Digital Twins for Cyber-Physical Systems – DT4CPS_V3

The first result concerns a proposal for a methodology implementing a digital twin integrating the models and data required by both the automation engineer (at the controller level) and the production engineer (at the line level). For this, a data-oriented JN architecture will be proposed, allowing the management of the different data flows between twins. As part of the project, a study on the interoperability of information systems will be necessary. The second result is a proposal for adjusting the digital models of the JN to align its behavior with the JP. This major challenge, expected by academics and industrialists, involves dynamic synchronization of the twins. Ensuring alignment will allow better exploitation of data from the twins at different scales (automation engineer, production engineer) and therefore provide support for reconfiguration decisions based on production or energy capacities and consequences. For this purpose, methods and tools are being considered for the control and reconfiguration of SCPPs, with a deliberate desynchronization of the JN in order to analyze its reconfiguration possibilities during virtual commissioning, before decision and resynchronization. A benchmark/demonstrator incorporating all the results obtained and reference data sets from the DT4CPS project will be made available to the scientific community.

/

/

Philippot A.; et al. Digital Twins for Cyber-Physical Systems. Session du CT JNum - 2èmes journées nationale de la SAGIP, Lyon, mai 2024.

Roisin M.; Yvars P.A.; Riera B. Constraint Programming for Logic controller Synthesis. 10th International Conference on Control, Decision and Information Technologies. 01-04 July, 2024, Valetta, Malta.

The digital twin, a virtual clone of a physical system, is seen as a key disruptive technology in the industry of the future. It is seen as a tool for steering Cyber-Physical Production Systems (CPPS) and twin should theoretically allow for improved commissioning, predictive maintenance and optimisation of the Production 4.0 system. Currently, the automation engineer uses simulation models of operating parts in a relatively empirical way for virtual commissioning and the implementation of different test scenarios in order to verify/correct the control part and thus save time during the real commissioning. Manufacturing engineers, for their part, use digital models for flow simulation, to evaluate decision-making scenarios (sequencing of operations, reconfiguration, impact of hazards, etc.). These simulation models are currently limited to testing the control system (once it has been completed) and to sizing and reconfiguring the installations, and are only used to a limited extent during the operating phase of the SCPP.
The DT4CPS (Digital Twins for Cyber-Physical Systems) project aims to federate and enrich the simulation models used by automation and manufacturing engineers in the design phase, to make a digital twin. By ensuring coherence with the physical system throughout the life cycle (operation and improvement in particular), this digital twin should make it possible to improve the commissioning of SCPPs, such as reconfigurable manufacturing systems, and become a decision-making tool during the operation phase. To achieve this, the DT4CPS project aims to remove several scientific (command and control of SCPPs), methodological (what methods and tools for automaticians/manufacturing engineers/operators 4.0?), technical (interoperability of information systems) and human barriers.