent prediction module for quality fluctuations, the intelligent optimization module for parameter adjustment strategies, the intelligent dynamic adjustment module for production equipment, and the intelligent generation module for quality monitoring reports are connected in sequence.

3. The real-time traceability system for production process quality based on AI vision as described in claim 1, characterized in that, The data acquisition and processing module also includes a data backup module.

4. The real-time traceability system for production process quality based on AI vision as described in claim 1, characterized in that, The data backup module is connected to the intelligent data acquisition and processing module.

5. The real-time traceability system for production process quality based on AI vision as described in claim 1, characterized in that, The data intelligent acquisition and processing module is equipped with an AI vision processing module, which is used to denoise and enhance images and clean image data.

6. The real-time traceability system for production process quality based on AI vision as described in claim 1, characterized in that, The data feature intelligent extraction module embeds a deep learning model, which has an AI vision branch to simultaneously extract parameter trend features and image visual features.

7. The real-time traceability system for production process quality based on AI vision as described in claim 1, characterized in that, The intelligent optimization module for parameter adjustment strategy includes an AI visual feedback quality result module.

Images