Multiple-Vector-Based Model Predictive Flux Control to Reduce Common-Mode Voltage for Induction Motors Considering Dead-Time Effects

Abstract

To reduce common mode voltage (CMV) and improve steady-state performance for induction motors (IMs) fed by two-level voltage source inverter (2L-VSI), a multiple-vector based model predictive flux control with CMV reduction (RCMV-MV-MPFC) strategy is proposed. First, a simplified sector identification method based on cost function is proposed to obtain the optimal voltage vector (VV) and the corresponding duty cycle. Second, the switching sequence is designed to suppress CMV, but the CMV spike caused by dead-time effect still exists. Therefore, an optimal switching sequence (OSS) including hysteresis is proposed to completely suppress CMV spikes, and the switching frequency (SF) can be reduced by redundant OSS. In addition, the design of hysteresis only relies on dead time and has a low complexity. The simulation and experimental results show that the proposed RCMV-MV-MPFC strategy can not only completely restrict the CMV within $\pm V_{\mathrm {dc}}$ /6, but also improve steady-state performance, reduce SF and execution time.

Keywords

• Common-mode voltage (CMV)
• dead-time effects
• multiple-vector based model predictive flux control
• optimal switching sequence (OSS)
• induction motors (IMs)