Digital factory visual management system and method based on AI

Abstract

The invention discloses an AI-based digital factory visual management system and method, and belongs to the field of factory management, and the system comprises a data collection layer which employs an Internet of Things device, an intelligent sensor and a device communication protocol, and is used for collecting data from each factory, a production line and a system in real time, and achieves the cross-region and cross-system data interconnection and intercommunication; the data processing and integration layer performs standardization and format conversion on heterogeneous data by using natural language processing and deep learning technologies, fuses multi-source data by using a machine learning algorithm, eliminates data islands, and ensures timeliness and accuracy of the data; according to the method, cross-region and cross-system data integration and sharing are realized, data islands are broken, and the resource allocation efficiency is improved; the production efficiency and the flexibility are remarkably improved and the production cost is reduced based on intelligent decision making and scheduling of AI; visualization and real-time monitoring functions provide visual and comprehensive production information for managers, and are convenient for rapid decision making.

Description

Technical Field

[0001] The present invention relates to the technical field of factory management, and in particular to a digital factory visualization management system and method based on AI. Background Art

[0002] With the in-depth development of Industry 4.0, digital factories have become increasingly important in the manufacturing industry. However, most current digital factory management systems can only manage a single factory or production line, with serious cross-regional and cross-system collaboration obstacles. The phenomenon of data islands is common, and information between different systems (such as ERP, MES, SCADA, etc.) cannot be effectively shared, resulting in difficulties in achieving optimal resource allocation, limited improvement in production efficiency, and the inability of existing technologies to meet the enterprise's demand for efficient collaborative production management.

[0003] Therefore, there is a need for a digital factory visualization management system and method based on AI to meet the enterprise's demand for efficient collaborative production management. Summary of the Invention

[0004] In view of the deficiencies of the prior art, the present invention provides a digital factory visualization management system and method based on AI to solve the problem of insufficient cross-regional and cross-system collaboration in existing digital factory management, and to achieve optimal allocation of production resources and significant improvement in production efficiency.

[0005] Technical Solution: To solve the above technical problems, according to one aspect of the present invention, more specifically, a digital factory visualization management system based on AI includes a data acquisition layer, a data processing and integration layer, an intelligent decision-making and optimization layer, and a visualization and user interaction layer:

[0006] The data acquisition layer uses Internet of Things devices, intelligent sensors, and device communication protocols to collect data in real time from various factories, production lines, and systems, realizing cross-regional and cross-system data interconnection and interoperability;

[0007] The data processing and integration layer uses natural language processing and deep learning technologies to standardize and convert heterogeneous data, and uses machine learning algorithms to fuse multi-source data, eliminating data islands and ensuring the timeliness and accuracy of data;

[0008] The intelligent decision-making and optimization layer uses intelligent optimization algorithms such as reinforcement learning, genetic algorithms, and ant colony algorithms to dynamically adjust production plans and resource allocations based on real-time data, constructs a prediction and early warning system by analyzing historical and real-time data through AI algorithms, and the AI decision-making engine generates real-time production scheduling and emergency handling solutions in combination with data analysis;

[0009] The visualization and user interaction layer uses AI to generate a multi-dimensional dynamic visualization dashboard to display key data, and through AR technology, it enables managers to remotely view and intervene in the status of each factory, production line, and equipment in real time. A collaborative work platform is built to achieve efficient communication and decision-making among managers in different regions and functions.

[0010] Furthermore, the device communication protocols of the data acquisition layer include OPC-UA and MQTT.

[0011] Furthermore, the data processing and integration layer uses deep learning algorithms to mine potential trends and anomalies in cross-regional production.

[0012] Furthermore, the prediction and early warning system of the intelligent decision-making and optimization layer can send early warnings of equipment failures, production bottlenecks, or demand fluctuations to relevant factories or personnel in advance.

[0013] Furthermore, the collaborative work platform of the visualization and user interaction layer integrates real-time data, task scheduling, and feedback mechanisms.

[0014] According to another aspect of the present invention, more specifically, it is a visualization management method for a digital factory based on AI. Using the visualization management system for a digital factory based on AI described above, it includes the following steps:

[0015] S1. Requirement analysis and planning

[0016] Based on the factory scale, production process, and business requirements, clarify the technical requirements and application scenarios for cross-regional and cross-system collaboration. For the data acquisition layer, plan the deployment locations and quantities of IoT devices and intelligent sensors, and determine the types and sources of data to be collected. At the same time, consider the adaptation of communication protocols for each system; for the subsequent layers, clarify the data processing requirements, intelligent decision-making goals, and key content for visualization display.

[0017] S2. Data collection and preprocessing

[0018] Use the IoT devices and sensors deployed in the data acquisition layer to collect real-time data, and use the AI technology in the data processing and integration layer to clean the collected data, remove noise and outliers, and unify the data format to ensure data quality. Store the processed data in the data warehouse;

[0019] S3. Development and optimization of the intelligent decision-making model

[0020] Based on historical and real-time production data, machine learning algorithms in the intelligent decision-making and optimization layer are used for model development and optimization. The Q-learning algorithm in reinforcement learning is improved. Through the improved algorithm, AI optimization algorithms and prediction models for scheduling optimization and fault prediction are developed. During the actual operation process, model parameters are continuously adjusted according to real-time data to optimize model performance;

[0021] S4. Visualization Interface Design and Deployment

[0022] With the help of the functions of the visualization and user interaction layer, an intelligent visualization interface is developed, including AR / VR interface display and multi-dimensional data visualization dashboard, to present cross-regional collaboration information in real time. When designing, full consideration is given to the convenience of user operation and how to better display the data and analysis results from the data acquisition layer, data processing and integration layer, and intelligent decision-making and optimization layer;

[0023] S5. System Integration and Testing

[0024] Integrate the modules of the data acquisition layer, data processing and integration layer, intelligent decision-making and optimization layer, and visualization and user interaction layer for system joint debugging. Simulate various production scenarios for testing, discover and solve problems in system operation, ensure the stable and efficient operation of cross-regional and cross-system collaboration functions, and focus on testing whether the data interaction and collaborative work between layers are normal;

[0025] S6. Operation and Maintenance and Continuous Optimization: Establish a system operation and maintenance team to monitor the system operation status in real time and handle faults in a timely manner. Regularly collect user feedback, and according to the enterprise's business development and technological progress, update or adjust the equipment in the data acquisition layer, optimize the algorithms in the data processing and integration layer, improve the models in the intelligent decision-making and optimization layer, and upgrade the interfaces in the visualization and user interaction layer to continuously improve system performance.

[0026] Furthermore, in the S2 data collection and preprocessing step, AI technology is used to remove noise and outliers in the data and unify the data format.

[0027] Furthermore, in the S3 intelligent decision model development and optimization step, the core formula of the Q-learning algorithm in reinforcement learning is:

[0028]

[0029] Among them, is the value function of taking action in state , is the learning rate is the reward obtained after taking action , is the discount factor is to take an action is the new state transferred to after that, is the one with the maximum value among all possible actions in the new state To improve the algorithm performance, the learning rate and the discount factor are improved. The improved learning rate is as follows:

[0030]

[0031] wherein, is the initial learning rate, is the current learning step number, is the total learning step number; the improved discount factor is as follows:

[0032]

[0033] wherein, is the initial discount factor, is the adjustment coefficient is the state 's uncertainty entropy; the adaptive discount factor optimizes resources and reduces costs during stable production, ensures short-term delivery during production fluctuations, and improves the system flexibility and responsiveness; the adaptive discount factor discovers potential faults according to short-term data changes when the equipment state is unstable; it accumulates long-term experience during stability to improve prediction accuracy and ensure equipment maintenance and production continuity.

[0034] Furthermore, in the S4 visualization interface design and deployment steps, the visualization system is deployed to the enterprise internal network to ensure that each department can access it in real time.

[0035] Furthermore, in the S6 operation and maintenance and continuous optimization steps, a system operation and maintenance team is established to monitor the system operation status in real time and handle faults in a timely manner.

[0036] The beneficial effects of the digital factory visualization management system and method based on AI of the present invention are as follows:

[0037] (1), The present invention realizes cross-region and cross-system data integration and sharing, breaks data islands, and improves resource allocation efficiency.

[0038] (2), The intelligent decision-making and scheduling based on AI of the present invention significantly improve production efficiency and flexibility and reduce production costs.

[0039] (3), The visualization and real-time monitoring functions of the present invention provide intuitive and comprehensive production information for managers, facilitating quick decision-making.

[0040] (4) The present invention uses blockchain and distributed ledger technologies to ensure data security and privacy and enhance system reliability. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] The present invention will be further described in detail below with reference to the drawings and specific implementation methods.

[0042] Figure 1 It is a schematic structural diagram of a visualization management system for an AI-based digital factory according to the present invention;

[0043] Figure 2 It is a schematic flowchart of a visualization management method for an AI-based digital factory according to the present invention. SPECIFIC IMPLEMENTATION MANNER

[0044] The present invention will be described in detail below with reference to the drawings and embodiments. It should be noted that, without conflict, the embodiments in the present application and the features in the embodiments can be combined with each other.

[0045] To make the technical solution of the present invention clearer, the present invention will be further described in detail below with reference to the drawings and specific embodiments.

[0046] Referring to Figure 1 , Figure 2 , a visualization management system and method for an AI-based digital factory are implemented as follows:

[0047] Requirement analysis and planning stage: Communicate deeply with the production, management and other departments of the enterprise to understand in detail the factory layout, production process, usage of each system and business development plan, determine the functional modules and technical indicators of the system, and formulate a detailed project implementation plan.

[0048] Data collection and preprocessing stage: According to the actual situation of the factory, reasonably deploy IoT devices and sensors to ensure coverage of all key production links and equipment. Use AI technology to clean the collected data, remove noise and outliers, unify the data format, and store it in the data warehouse.

[0049] Intelligent decision-making model development and optimization stage: Collect a large amount of historical production data, use machine learning algorithms for training, and build a production scheduling optimization model and an equipment failure prediction model. During the actual operation process, continuously adjust the model parameters according to real-time data to optimize the model performance.

[0050] Visualization interface design and deployment stage: Design a simple and intuitive visualization interface to display production data through graphics, charts, etc. Use AR technology to develop a mobile application to facilitate managers to view the production site situation at any time and place. Deploy the visualization system to the enterprise internal network to ensure that each department can access it in real time.

[0051] System Integration and Testing Phase: Integrate each module of the data acquisition layer, data processing and integration layer, intelligent decision-making and optimization layer, and visualization and user interaction layer, and conduct system joint debugging. Simulate various production scenarios for testing, discover and solve problems in system operation, and ensure the stability and reliability of the system.

[0052] Operation and Maintenance and Continuous Optimization Phase: Establish a system operation and maintenance team to monitor the system operation status in real time and handle faults promptly. Regularly collect user feedback, update the AI algorithm, optimize the scheduling model according to the enterprise business development and technological progress, and continuously improve the system performance and user experience.

[0053] Working Principle: The present invention is achieved through multi-level technical integration and collaboration around solving the problems of cross-regional and cross-system collaborative management.

[0054] Data Acquisition Principle: As the starting point of the system operation, the data acquisition layer uses Internet of Things (IoT) devices, intelligent sensors, and specific device communication protocols (such as OPC-UA, MQTT) to obtain data. IoT devices and intelligent sensors are widely deployed at key nodes in various factories, production lines, and different systems. For example, sensors are installed on production equipment to monitor the operating parameters of the equipment in real time, such as temperature, rotation speed, vibration, etc.; material sensors are set on the production line to sense the flow and inventory of materials. These sensors convert the collected physical signals into electrical signals or digital signals. Communication protocols such as OPC-UA and MQTT are responsible for transmitting these signals to the data processing and integration layer in a standardized and reliable manner. Its working mechanism is that sensors collect data at a preset frequency, and after the data is encapsulated by the communication protocol, it is transmitted in the form of data packets through wired or wireless communication networks, ensuring that data can be collected in real time and accurately in a cross-regional and cross-system environment.

[0055] Data Processing and Integration Principle: The data processing and integration layer receives heterogeneous data from the data acquisition layer. Since there are differences in data formats, encoding methods, semantics, etc. of different data sources, natural language processing (NLP) and deep learning (DL) technologies in AI play a key role here. NLP technology processes data containing text information through means such as text analysis and semantic understanding. For example, it parses the text descriptions in device logs and extracts key information. DL technology uses neural network models to extract features and identify patterns from various types of data such as images and numerical values, and converts non-standard format data into a unified and recognizable format. Subsequently, machine learning algorithms intervene to fuse multi-source data. By constructing a data fusion model, according to factors such as the timestamp, source, and correlation of the data, data from different channels are integrated to eliminate data islands. For example, correlate and analyze the operation data of production equipment with the material inventory data to ensure the timeliness and accuracy of the data and provide a reliable data basis for subsequent intelligent decision-making.

[0056] Principles of Intelligent Decision-making and Optimization: The intelligent decision-making and optimization layer relies on intelligent optimization algorithms such as reinforcement learning (RL), genetic algorithm (GA), and ant colony optimization (ACO). These algorithms take real-time data as input and dynamically adjust production plans and resource allocations. Taking reinforcement learning as an example, the system regards the production environment as a state space and different production decisions (such as equipment start / stop, task assignment) as actions. The algorithm continuously tries different actions and adjusts the decision-making strategy according to the obtained rewards (such as increased production efficiency, reduced costs). In terms of equipment failure prediction, AI algorithms analyze a large amount of historical and real-time data to build a prediction model. For example, using deep learning algorithms to perform time series analysis on equipment operation data, learning the characteristics of normal operation patterns and failure patterns. When the real-time data deviates from the normal pattern by a certain threshold, the prediction and early warning system issues an equipment failure warning. The AI decision-making engine combines the data analysis results to generate real-time production scheduling and emergency handling plans to achieve the optimal allocation of resources across regions and factories and maximize production efficiency.

[0057] Principles of Visualization and User Interaction: The visualization and user interaction layer uses AI to generate multi-dimensional dynamic visualization dashboards. AI algorithms analyze the processed data and present key data such as production line status, equipment operation conditions, and inventory levels in intuitive graphs (such as bar charts, line charts), diagrams (such as Gantt charts), etc. according to data characteristics and user needs. The application of AR technology enables managers to view the status of each factory, production line, and equipment in real time through mobile devices or head-mounted display devices. Managers can remotely intervene and adjust the production process through interaction methods such as gestures and voices. The collaborative work platform integrates real-time data, task scheduling, and feedback mechanisms, enabling managers in different regions and functions to share information, issue tasks, and receive feedback on this platform, achieving efficient communication and decision-making and improving cross-regional collaborative management efficiency.

[0058] In the entire workflow, from data collection, processing, decision-making to visualization and interaction, each link is closely connected, forming a closed-loop intelligent management system, continuously optimizing the cross-regional and cross-system collaborative management of the digital factory, and achieving the efficient allocation of production resources and the improvement of production efficiency.

[0059] The above-described embodiments merely represent several implementation manners of the present invention. Their descriptions are relatively specific and detailed, but should not be construed as limiting the scope of the invention patent of the present invention. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present invention, several modifications and improvements can still be made, and these all belong to the protection scope of the present invention. Therefore, the protection scope of the invention patent of the present invention shall be subject to the appended claims.

Claims

1. An AI-based digital factory visualization management system, including a data acquisition layer, a data processing and integration layer, an intelligent decision-making and optimization layer, and a visualization and user interaction layer, characterized by: The data collection layer uses IoT devices, smart sensors, and device communication protocols to collect real-time data from various factories, production lines, and systems, enabling cross-regional and cross-system data interconnection and interoperability. The data processing and integration layer uses natural language processing and deep learning technologies to standardize and convert heterogeneous data, and uses machine learning algorithms to fuse multi-source data, eliminating data silos and ensuring data timeliness and accuracy. The intelligent decision-making and optimization layer uses intelligent optimization algorithms such as reinforcement learning, genetic algorithms, and ant colony algorithms to dynamically adjust production plans and resource allocation based on real-time data. It uses AI algorithms to analyze historical and real-time data to build a prediction and early warning system. The AI decision engine combines data analysis to generate real-time production scheduling and emergency response plans. The visualization and user interaction layer uses AI to generate multi-dimensional dynamic visualization dashboards to display key data, and uses AR technology to enable managers to remotely view and intervene in the status of each factory, production line and equipment in real time, and build a collaborative work platform to achieve efficient communication and decision-making among managers in different regions and functions.

2. The AI-based digital factory visualization management system according to claim 1, characterized in that: The device communication protocols of the data acquisition layer include OPC-UA and MQTT.

3. The AI-based digital factory visualization management system according to claim 1 is characterized by: The data processing and integration layer uses deep learning algorithms to explore potential trends and anomalies in cross-regional production.

4. The AI-based digital factory visualization management system according to claim 1, wherein: The prediction and early warning system of the intelligent decision-making and optimization layer can issue early warnings of equipment failures, production bottlenecks or demand fluctuations to relevant factories or personnel.

5. The AI-based digital factory visualization management system according to claim 1, characterized in that: The collaborative work platform of the visualization and user interaction layer integrates real-time data, task scheduling and feedback mechanisms.

6. A visualization management method for a digital factory based on AI, using a visualization management system for a digital factory based on AI as described in claim 1, characterized in that, The following steps are involved: S1. Demand Analysis and Planning Based on the factory scale, production processes, and business needs, clarify the technical requirements and application scenarios for cross-regional and cross-system collaboration. For the data collection layer, plan the deployment location and quantity of IoT devices and smart sensors, and determine the type and source of the required collected data, while considering the adaptability of the communication protocols of each system. For subsequent layers, clarify the data processing requirements, intelligent decision-making goals, and key contents of visualization. S2. Data Collection and Preprocessing The IoT devices and sensors deployed in the data collection layer collect real-time data. AI technology in the data processing and integration layer is used to clean the collected data, remove noise and outliers, and unify the data format to ensure data quality. The processed data is then stored in the data warehouse. S3. Development and optimization of intelligent decision-making models Based on historical and real-time production data, we use machine learning algorithms in the intelligent decision-making and optimization layer to develop and optimize models. We also improve the Q-learning algorithm in reinforcement learning. With this improved algorithm, we develop AI optimization algorithms and prediction models for scheduling optimization and fault prediction. During actual operation, we continuously adjust model parameters based on real-time data to optimize model performance. S4. Visual interface design and deployment With the functions of the visualization and user interaction layer, an intelligent visualization interface is developed, including AR / VR interface display and multi-dimensional data visualization dashboards, to present cross-regional collaboration information in real time. When designing, full consideration is given to the convenience of user operation and how to better display the data and analysis results from the data collection layer, data processing and integration layer, and intelligent decision-making and optimization layer; S5. System Integration and Testing Integrate the modules of the data collection layer, data processing and integration layer, intelligent decision-making and optimization layer, and visualization and user interaction layer, and conduct system joint debugging. Simulate various production scenarios for testing, discover and solve problems in system operation, ensure the stable and efficient operation of cross-regional and cross-system collaboration functions, and focus on testing whether the data interaction and collaborative work between layers are normal; S6. Operation and Maintenance and Continuous Optimization: Establish a system operation and maintenance team to monitor the system operation status in real time and handle faults in a timely manner. Regularly collect user feedback, and according to the business development and technological progress of the enterprise, update or adjust the equipment in the data collection layer, optimize the algorithms in the data processing and integration layer, improve the models in the intelligent decision-making and optimization layer, and upgrade the interfaces in the visualization and user interaction layer to continuously improve the system performance.

7. The AI-based digital factory visualization management method according to claim 6, characterized in that: In the S2 data collection and preprocessing step, AI technology is used to remove noise and outliers in the data and unify the data format.

8. The AI-based digital factory visualization management method according to claim 6, wherein: In the S3 intelligent decision-making model development and optimization step, the core formula of the Q-learning algorithm in reinforcement learning is: ； Among them, is the state to take action of the value function, is the learning rate is to take action the reward obtained after is the discount factor is to take action the new state transferred to after is the one with the maximum value among all possible actions in the new state To improve the algorithm performance, the learning rate and the discount factor are improved. The improved learning rate is: ； in, is the initial learning rate, is the current number of learning steps, is the total number of learning steps; the improved discount factor for: ； Among them, is the initial discount factor, is the adjustment coefficient is the state of the uncertainty entropy; The adaptive discount factor optimizes resource cost reduction during stable production, ensures short-term delivery during production fluctuations, and improves system flexibility and responsiveness; When the equipment state is unstable, the adaptive discount factor discovers potential faults according to short-term data changes; When stable, it accumulates long-term experience, improves prediction accuracy, and ensures equipment maintenance and production continuity.

9. The AI-based digital factory visualization management method according to claim 6, characterized in that: In the S4 visualization interface design and deployment step, the visualization system is deployed to the enterprise internal network to ensure that each department can access it in real time.

10. The AI-based digital factory visualization management method according to claim 6, wherein: In the S6 operation and maintenance and continuous optimization step, a system operation and maintenance team is established to monitor the system operation status in real time and handle faults in a timely manner.

Images