From safety analysis of reconfigurable systems to design of fault-tolerant control strategies

Abstract

The design of fault-tolerant control strategies requires a perfect knowledge of both the possible reconfigurations of the system and of the behavior of this system when failures occur. In this paper it is shown that the use of a model-based safety analysis (MBSA) framework, able to cope with repairable and reconfigurable phased-mission systems, is helpful for the choice of the best reconfiguration strategies to be implemented in the control system. The core of this approach is based on the integration of a model of the system structure (Fault Tree), a model of the dysfunctional behaviors of the components of the system (Switched Markov Processes) and a model of the reconfiguration mechanisms (Moore Machines). The syntax and semantics of the different models and their integration is first defined. The benefits of this approach for performance evaluation of fault-tolerant control strategies are afterwards illustrated through an application example.