Applications of adaptive genetic algorithm to radar engine fault diagnosis

Abstract

According to the problem on calculating the synthetic exponent characterizing the whole performance of radar engine by using the synthetic weighted method, the weights of every parameter are difficult to be determined. To solve this problem, a method of determining the weights of every parameter by adaptive genetic algorithm is presented. The synthetic exponent gained by AGA is more sensitive and exact than the one gained by the expert investigated method in reflecting the whole performance of the engines. Mean while, this method improves the rate of identifying whether the performance of the engine is normal or not, finds the potential forepart fault of engine and prevents the spread of the fault. The validity of the method is testified by monitoring certain type of turbine-fan engine. The adaptive crossover probability and adaptive mutation probability are proposed, which consider the influence of every generation to algorithm and the effect of different individual fitness in every generation.