Abstract:This paper presents a method to build a formal model of a plant, in the form of a network of timed automata, to be used for model-based verification of controllers. To ensure re-usability, this model is built by instantiation of generic components models. When the instantiated components models are assembled, spurious evolutions leading to states which do not represent the real behavior of the plant, can occur, owing to the rich semantics of the modeling formalism. Then a modeling strategy is proposed in order to discard these evolutions so as to reduce the state space of the plant model to the only meaningful states. The method is exemplified and discussed on a small case study.

Formal verification methods require that a model of the system to analyze, in the form of a network of automata for instance, be built previously. Every evolution of this formal model must represent a real evolution of the modeled system; if the model contains indeed spurious evolutions, meaningless states, which do not correspond to physically possible states, can be reached and the verification results are surely not trustworthy. This paper focuses on construction of the formal model of a closed-loop system which can be represented as a Discrete Event System (DES) and where all evolutions and states are meaningful wrt to the real system behavior. A closed-loop system is composed of a physical system to control, named plant, and a controller. A modular approach to build the plant model is presented in the first part of the paper; to prevent from meaningless evolutions and states in this model, a solution based on the concept of urgent edges is proposed and exemplified. Then, construction of the formal model of the closed-loop system is addressed; it is shown that restriction of the evolutions of this model to the only meaningful ones can be easily achieved by introducing variables that represent the modification of the inputs of the logic controller and the stability condition of the control specification.

Model-driven engineering is a promising approach used to develop and analyze complex systems from different domains. In this paper, we focus on the safety aspect and introduce a methodology and associated framework for modeldriven safety analysis (SA) of large critical systems. The methodology is meant to cope with design complexity and reduce time of SA process. The framework, called Sophia, supports proposed methodology and includes facilities (i) to automatically perform various SA methods, (ii) to make semantic connections with formal SA tools, (iii) to represent SA results in the system modeling environment. We illustrate our approach using a case study from transport domain.

No abstract