GRNN model for fault diagnosis of unmanned helicopter rotor’s unbalance

Abstract

In order to diagnose the unmanned helicopter rotor’s unbalance fault accurately, a method based on particle swarm optimization algorithm and generalized regression neural network (PSO-GRNN) is proposed. The average mean square error got from cross-validation is used as the fitness function of particle swarm, then the optimal GRNN smooth factor is attained by using particle swarm optimization algorithm, and an optimal model for fault diagnosis is achieved finally. It can be concluded that, based on the PSO-GRNN model, the type and the grade of the helicopter rotor’s unbalance can be diagnosed effectively, the diagnosis accurate rate of fault type is up to 94.29 % and the maximum error of fault grade is only 6.54 %, which is perfectly satisfied for the requirement of project.