International audience

The design of Cyber Physical Systems (CPS) is becoming increasingly complex due to the dynamic changes in their environments and infrastructures, requiring them to be self-adaptive. An important class of CPS is Industrial Control Systems (ICS), where a major trend is to upgrade from historically specific hardware and technologies towards more software-defined, virtual approaches involving the Could-Fog-Edge continuum. In this work, we propose a model-based approach for the design of the self-adaptation in ICS, inspired by, and applied to an industrial case study in Smart Grids, more particularly an electrical substation from RTE (the French Energy Transmission company). The problem is to allocate and reallocate dynamically a set of control functions upon a distributed computing infrastructure, with self-adaptation to variations and perturbations. We define and implement the model-based autonomic management feedback loop using constraint programming, to describe the space of possible configurations, as well as the constraints and objectives formalizing the operators strategies. This model is used in simulation, calling the constraints solver at each cycle of the loop.

Cyber-Physical Systems (CPS) contain complex computational components that control physical entities.The design of these components must take into account the realtime and concurrent nature of these systems.Formulating requirements that describe CPS behaviors precisely, ruling out misunderstandings, is a crucial yet difficult endeavor.To increase trust in the requirements, formal methods can be used to check relevant properties of the requirements.We investigate a process algebra to capture real-time behaviors and concurrency in CPS requirements in order to automate their analysis.We use a structured natural language to first express CPS requirements: this takes into account current practice, indeed requirements should be easily writable as well as graspable by stakeholders with various points of view and ease communication among them.At the same time, requirements analysis using simulation or formal validation is possible by taking advantage of the requirements structure.We discuss translation from the structured requirements into the process algebra to automate the overall process.Our approach is implemented and is illustrated by an example issued from CPS4EU project 1 .