Adaptive ANN-Tuned PID Controller for Speed Regulation of BLDC Motor with Simulation and Hardware Validation

Abstract

Brushless DC (BLDC) motors have gained widespread acceptance in electric vehicles, industrial automation, robotics, and renewable energy applications owing to their superior efficiency, reliability, and superior dynamic characteristics. However, achieving accurate speed regulation under diverse system conditions remains a challenging task driven by the nonlinear nature of motor dynamics. This research examines the implementation and real-time implementation of an ANN-assisted PID-controller for BLDC motor speed regulation. The proposed controller utilizes an ANN to adaptively tune the PID parameters according to the motor operating conditions. Initially, the controller performance is verified through MATLAB/Simulink simulation. Subsequently, the trained ANN model is deployed on a microcontroller-based hardware platform for experimental validation. The response of the implemented system controller is evaluated and compared with a conventional PID controller using speed response characteristics and error-based performance indices. Experimental results demonstrate improved speed tracking, lower overshoot and shorter settling duration and enhanced robustness under varying operating conditions. The study confirms the practical feasibility of ANN-assisted intelligent control strategies for real-time BLDC-motor drive applications.