Implementation of Convolutional Neural Network for Speed Control of BLDC Motor

This paper presents a modern control strategy for the speed control of a Brushless DC (BLDC) motor based on Convolution Neural Network (CNN). The PID loop employed to manage the quickness of the BLDC motor is attenuated using the multi layered perception-based CNN. The modern technique is successful in optimizing the parameters of integral square error of the PID controller. A convolution model is created, compiled and trained based on the verified data and the evaluated model performance. The sequence of speed-controlled data from the controller is to be multiplexed into convolution region, Rectified Linear (ReL) Unit, pooling zone, finally the rest of the error is minimized in fully connected area.