Aiming at the identification of typical faults of planetary gear transmission systems,a fault diagnosis method based on signal mixing characteristics and chaos fruit fly optimization algorithm-generalized regression neural network (CFOA-GRNN) was proposed.Several typical time-domain statistical characteristics of the signal were calculated,the frequency-domain energy characteristics of the signal were obtained through wavelet packet decomposition,then the signal mixed characteristic vector was obtained as the input of the generalized regression neural network (GRNN);the improved fruit fly optimization algorithm with chaotic disturbance was used to find the optimal parameters of GRNN,and the optimal diagnosis model was constructed;the collected data under different working conditions of the gearbox test bench were used for application and comparison.The results show that the proposed method can be used to effectively identify different faults of the gearbox under different working conditions；compared with other models,the proposed method has the advantages of simple parameter setting,small subjective factors influence,fast optimization speed and better practicability.