Predictive Flux Control-Based Efficiency Optimization Strategy for Three-Level Inverter-Fed Linear Induction Machine Adapted to Linear Metro

The linear induction machine (LIM) in linear metro suffers from poor efficiency due to the large air gap, end effects, and long-term light load operating conditions. Although such issue has been improved by some LIM efficiency optimization strategies (EOSs), poor performance, and high harmonic loss caused by low switching frequency of large-capacity inverters have not been well addressed. To further improve the efficiency of LIM system under low switching frequency, this article proposes one predictive flux control-based EOS for three-level inverter-fed LIM system. First, one novel primary flux-based LIM loss model is built, with which the optimal primary flux for minimum loss can be derived analytically. Then, considering multiple control demands, a computationally efficient predictive control is proposed to manipulate optimal flux under low switching frequency. Through control targets conversion, sequential optimization and online adaptive adjustment, tuning work for all weighting factors is avoided. Finally, comprehensive experimental comparisons with the mature control strategy and the existing EOS in linear metro are investigated on one 3 kW arc induction machine. It is confirmed that more loss reduction and efficiency improvement for LIM can be effectively achieved with the proposed method under low switching frequency.