An Analytical Cost Function Design and Implementation for Predictive Control of Induction Machine Drives

Abstract

In finite control set-model predictive torque control (FCS-MPTC) of induction machine (IM), the optimal design of weighting factors for the cost function has always been a research difficulty in community of scholars. Generally, for weighting factors of FCS-MPTC, the rating method or cut-and-trial method are often utilized. These methods adopts the fixed value for weighting factor, which can not adapt to multiple operating modes of IM. In addition, the cut-and-trial method is cumbersome and difficult to achieve multi-objective balanced regulation. To solve this problem, this paper proposes an analytical cost function design for predictive control of induction machine drives (abbreviated as ACF-MPTC). According to the internal electromagnetic relationship of IM, the analytical expression of weighting factor is obtained through theoretical derivation. The control performances of traditional method and ACF-MPTC, which includes root mean square (RMS) of electromagnetic torque, RMS of stator flux and total harmonic distortion (THD) of stator current, for different operating modes of IM is studied. The parameter sensitivity of ACF-MPTC is analyzed, and a robust ACF-MPTC based on online parameter identification technology is proposed. The experimental results verify the effectiveness of the proposed algorithm.