A new approach to minimize torque ripple and noise in model predictive control of permanent magnet synchronous motor drives

Abstract

Model predictive control widely advocates in the better performance control technique for permanent magnet synchronous motor drives for its exceptional flexibility of incorporating nonlinear parameters. However, the higher ripples in torque and harmonics in stator current are the major criticisms in this method. Enlightened by the idea of including nonlinear constraints in the cost function, this article proposes a novel cost function with reduced ripples in torque, stator flux, and current harmonics. The improvised cost function uses torque tracking, maximum torque per ampere, minimization of switching losses, and system constraints. This approach attains effectiveness in the minimization of the ripples in torque, stator flux, and stator current harmonics and also reduces the acoustic noise of the system. The claims of the proposed work are supported by the quantitative comparison of the simulated response with the standard model predictive torque control that uses errors in the stator flux and electromagnetic torque in its cost function. The compared simulation results prove the effectiveness of the proposed control technique.