A theoretical multiple faults diagnosis approach of hybrid systems applied to a two tank system

The common approach in fault diagnosis is based in the assumption of single faults. This assumption may result to erroneous diagnosis statement in case where multiple faults occurs. Thereby multiple fault diagnosis is a challenging task especially in the control of large scale complex systems that can be viewed as hybrid systems. This owed to the fact that multiple faults are hard to detect because there consequences can mask or compensate to each other. The goal is to detect multiple faults as early as possible and provide a timely warning. A key issue is to prevent local faults to be developed into system failures that may cause safety hazards, stop temporarily the production and possible detrimental environment impact. In this work we introduce the notion of multiple faults diagnosability of hybrid systems in the framework of hybrid input output automata (HIOA). We present a methodology for detection of multiple faults imposing the conditions for a hybrid system to be diagnosable. This approach is applicable to a wide rage of systems since hybrid systems involve both continuous and discrete dynamics. The feasibility of the proposed approach is shown to a two tank system.