Abstract:As the faults of computer numerical control (CNC) machine tool having the characteristics of concealment and complexity, in order to quickly and accurately identify the faults, a fault diagnosis method for CNC machine tool was presented based on rational combination of the fuzzy logic, RBF neural network and particle swarm optimization (PSO) algorithm, which integrated with properties of strong selfadaptability, fast searching and strong compatibility of fuzzy neural network, and global strong searching ability of PSO algorithm. A modified velocity updating formula and inertia weight for particle swarm algorithm was proposed to improve the local searching ability to optimize the structure parameters of fuzzy neural network on basis of standard particle swarm algorithm. Thus, the fault diagnosis model of CNC machine spindle servo system with improved PSO optimization of fuzzy neural network was established. The experiment and simulation results show that the proposed method has higher fault identification accuracy and stronger generalization ability, as compared with RBF neural network and the standard PSO optimization of fuzzy neural network.