A Fault Detection and Isolation Model Based on Conditional Finite State Machine for Gas Turbine

How to construct the appropriate and feasible fault diagnosis and isolation model is the key technology to guarantee gas turbine system running normally. This paper introduces the notion of fault states represented by finite state machine to describe the process of fault detection. Fault relationship type is analyzed as three types. Then conditional finite state machine of a fault is defined. And we construct a fault detection and isolation model with cause and effect, time conditions. The fault state and their transitions is analyzed in detail. Then running algorithm of above FSM is presented and the applied algorithm is further given. In the end, we apply above model and algorithm to a fault detection case of fuel system in the gas turbine, construct relative conditional finite state machine and describe the detection process.