Application of RBF Neural Network in Intelligent Fault Diagnosis System

With the problem of large-scale system safety and stability increasingly prominently, intelligent fault diagnosis is becoming very important, and this paper puts forward the application of neural network in fault diagnosis system and analyzes in detail the principle, structure model, learning algorithm of radial basis function (RBF) neural network based on the basic principle of neural network knowledge. At last, taking the numerical control (NC) module in a system as an example, combined with the specific characteristics of NC module, this paper gives the structure of neural network diagnosis system and builds the RBF network model for simulation, training, and learning, the result of which shows that the intelligent fault diagnosis method can improve the back-propagation (BP) neural network, is feasible, and has a strong practical value.