A Fast Learning Neuro Adaptive Control of Buck Converter driven PMDC Motor: Design, Analysis and Validation

This paper presents a novel fast learning neural network for the estimation of load torque in PMDC motor. The control objective of angular velocity trajectory tracking is achieved by designing a controller for cascaded Buck converter PMDC motor system by utilizing an adaptive backstepping methodology augmented with a new Type-II Chebyshev neural network (CNN). The online learning laws for the neural network are developed, satisfying overall closed loop system stability using Lyapunov stability criterion. A rigorous stability analysis has been provided. Performance of the proposed control method is validated on a digital platform using dSPACE Control Desk DS1103 set-up with TM320F240 Digital Signal Processor. The dynamic response of Buck converter driven PMDC motor is examined for settling time, peak undershoot and overshoot guaranteeing the transient performance under conventional adaptive backstepping control and Type-I CNN based adaptive backstepping control techniques. Further, such results are compared with those obtained using the proposed method under start-up, wide range variations in load torque and reference trajectory.