Research on intelligent improvement technology of glass equipment based on visualization.

With the increasing market demand for glass products, the output and quality of glass products continue to improve, which puts forward higher requirements for the manufacture of glass processing equipment. The production process of glass processing equipment involves multi-disciplinary and multi-field technical knowledge. It is particularly important to visually present and analyze the process parameters and processing status information involved in it through visual means. As a new technology, visual analysis technology is widely used in the fields of machinery, electronics, control, and automation, and has been widely used in industrial design. With the continuous development and maturity of computer vision technology and visual analysis technology, the application of visual analysis technology in industrial design has become more and more in-depth, and visualization research has gradually become an important means for industrial designers to carry out research work. Display design products, process parameters, and processing status information in three-dimensional space, so that people can intuitively see the appearance structure and various parts of the product, and can effectively manage and process the data, improving the work efficiency of designers. LIJIANG Glass mainly started from the field of industrial design, and briefly introduced visual analysis technology and related literature review; focused on improving the design quality and reducing production costs of domestic glass manufacturing equipment process parameters and processing status information through visual analysis. Do some research, and discuss the future development trend of the glass equipment industry in combination with relevant research results at home and abroad.

1. Literature review

1.1 Visual Analysis and Design

The concept of visual analysis and design was first proposed by German scholar Boltz. Its essence is to change the product design process from static description to dynamic design. It is a method of analyzing, evaluating, and optimizing products, and then optimizing their appearance, function, and performance. From the perspective of industrial design (Industrial Design), Charles defined it as "theoretical research and application based on the human cognitive process, so that computers can understand the content expressed by human beings and organize, explain or predict their behavior and behavior in a certain way." A theoretical approach to the results". Andrew In the field of glass processing equipment, visual analysis, and design are mainly used in equipment function and structure analysis, appearance visual effect design, process parameter optimization, state evaluation and product quality control, etc. More intelligent mechanical experts believe that visual analysis mainly includes mechanical structure analysis, optical characteristic analysis, and material performance analysis; while visual design mainly includes the product's overall shape and appearance color design, etc.

1.2 Condition monitoring and fault diagnosis of glass equipment on the production line

Glass equipment working on the production line not only needs to maintain and maintain its internal key components such as sealing elements and transmission components, but also needs to carry out fault diagnosis and testing regularly to ensure normal and orderly production. Some intelligent mechanical experts propose that manual inspection or online monitoring is usually used for the leakage status monitoring of sealing elements and the fault diagnosis and maintenance of transmission components. To better realize intelligent maintenance, it is necessary to detect defects on key components and arrange maintenance tasks in time according to the defect detection results. A glass processing equipment health monitoring system based on the combination of machine vision technology and deep learning technology is also proposed. With an intelligent algorithm as the core, the motion posture analysis method based on machine vision is used as the main technical means, combined with industrial sensor networks and industrial big data. The real-time data collection and processing capabilities provided by the data platform can automatically identify the processing trajectory of glass processing equipment and monitor the online operation status. In addition, some intelligent mechanical experts analyzed the influence of the combination of glass processing equipment motion sensor array and signal acquisition module on its measurement accuracy, measurement stability, dynamic response performance, and other indicators through experiments, and proposed a network model based on deep learning Methods for fault diagnosis and identification of moving parts.

1.3 Analysis of process parameters and processing status of glass processing equipment

Glass processing equipment can analyze its process parameters and processing status through visual analysis methods, to better understand the production process of glass products and guide glass production. Some intelligent mechanical experts have proposed that the manufacturing process and each process of glass processing equipment can be visually displayed and analyzed using visualization. Through visual analysis means, data such as process parameters and processing status can be processed, and intuitive graphics can be formed, which is convenient observe and understand. At present, there are many software that can apply visual analysis methods and realize visual analysis and processing. However, due to the complex structure and bulky size of glass processing equipment, most of the software is developed for specific equipment or a specific process. Other intelligent machinery experts pointed out that for different types, multiple processing equipment, and processes, visual analysis methods are not the only feasible solution. Through the collection and preprocessing of visual data, it is possible to obtain information such as different types of glass processing equipment, process parameters, and the processing status of corresponding processing links based on a large amount of data. Visual analysis can help us intuitively understand the operation process of the glass automatic production line, especially the automatic insulating glass production line, and further understand its process parameters and the interrelationships between the various processes. At present, glass production and processing enterprises in developing countries have little research on this aspect, but more in-depth and comprehensive research on it will have a profound impact on the development of the glass automation industry.

1.4 3D visualization

3D visualization is the three-dimensional and intuitive reproduction of complex products so that users can have a better understanding of their processing process in a visual situation, to arrange production plans more reasonably. In the whole life cycle of glass equipment, the production and processing process, process parameters and processing status of glass equipment can be intuitively understood through 3D visualization means, which provides a decision-making basis for improving the quality of glass product manufacturing and reducing costs. 3DMAX is a rapid modeling tool, which can help designers quickly model, and provide intuitive visual data analysis and interactive results for subsequent process adjustments and improvements. Some experts believe that with the continuous expansion and deepening of the application scope of 3DMAX, many problems will be encountered in its visual analysis and optimization, such as inconsistent performance of 3DMAX data in different environments, the inability to obtain 3D surface models in real-time, and the difficulty of quickly modifying 3D models wait. The key problem to be solved in the visualization process of glass processing equipment is to efficiently process and analyze various types of information. Other glass processing equipment practitioners have proposed that the integration of visual information and analysis technology will provide more efficient and accurate visual support for process parameters and processing status in the whole life cycle of glass processing equipment.

2. Research process

2.1 Main technical content

2.1.1 Infrastructure

The basic environmental equipment (PLC) of the glass production line communicates with the data acquisition station (data PC), and the acquired data information is passed through the OPC-UA communication protocol to improve accuracy and reliability. Pass the valid data after edge computing to the upper data layer through the mutually authenticated OPC-UA protocol. The data collection station (data PC) acts as a data bridge, responsible for collecting raw data and preprocessed information, performing simple edge computing, and eliminating invalid upper-level local data servers or cloud data servers.

1. Cloud deployment: The cloud deployment method supports mainstream cloud service providers such as Amazon Cloud, and also supports user-built private clouds, and uses 4G/5G mode for data upload. The establishment of a cloud platform is of great value for mobile applications, massive data storage, and flexible expansion of servers. At the same time, the data security technology of cloud service providers can be used to ensure the security of data storage, while reducing hardware costs and subsequent maintenance costs.
2. Local privatization deployment: According to the actual needs of users, localization deployment is supported, and the user's glass equipment data is transmitted to a fixed private server cluster through the Internet to achieve unified management of glass equipment.

2.1.2 Intelligent Management and Control

Data analysis, state prediction, and intelligent management of glass equipment are aimed at pushing the operating status of glass equipment to the visualization terminal. The specific process is as follows:

1. Intelligent management: equipment management, material management technical supervision, operation management;
2. Intelligent analysis: data control, fault warning, intelligent monitoring, intelligent prediction;
3. Intelligent network: data analysis, intelligent inspection operation assistance.

2.1.3 Unified display - various types of human-computer interaction interfaces

Using rich graphics and image methods, combined with three-dimensional space expression means and intelligent control, through the image of a professional rendering engine to three-dimensionally realize the integrated platform of data multi-dimensional visualization display, interaction, and design, and create a full range of big data visualization decision-making Center to help users intelligently perceive the situation of glass equipment and scientific decision-making analysis.

2.2 The main technical difficulties and solutions

2.2.1 Data collection and establishment of data structure

1. Data collection: Multi-source data integration is carried out by collecting control system data, equipment inspection data, process-related data, etc. The method of integration is to import control system/equipment monitoring historical data, access real-time data, enter/upload fault notification and analysis data, and combine data related to process design, through convenient human-computer interaction, to realize the close connection between data and industry knowledge and experience Combined, it provides a solid foundation for subsequent analysis.
2. Data collection layer: Collect relevant data at the bottom layer through interfaces and devices.
3. Data storage layer: The data storage layer mainly completes the real-time storage function of field data. Historical data provides data support for applications through data processing.
4. External interface layer: through a unified interface platform, the factory database provides the integrated and calculated data to other application windows.

2.2.2 Glass equipment data development

Using non-linear multivariate dynamic correlation modeling technology, through big data feature mining, learning, dynamic model optimization, and knowledge base management, two sets of models are developed, namely an early warning model for operating status and an intelligent health diagnosis model.

1. Early warning model of operating status: use various types of data generated in the process of equipment production control to analyze the health status of important equipment, including multi-dimensional correlation analysis of equipment data, horizontal and vertical comparative analysis of equipment operating status, and based on actual use And equipment health assessment of operating conditions, etc., to provide auxiliary support and reliable basis for equipment management decision-making.
2. Intelligent equipment health diagnosis model: Based on historical fault records, online data analysis of the knowledge base provides users with suggestions for solving difficult faults, and self-improvement by continuously updating the knowledge base, as shown in Figure 1.

^{Figure 1 Data modeling and troubleshooting process diagram}

2.2.3 Establish a device management library

Based on the development of glass equipment data, combined with glass equipment and technology, through the probability of equipment failure and the severity of the consequences of equipment failure, intelligently establish equipment risk assessment and establish a 4-level equipment management library to realize glass equipment. Carry out scientific procurement management of spare parts. Level I warehouses are extremely risky, factory-level inventory of spare parts and operating equipment must be inspected irregularly; n-level warehouses are risky, enterprise-level inventory of spare parts and operating equipment must be checked regularly; disk-level warehouse equipment risks are average, and "cloud" inventory spare parts. No spare parts or fewer spare parts may be stored in the W-level library for routine inspection of operating equipment.

2.2.4 Full life cycle monitoring of glass equipment (intelligent command center)

The whole life cycle monitoring of glass equipment mainly includes 5 parts.

1. Platform management: including equipment model, project model, user and role, mail service, SMS service, system log, platform general management, and other functions;
2. Authority management: adopt the authority management mechanism based on user roles;
3. Equipment management: real-time equipment information. Support glass equipment 3D modeling, real-time data monitoring, data log analysis, audit trail, switch records, equipment alarms, equipment maps, and network topology;
4. Event warning: customize alarm and early warning information in combination with glass production business systems and equipment;
5. Web configuration and application: A web cloud configuration function is provided, and exclusive vector graphics components can be customized.

2.2.5 Functional Classification

1. Alarm classification: manage the priority classification of the alarm content, and set the response mechanism, duration, responsible person, and other information
2. Contact management: Precisely bind the relationship between maintenance and the specific person in charge, contact information, and set the maintenance response time. When an alarm occurs, the maintenance personnel will be notified to prepare tools and spare parts in time to rush to the site of the alarm equipment through message push to improve maintenance efficiency;
3. Locate faults and alarm points: sort out and code faults and alarms based on classification, and analyze parameter abnormalities during fault periods according to fault period feature location technology, quickly locate faults and alarm points, and combine combing and coding to quickly Provide operation and maintenance schemes and methods to improve operation and maintenance efficiency and reduce operation and maintenance costs;
4. Predictive operation and maintenance: Abnormality analysis and early warning of glass equipment adopt abnormal feature detection technology, which can automatically detect the abnormal state corresponding to the abnormal data from the correlation change of glass equipment sensor data, issue operation, and maintenance instructions in advance, reduce Unscheduled downtime increases the utilization rate of glass equipment.

3. Economic and social benefits

At present, in the field of glass processing equipment, the system construction method based on the automatic identification of process parameters is mainly used to realize the visualization of process parameters and status information. This research is accompanied by the development of automatic identification technology of process parameters to the direction of intelligence and visualization, which can effectively reduce the cost of manual work and improve the degree of automation and production efficiency. In the future, the field of glass processing equipment will further improve and optimize the visualization technology, so that it can better serve the intelligent upgrading and transformation of the glass industry, and promote the high-quality development of automation equipment.

4. Conclusion

With the development of computer technology, visual design, and analysis technology has become one of the hot spots in the field of industrial design. It not only provides better theoretical support for the intelligent design and development of glass processing equipment but also provides a basis for the development and appearance design of glass products. And performance analysis provides richer and more convenient data sources, thereby improving the efficiency of designers and product development. This research mainly introduces the large-screen visualization system and its functional modules and algorithms, realizing the visualization of glass processing equipment. Through the integration of computer vision technology and visual analysis technology, not only the rapid construction, virtual assembly, and simulation of complex 3D solid models and structures can be realized, but also the visual output of process parameters and processing status in the whole life cycle of glass equipment can be realized to meet different needs. Customers' customization requirements can also improve production efficiency. However, in the actual application process, because the design and construction of the traditional visualization system are based on the integration of hardware, software, and services, it takes a lot of time, cost, and financial resources, which leads to the ineffectiveness of the visualization system in practical application. With the rapid development of computer technology and related technologies (such as image processing, speech recognition, semantic recognition, etc.), the visual analysis system will gradually become an indispensable tool in the intelligent development and research of modern glass manufacturing equipment.

^{For more information about LIJIANG Glass insulated glass production line and insulated glass producing machine, please click here to learn more}