Enhancement of Weighting Coefficient Selection using Grey Relational Analysis for Model Predictive Torque Control of PMSM Drive: Analysis and Experiments

Abstract

Permanent magnet synchronous motor (PMSM) with model predictive torque control (MPTC) is popular for its simplified control structure and adaptable in incorporating control parameters into the control algorithm. However, in control technique the primary concern for objective function (OF) depends on the selection of appropriate weighting coefficient (WC). Basically, for weighting coefficient selection, empirical methods are used but it takes additional time and heuristic process. In this paper, Grey Relational Analysis (GRA) technique is introduced in optimization of objective function for selection of appropriate weighting coefficient. In this methodology, stator flux and torque having individual OF are modified from single-OF. This ensures that in each sampling period, selection of grey relational optimal control action is dependent on the preference given to the control parameters in OF. For each sampling, a Grey Relational Grade (GRG) is employed to determine the appropriate control action. The models for two-level inverter fed PMSM are developed in MATLAB/Simulink to test the various operations of PMSM drive and the results are validated on the experimental test bench using dSPACE-1104 R&D controller. In order to highlight the effectiveness of the proposed technique, the results are compared with DTFC and MPTC approach.