This paper makes an in-depth analysis and reflection on the following ten questions, with a view to providing references for researchers to better understand digital twins, for decision makers to treat digital twins rationally and correctly, and for practitioners to better apply digital twins.

① What is a digital twin?

② Who is paying attention to digital twins?

IC660BRD021 Digital twins: Who is hotter, China, the United States or Germany?

What is the relationship between digital twins and intelligent manufacturing?

Can digital twins be integrated with the New generation of information technology (New IT)?

⑥ Are there scientific problems with digital twins?

⑦ What is the use of digital twins?

⑧ What are the applicable criteria for digital twins?

⑨ Do digital twins need standards?

⑩ Do digital twins require commercial tools/platforms?

What is a digital twin?

At present, more and more scholars and enterprises pay attention to digital twins and carry out research and practice, but from different angles, there are different understandings of digital twins. Starting from different dimensions, this section summarizes and analyzes the current understanding of digital twins, and tries to discuss the ideal characteristics of digital twins for reference.

(1) Model dimension One view holds that digital twins are three-dimensional models, copies of physical entities, or virtual prototypes. These understandings focus on the model dimension of digital twins from the perspective of model requirements and functions. Based on the analysis of existing literature, the IC660BRD021 ideal digital twin model involves multi-dimensional, multi-space-time and multi-scale models such as geometric model, physical model, behavior model and rule model, and it is expected that the digital twin model has the characteristics of high fidelity, high reliability and high precision, so that it can truly depict objects

Manage the world. In addition, different from the traditional model, the digital twin also emphasizes the interaction between virtual and real, and can be updated and dynamically evolved in real time, so as to realize the dynamic real mapping of the physical world.

(2) Data dimension Professor Grieves once put forward the concept related to digital twins in the product Lifecycle management (PLM) course at the University of Michigan, so there is a view that digital twins are PLM. Similarly, there are arguments for Digital twins as data/big data, Digital Shadow, or DigitalThread. These insights focus on the value of digital twins in product lifecycle data management, data analysis and mining, data integration and fusion. Data is the core driving force of digital twin, digital twin data not only includes the whole factor/whole process/whole business data throughout the product life cycle, but also emphasizes the integration of data, such as information physical virtual-real integration, multi-source heterogeneous integration. In addition, the digital twin should also have real-time dynamic update, real-time interaction, timely response and other characteristics in the data dimension.

(3) Connectivity dimension A kind of view that the digital twin is the Internet of Things platform or industrial Internet platform, these views focus on perceptual access from the physical world to the virtual world, reliable transmission, intelligent services. From the perspective and IC660BRD021 requirements of meeting the physical connection mapping and real-time interaction of information, the ideal digital twin should not only support cross-interface, cross-protocol and cross-platform interconnection, but also emphasize the two-way connection, two-way interaction and two-way driving between different dimensions of the digital twin (physical entity, virtual entity, twin data, service/application), and emphasize real-time. Thus forming the information physical closed loop system.

(4) Service/function dimension one view that digital twin is simulation, virtual verification, or visualization, this kind of understanding is mainly from the perspective of functional requirements, to understand some functions/services that digital twin can support. At present, the digital twin has been applied in different industries and different fields, based on the dual drive of model and data, the digital twin not only reflects its application value in simulation, virtual verification and visualization, but also can be targeted at different objects and needs. Provide corresponding functions and services in product design, operation monitoring, energy consumption optimization, intelligent management and control, fault prediction and diagnosis, equipment health management,IC660BRD021  recycling and reuse. It can be seen that the services/functions of digital twins are diversified.

(5) Physical dimension one view holds that digital twins are only digital expressions or virtual bodies of physical entities, and their conceptual category does not include physical entities. Practice and application show that physical object is an important part of digital twin, and the model, data, function/service of digital twin are inseparable from physical object. Digital twins vary by physical entity object, data by physical entity characteristics, and functions/services by physical entity needs. In addition, information physical interaction is one of the important characteristics that distinguish digital twins from other concepts. If the concept category of digital twins does not include physical entities, the interaction lacks objects.

To sum up, there are many different understandings and understandings of digital twins at present, and there is no unified consensus definition yet, but physical entity, virtual model, data, connection, and service are the core elements of digital twins. Digital twins in different stages (such as different stages of products) show different characteristics, the understanding and practice of digital twins can not be separated from specific objects, specific applications and specific needs. From the perspective of application and solving practical needs, the practical application process does not necessarily require the “digital twin” to have all the ideal characteristics, and can meet the specific needs of users.