Using Hybrid Automata for Diagnosis of Hybrid Dynamical Systems

Physical systems can fail. For this reason the problem of identifying and reacting to faults has received a large attention in the control and computer science communities. In this paper we study the fault diagnosis problem and modeling of Hybrid Dynamical Systems (HDS). Generally speaking, HDS is a system mixing continuous and discrete behaviors that cannot be faithfully modeled neither by using formalism with continuous dynamics only nor by a formalism including only discrete dynamics. We use the well known framework of hybrid automata for modeling hybrid systems, because they combine the continous and discretes parts on the same structure. Hybrid automaton is a states-transitions graph, whose dynamic evolution is represented by discretes and continous steps alternations, also, continous evolution happens in the automaton apexes, while discrete evolution is realized by transitions crossing (arcs) of the graph. Their simulation presents many problems mainly the synchronisation between the two models. Stateflow, used to describe the discrete model, is co-ordinated with Matlab, used to describe the continuous model. This article is a description of a case study, which is a two tanks system