Position Sensorless Control for Brushless DC Motor Based on RBFNN Optimized by Fast Recurvise Algorithm

摘要

The principle of position sensorless control for brushless DC Motors (BLDCM) is analyzed in this paper, and a new control method for BLDCM which is based on radial basis function (RBF) neural network optimized by Fast Recursive Algorithm, is proposed due to perfect nonlinear mapping characteristic of neural network. Using FRA, the proposed method can determine the numbers and locations of the centers, and derive the weights between the hidden layer and the output layer. The effectively of this proposed position sensorless control method is verified by the simulation results.