Electrothermal Performance-Based FCS-MPC for Dynamic Thermal Balance Control of Traction Converters

Abstract

The thermal stress imbalance is a challenge for the reliability and in-service lifetime of modular multilevel converters, especially in traction drive applications. The finite-control-set model predictive control (FCS-MPC) can offer an effective solution to this issue, but it is faced with two difficulties, that is, the weight allocation of electrical and thermal variables with different natures in control objectives functions, and dynamic weighting factor adjustment during the switching of variable operations. Therefore, an electrothermal performance-based FCS-MPC (ETP-FCS-MPC) for dynamic thermal balance control is proposed, which can not only achieve thermal stress balance in converters with more intuitive and time-saving ways but also offer dynamic compromise schemes between the electrical and thermal performance of converters in the speed switching process. First of all, dynamic electrical and thermal models of converters are presented. Then, the absolute error-based electrical control objective and the energy loss variance-based thermal control objective are designed. Moreover, a performance-based normalized control objective function and an outer loop control feedback-based dynamic weighting factor are proposed. Finally, hardware-in-the-loop (HIL)-based test is carried out to verify the effectiveness of the proposed method, and the test results show the merits of the proposed method.