GRA-TOPSIS-Based Model Predictive Torque Control for Induction Motor Without Weighting Factors

Model predictive torque control (MPTC) has been recognized as a promising control strategy for induction motor (IM) drives. However, the weighting factor for adjusting the relative weights of torque and flux terms must be tuned correctly under different operating conditions. Up to now, the tuning process of weighting factor is still mainly based on engineering experience. In this paper, a grey correlation analysis-based Technique for Order Preference by Similarity to Ideal Solution (GRA-TOPSIS) is introduced as a solution for this issue, the proposed GRA-TOPSIS-MPTC eliminates the weighting factor tuning procedure and reduces the control complexity. As known, TOPSIS is one of the popularly multi-criteria decision-making methods used for the selection of optimal alternative from the various available alternatives. This paper combines the grey relational analysis (GRA) and TOPSIS to further enhance the decision-making rationality. In the GRA-TOPSIS-MPTC system, the torque and flux control objectives are divided into two separate sub-cost function. Then the torque and flux control errors are both standardized and evaluated based on the GRA-TOPSIS theory. The simulation and experimental results demonstrate that the proposed GRA-TOPSIS-MPTC, which has no weighting factor, achieves similar or even better control performance, and reduces the switching frequency than conventional MPTC.