Sensorless Speed Control of Brushless DC Motor Using Artificial Neural Network Predicted Back EMF

Electric Vehicles (EVs) are increasingly preferred over Internal Combustion Engine (ICE) vehicles due to their eco-friendliness. Brushless DC (BLDC) motors are ideal for EVs because of their efficiency and low maintenance. Traditional BLDC motor control uses position sensors, which are expensive and complex. Sensorless control, which does not rely on these sensors, is more desirable. A typical sensorless technique is to detect the Back Electromotive Force (BEMF) in the motor phase, but this is difficult at low speeds and sensitive to noise. The suggested method is a Sensorless Speed Control for BLDC motors using an Artificial Neural Network (ANN) to analyze BEMF. The ANN learns the correlation between BEMF zero crossing signals and the hall sensor's equivalent emf. This aids in precise rotor position and speed determination, leading to correct motor drive signals. This method's benefits include noise resistance, the absence of extra sensors or circuits, and effective operation at all speeds. TI Motor Control Board is used for testing and MATLAB Simulink is used for comparative analysis. The torque and current ripples are significantly reduced by using ANN BEMF method, and the speed regulation is improved.