Home >

Professor Wen Lisheng Of Donghua University: Digital Twin Is The Base Of Intelligent Garment Manufacturing

2022/5/10 15:01:00 0

한국어 日本語 Français English 中文

Digital Intelligent Manufacturing Wen Lisheng

China garment Intelligent Manufacturing Alliance

Deputy leader of expert group

Professor Wen Lisheng, Donghua University

1. We recognize the digital twin

In the 1960s, NASA carried out the Apollo program. The purpose of this program is to realize the manned lunar landing flight and the field investigation of the moon. After the successful implementation of the Apollo program, NASA released the area 11 technology roadmap in 2010 The concept of digital twins is put forward for the first time 。 It is defined as: A digital twin is an integrated simulation of a vehicle or system with multiple physical quantities and spatial scales. The simulation uses the most effective physical model, update of sensor data, flight history, etc. to mirror the survival state of the corresponding twin in flight 。

In 2002, Professor Michael grives of the University of Michigan presented a slide presentation to industry for the establishment of PLM (product life cycle management) center, The concept model of PLM is put forward for the first time, From a virtual space and a virtual space to a real-world data stream.

Figure 1 the original conceptual model of digital twin and its terminology

Gartner describes digital technology as one of the most important twin of digital technology in the world.

The definition recommended by cimdata is: Digital twin (i.e. digital clone) is a system description based on physical entities, which can realize the creation, management and application of data, models and information from trusted sources across the whole life cycle of the system.

The definition of Professor Tao Fei of China Beihang University is: as a key way to realize bidirectional mapping, dynamic interaction and real-time connection between virtual reality and reality, digital twin can map the attributes, structure, state, performance, function and behavior of physical entities and systems to the virtual world, forming a high fidelity dynamic multi-dimensional / multi-scale / multi-physical quantity model It provides an effective means to understand, control and transform the physical world.

In the white paper on digital twin Technology (2019) of China in 2019, the definition of digital twin is as follows: Digital twin is a digital model of existing or existing physical objects, which can perceive, diagnose and predict the state of physical objects in real time through measurement, simulation and data analysis, The behavior of physical objects is regulated through performance and state optimization and instruction sending, and self evolution is achieved through mutual learning among relevant digital models, and the decision-making of stakeholders in the life cycle of physical objects is improved.

In 2020, the white paper on the application of digital twins issued by the Institute of standardization of the Ministry of industry and information technology of China defines that digital twins are virtual entities that create physical entities in a digital way, and simulate, verify, predict and control the whole life cycle process of physical entities with the help of historical data, real-time data and algorithm models.

It can be seen that there are many different understandings and understandings on the above digital twin, and there is no unified consensus definition yet. But from these definitions, we more agree with the definition given by the Standardization Institute of the Ministry of industry and information technology, which is more accurate, comprehensive and easy to understand.

Structure and composition of digital twin system

The overall structure of digital twin system is shown in Fig. 2. This architecture is provided by the Standardization Institute of MIIT in 2020, which can guide the construction and implementation of digital twin.

Figure 2 Architecture of digital twin system

As can be seen from Figure 2, the overall architecture is divided into two parts: the digital twin system framework and the physical objects. The upper part is the digital twin system framework, and the lower part is the physical object. The physical object is used to express the relationship between the digital twin world and the physical world. It is not an element of the digital twin frame 。 Digital twin architecture includes the following elements: digital modeling, measurement and control, simulation, data analysis, digital assets and human-computer interaction.

As a manufacturing industry, if you want to understand the digital twin system architecture more clearly, see Figure 3. The physical object in Figure 2 is what we usually call the real world or physical world (or physical space). In the manufacturing industry, it can be understood as the enterprise physical world composed of products, equipment, production lines, materials, people and other elements, while the framework of digital twin system can be understood as modeling, simulation, simulation, and other elements Data analysis and other technologies build digital twin virtual world (or cyber physical space). The two worlds are connected by the network, and the real-time data mapping and feedback control are carried out, so as to achieve the optimal production of real-world enterprises.

Source: advanced manufacturing Figure 3 Twin system

The key technology and characteristics of digital twin

(1) the key technology of digital twinning

The key technologies of manufacturing oriented digital twin include modeling, simulation, Internet of things, big data, artificial intelligence, cloud computing, digital clues, edge computing, virtual reality (VR / Ar / MR), MES, GIS, CAD, CAE, cam, industrial control, blockchain, etc., as shown in Figure 4.

Figure 4 Key technologies of digital twin

From these technologies, most of the key technologies of digital Twin are related to new information technology (newIt), as shown in Figure 5. NewIt plays an important role in the realization and application of digital twin 。 For example, the Internet of things can provide help for real-time, reliable and efficient transmission of twin data through wired or wireless networks; VR technology uses computer graphics, detail rendering and dynamic environment modeling to realize the visualization and dynamic display of physical entity attributes, behaviors and rules; AR and Mr technologies use real-time data acquisition, scene capture, real-time tracking and registration to realize the synchronization and integration of virtual model and physical entity in space-time, and enhance the function of physical entity in detection, verification and guidance through virtual model; The edge computing technology can filter, regulate and process the data collected from the physical world in real time on the edge side, so as to realize the user's local instant decision-making, quick response and timely execution; System level and complex system level digital twins need more computing and storage capacity. Cloud computing can be used on demand and distributed sharing mode, which can dynamically meet the different computing, storage and operation requirements of digital twins when using huge cloud computing resources and data centers; Big data can extract more valuable information from the massive data generated by digital twin to interpret and predict the results and processes of real events; The blockchain can provide a reliable guarantee for the security of digital twins, and ensure that the twin data can not be tampered with, marked in the whole process, traceable, traceable, etc; Artificial intelligence can automatically perform data preparation, analysis and fusion without the participation of data experts through intelligent matching optimal algorithm, and carry out deep knowledge mining for twin data, so as to generate various types of services.

Source: the relationship between digital twin and newIt

The premise of digital twin application is the model of each link and a large amount of data. How to generate, exchange and transfer the data similar to product design, manufacturing, operation and maintenance? How to realize the seamless flow of data between some relatively independent systems? These are exactly what digital cues are trying to solve. Therefore, digital clue technology is also one of the key technologies of digital twin. Only by understanding the digital clue can we understand the digital twin more deeply. The so-called digital clues are sometimes called digital threads, digital ties, digital mainlines, etc, In the past, there was a whole life cycle of data integration between cimdata and cimdata 。 It can be seen that digital clues can connect the digital twins of different life cycles to form the automatic flow of data. Therefore, digital clues are a bridge between the real physical world and the digital virtual world, as shown in Figure 6.

Six sources of advanced manufacturing industry

(2) the typical and ideal characteristics of digital twins

Digital twins have the following typical characteristics:

(1) interoperability ——The physical objects and digital space in digital twins can be bi-directional mapping, dynamic interaction and real-time connection. Therefore, digital twins have the ability to map physical entities with various digital models, and have the ability to transform, merge and establish the equivalence of expression between different digital models;

(2) scalability ——Digital twin technology has the ability to integrate, add and replace digital models, and can expand the content of multi-scale, multi-physical and multi-level models;

(3) real time ——Digital twin technology requires digitization, that is, to manage data in a way that can be recognized and processed by computers to represent physical entities that change with time axis. The objects represented include appearance, state, attribute and internal mechanism, forming the digital virtual body mapping of real-time state of physical entities;

(4) fidelity ——Fidelity of digital twins refers to the proximity between the digital virtual body model and the physical entity. It is required that virtual bodies and entities should not only keep the high degree of simulation of geometric structure, but also be simulated in the state, phase state and temporal state;

(5) closed loop ——The digital virtual body in digital twins is used to describe the visual model and internal mechanism of physical entities, so as to monitor, analyze and reason the state data of physical entities, optimize process parameters and operation parameters, and implement decision-making function, that is, to endow digital virtual bodies and physical entities with a brain.

The understanding and practice of digital twins can not be separated from specific objects, specific applications and specific needs, such as a product, an equipment, a production line, a workshop, etc. Their digital twins in different stages will show different characteristics, which requires us to start from the application and solve practical problems. In the process of application, it is not necessary to establish "digital twins" with all "ideal features", as long as they can meet the needs of users. Table 1 is the ideal characteristics of digital twin based on various perspectives, which is worthy of our reference in application.

Table 1 ideal characteristics of digital twins based on different understandings

Digital twin is the base of intelligent manufacturing

The core technology of CPS is the digital production system. Information physical fusion production system is the application of information physical system (CPS) in the production field, and it is a multi-dimensional intelligent technology system. By means of integration of computing, computing, intelligent computing, intelligent computing and intelligent production, the three technologies are integrated with each other. The information obtained is combined with the physical performance representation of the object to form the deep integration, real-time interaction, mutual coupling and timely updating of virtual space and entity space, and the virtual image of entity is constructed in cyberspace. Through the operation and analysis of self perception, self memory, self cognition, self decision and self reconstruction, the digital, intelligent and networked production system is realized.

Source: intelligent manufacturing im Figure 7 digital twin and intelligent manufacturing

Intelligent manufacturing system based on digital twin is a predictive intelligent manufacturing system. As we all know, the trouble of intelligent manufacturing is the failure of equipment and devices. As time goes on, the equipment will wear out and its performance will decline, eventually leading to failure and shutdown. By mastering the actual state of equipment, predictive intelligent manufacturing adopts planned maintenance instead of rush repair after failure or unnecessary replacement of parts that can continue to be used, so as to reduce maintenance costs and production costs. Knowing when the equipment may fail can reasonably arrange the maintenance plan, realize "on time" maintenance, maximize the availability of equipment and extend its normal operation time, and improve the operation efficiency of the plant. The equipment degradation mode and real-time state evaluation are combined with process control to ensure the stability of product quality and move towards the worry free intelligent manufacturing when the performance of equipment or system changes with time.

From "informatization drives industrialization, industrialization promotes informatization", to the development strategy of "integration and evolution" of "made in China 2025", China expects to realize intelligent manufacturing through information physical integration. Digital twin is not only the basis of building CPS system, but also an enabling technology of intelligent manufacturing. It can be said that without digital twin, there will be no CPS and no intelligent manufacturing. Digital twin is the "base" of Intelligent Manufacturing in the future, which will have a bright future in the practice of intelligent manufacturing.

5. Digital twin approach of intelligent garment manufacturing

Gartner, together with Shugen Internet, once published the white paper "how to use digital twins to help enterprises create value" (hereinafter referred to as "white paper") in the industrial Internet industry. In the white paper, it is predicted that digital twin will be an important driving force for enterprises to realize transformation and create cost value in the future. By 2021, half of the large manufacturing industries will use digital twins, which will increase the efficiency of these enterprises by 10%. By 2024, more than 25% of the new digital twins will be adopted as functions of new lot business applications.

The reason why we pay so much attention to digital twin technology is that it can be directly applied in intelligent garment manufacturing factory. Digital twin technology is also an important part of garment intelligent factory practice. We use the digital twin technology to realize the intelligent manufacturing of clothing. The way is as follows: starting from the digital virtualization of the physical garment processing equipment (i.e. the establishment of simulation model), with the continuous development and application of new IT technologies such as artificial intelligence, big data, cloud computing, etc., digital twin will gradually change from equipment digital virtualization to product line digital virtualization The development of digital factory. In the process of digital virtualization of these physical entities, it is necessary to generate data resource database independently through repeated simulation calculation, and use artificial intelligence technology such as deep learning to gradually realize the self-adaptive and self decision-making of digital twins for physical equipment and production line process. When the real production situation changes, it can realize the real-time interaction between the virtual production scene and the real-time production.

Figure 8 intelligent manufacturing concept planning of digital twin Technology

The physical equipment (physical world) in Figure 8 can be all kinds of equipment, production line, workshop and factory for garment processing, while the simulation model (digital virtual world) is a 3D digital virtual entity for equipment, production line, workshop and factory; Figure 9 is a digital twin system for robot equipment commonly used in our garment intelligent manufacturing. This system is for predictive health maintenance of robot equipment. Its principle is shown in Figure 10; Whether it is equipment, production line or factory, the application principle of digital twin technology is the same, but there are differences in hierarchy and value, so I will not repeat it here. The operation mechanism of digital twin clothing workshop is shown in Figure 11; The conceptual model of digital twin in garment factory is shown in Figure 12; The digital twin clothing stereoscopic warehouse is shown in Figure 13.

Source: Engineering 2021 Figure 9 Robot Digital twin system

Source: advanced manufacturing Figure 10 equipment health digital twin system

Source: zhizaoyuan Figure 11 operation mechanism of digital twin clothing workshop

Figure 12 digital twin concept of garment factory

Source: zhizaoyuan picture 13 digital twin clothing stereoscopic warehouse

The market research conducted by China Society of mechanical engineering in 2020 shows that:

As companies continue to learn how to use digital twins to improve productivity and reduce costs, digital twins will become more and more popular in the future. By 2025, up to 89% of IOT platforms will include digital twins; By 2027, digital twin will become the standard function of IOT platform; After nearly 36% of executives in various industries have learned about digital twin technology, 50% of them will use digital twin in their business by 2028. Our clothing manufacturing industry, of course can't Backward.