Influence of supervisory control on the dependability of a hybrid system

A dynamic hybrid system is described by a set of continuous variables and a set of discrete events interacting together. The reality also requires taking into account component failures and the stochastic behavior of the system. Some events or variables take then a stochastic character. The component failures are in fact unobservable events and partially controllable. We are going to show how the control theory might be a mean to help to the conception of safe systems assuming that changes in the system (introduction of safety barriers …) would act as a supervisor in the context of simulation. The behavior of the system so “controlled” may be established without having to rebuild a complete model of the modified system. This assumes that the propagation of dangerous initiating events is inhibited and then the probability of dangerous states of the system reduced to zero.