MODIFIED RADIAL BASIS FUNCTION NEURAL NETWORK-BASED SPEED CONTROL OF A CHOPPER-FED DC MOTOR DRIVE: PERFORMANCE ANALYSIS AND COMPARATIVE STUDY

Abstract

The control of DC motor drives is critical for achieving high performance in industrial applications requiring precise speed regulation. Conventional controllers such as proportional–integral (PI) controllers often exhibit limitations, when applied to nonlinear systems with varying load conditions and parameter uncertainties. This paper presents a radial basis function (RBF) neural network-based controller for the speed control of a chopper-fed DC motor drive.
The proposed controller is developed using MATLAB/Simscape Electrical and is designed to regulate motor speed by adjusting the duty cycle of the DC chopper. A modified RBF neural network architecture is employed to capture the nonlinear dynamics of the converter–motor system and improve control accuracy. The performance of the proposed controller is evaluated using three different DC motor drive configurations representing various operating quadrants.
Simulation results are analyzed in terms of duty cycle response, armature voltage, armature current, and motor speed. A comparative analysis with a conventional PI controller demonstrates that the RBF neural network controller provides improved dynamic performance, including reduced oscillations, faster settling time, and enhanced stability under varying load conditions.
The results indicate that the proposed RBF-based control strategy offers a reliable and effective solution for nonlinear motor drive systems and has strong potential for application in modern industrial control.