A New Approach of Position Sensorless Control for Brushless DC Motor

Abstract

Position sensorless control technique for Permanent Magnets-Brush Less Direct Current (PM-BLDC) motor drive is considered in this paper. A new estimation based on sensorless technique is proposed for PMBLDC motor. Artificial Neural Network (ANN) is aided for the purpose. The inputs to the ANN are the voltages of PM-BLDC motor and it estimates the sample signals to feed Zero Crossing Point (ZCP) detection circuit. The ZCP detection circuit provides ZCP signals for commutation logic which gives the commutation sequence to power switches. In order to provide the correct sample signal to ZCP detection circuit, the ANN is well trained by Genetic Algorithm (GA). The proposed sensor less control model is implemented in MATLAB/SIMULINK working platform. Field Programmable Gate Array (FPGA) is used to implement the proposed method. Experimental results verify the analysis and demonstrate the advantages of the proposed method.