Health Monitoring and Prognosis of Hybrid Systems

Maintenance and repair of complex systems are an increasing part of the total cost of final product. Efficient diagnosis and prognosis techniques have to be adopted to detect, isolate and anticipate faults. Moreover the recent industrial systems are naturally hybrid: their dynamic behavior is both continuous and discrete. This paper presents an architecture of health monitoring and prognosis for hybrid systems. By using model and experience-based approach we propose an implementation of an integrated diagnosis/prognosis process based on Weibull probabilistic model. This article focuses particularly on the prognosis algorithm description. The process has been implemented and tested on Matlab. Simulation results on a water tank system show how prognosis and diagnosis interact into the architecture.