An improved Bayesian Optimization Algorithm for fault identification on flight control system

Fault identification method provides a great enhancement by using evolutionary algorithms in complex mechatronics systems. A Mutation-based Bayesian optimization algorithm is presented to improve the efficiency of Bayesian optimization algorithm (BOA). The mutation operator which makes full use of local information is combined into BOA by diversity function. The original objective is to combine the global information and local information in order to avoid local optimum. According to the fault analysis of aircraft actuation systems, the program of BOA for fault identification is introduced. The scheme is illustrated through simulations applying the flight control system of a fighter. The simulation result show fault identification is achieved.