1. Analysis of digital twin maturity assessment requirements

As an effective means to realize cyber-physical integration, digital twin has been widely concerned and highly valued by domestic and foreign academic circles, government departments and people in related fields. In recent years, the author’s team focused on digital twin five-dimensional model, digital twin model construction theory, digital twin multi-scale modeling method, digital twin model evaluation index system, digital twin data, digital twin service, digital twin standard system, digital twin enabling Technology and tools have carried out exploratory theoretical research work, and a series of practical work has been carried out focusing on digital twin workshops, digital twin equipment, and digital twin satellites. However, in the process of communicating and discussing with domestic and foreign scholars, government departments and enterprises in related fields, we found the following common questions and confusions:

(1) How to judge whether it is a digital twin?

(2) How to judge whether the existing digital twin can meet the application requirements?
(3) How to optimize and improve if the application requirements are not met?

The above problems can be attributed to the lack of a systematic description and evaluation method for the digital twin category and development stage , while the maturity model is a systematic description of the concept category, development process and staged goals of the target system. At the same time, it also has the ability to evaluate A function of the target system’s current level of development and capability. Therefore, from the perspective of landing application, this paper proposes a set of digital twin maturity model, and designs a set of digital twin maturity evaluation application process.

2. Digital twin maturity level

Statistical analysis of existing theoretical research and application practices related to digital twins can be mainly divided into the following categories according to their functions and uses: ① physical entity design verification and equivalent analysis based on digital twins ; ② visualization of physical entity operation process based on digital twins Monitoring ; ③Remote operation and maintenance management and control of physical entities based on digital twins ; ④Diagnosis and prediction based on digital twins ; ⑤Intelligent decision-making and optimization based on digital twins ; Through the common analysis of the above-mentioned various types of digital twin research and applications, it is found that physical entities, digital twin models, and the connection and interaction between them constitute the “minimum concept” of digital twins. On this basis, based on the five-dimensional digital twin model proposed by the author’s team earlier, from the physical entity (PE), digital twin model (DM), digital twin data (DD), connection interaction (CI) and functional service (FS) five Starting from this dimension, according to the connection interaction mode and the degree of automation, taking the functional services that digital twins can provide as the main line, digital twins are divided into six maturity levels, as shown in Figure 1. Among them, the physical entity in the physical space and the digital twin model in the information space interact through the connection between the two, and the digital twin data contains all the information of the digital twin, running through the present-future, physical space-information space, physical entity- Digital twin model – connection interaction – function service. In addition, the physical entities, digital twin models, digital twin data, connection interactions, and functional services in Figure 1 have different levels of capability at each maturity level of the digital twin, but due to the limited information capacity of the picture, Figure 1 does not compare them. To expand and describe in detail, this part will be expanded in the third chapter .

2.2 Level 1 (L1): Mirroring reality  
with virtual reality refers to the use of digital twin models to reproduce the real-time state and change process of physical entities in real time. Digital twins with this capability are at the first level of their maturity level (L1) . At this level, the digital twin model is driven by real and time-sensitive data related to physical entities, synchronously and visually presents the same operating status and process as physical entities, and outputs the same results as physical entities, thereby breaking through time to a certain extent , Space and environmental constraints limit the monitoring process of physical entities, but the operation and control of physical entities still rely on the direct intervention of on-site personnel, and it is still impossible to realize remote visual control of physical entities.

2.3 Level 2 (L2): Controlling the real  
with the virtual refers to using the digital twin model to indirectly control the operation process of the physical entity. The digital twin with this capability is at the second level (L2) of its maturity level. At this level, the digital twin model in the information space has relatively complete motion and control logic, and can accept input instructions to realize a more complex operation process in the information space. At the same time, on the basis of mirroring the reality with the virtual, incrementally construct the data transmission channel from the digital twin model to the physical entity, realize the real-time two-way closed-loop interaction between the virtual and the real, thereby endowing the physical entity with the ability of remote visual control, and further breaking through space and environmental constraints. Limitations on manipulation of physical entities. Although this control is not necessarily intelligent or optimized, it can still greatly improve the efficiency of the management and control of physical entities.

2.4 Level 3 (L3): Using the virtual to predict the real  
and the virtual to predict the real refers to using the digital twin model to predict the operation process and state of the physical entity for a period of time in the future. The digital twin with this capability is at the third level (L3) of its maturity level . At this level, the digital twin model can dynamically reflect the current actual state of the physical entity based on the real-time two-way closed-loop interaction with the physical entity, and rationally utilize the explicit mechanism described by the digital twin model and the implicit laws contained in the digital twin data. , to realize the online preview of the future operation process of the physical entity and the speculation of the operation results, so that to a certain extent, the unknown is transformed into a prediction, and the sudden and occasional problems are transformed into routine problems.

2.5 Level 4 (L4):  
Optimizing physical entities with virtual excellence refers to using digital twin models to optimize physical entities, and digital twins with this capability are at the fourth level (L4) of their maturity levels. At this level, the digital twin can not only reflect the operating status of the physical entity in real time based on the digital twin model, but also predict the future development of the physical entity based on the digital twin data. Time-sensitive intelligent decision-making and optimization, and intelligent management and control of physical entities based on real-time interaction mechanisms.

2.6 Level 5 (L5): Virtual 
-real symbiosis As an ideal goal of digital twins, virtual-real symbiosis refers to the long-term synchronous operation of physical entities and digital twin models, and even the ability to realize autonomous twins through dynamic reconstruction during the entire life cycle. of digital twins are at level five (L5) of their maturity scale. At this level, the physical entity and the digital twin model can perceive and recognize each other’s updated content in real time based on two-way interaction, and based on the differences between the two, use 3D printing, robotics, artificial intelligence and other technologies to realize the integration of the physical entity and the digital twin model. Self-construction or dynamic reconstruction, so that both maintain dynamic consistency during long-term operation, so as to ensure the effectiveness of many functional services including visualization, prediction, decision-making, optimization, etc., and achieve low-cost, high-quality, sustainable digital twin.

3. Digital twin maturity evaluation factors

The digital twin maturity level describes the conceptual category and ladder goals of digital twins from a macro perspective, as well as the evolution path of digital twins formed by the series of goals at each stage. In order to realize the operable digital twin maturity evaluation, from the five dimensions of physical entity, digital twin model, digital twin data, connection interaction and functional service, further analysis obtains 19 relevant factors that can affect the maturity level of digital twin . It is used as an evaluation factor for maturity grading, and the specific requirements for each maturity level of the digital twin for the maturity level of the evaluation factor are specified, as shown in Table 1 to Table 6.

Table 1 Requirements for each maturity level of digital twin

The above digital twin maturity level and digital twin maturity evaluation factors constitute the digital twin maturity model , as shown in Figure 2. Among them, the digital twin maturity level systematically describes the conceptual category and phased goals of digital twin, as well as the capability characteristics of digital twin in each development stage, while the digital twin maturity evaluation factors include physical entities, digital twin models, The five subdivision dimensions of digital twin data, connection and interaction, and functional services sort out the different levels of capabilities required by digital twins at different maturity levels, laying the foundation for operable digital twin maturity evaluation.

In order to verify the digital twin maturity model and its application process proposed in this paper, the maturity evaluation of the unit-level digital twin (digital twin robot) and system-level digital twin (digital twin workshop) is carried out, respectively, as shown in Figure 4 and Figure 5 . Among them, the maturity evaluation results of the unit-level digital twin are used to guide its upgrade and optimization, and the maturity evaluation results of the system-level digital twin are used to compare the development levels of the two digital twin workshops.

and Maturity Evaluation Results of Digital Twin Workshop B

Some final learning experiences: