【目的】为提升电动汽车驱动系统中内置式永磁同步电机(IPMSM)在复杂工况下的运行效率与动态性能，针对传统最大转矩电流比(MTPA)控制策略参数依赖性强、稳态效率易受参数偏差影响及搜索法动态响应不足的问题，本文提出了一种中国轻型车测试循环(CLTC)工况下基于改进搜索法的最大转矩电流比(IS-MTPA)控制策略。【方法】首先，基于CLTC工况，采用K-Means聚类算法对驱动模式下的913个工况点进行聚类，从而提取6个典型工况点。其次，将典型工况点对应的MTPA集总参数作为搜索初值，并在线调整集总参数以补偿电机参数变化，实现快速、高效寻优。最后，通过仿真和试验对所提IS-MTPA-CLTC控制策略进行验证。【结果】仿真与试验结果表明，相较于传统MTPA控制策略，所提IS-MTPA-CLTC控制策略在典型车用工况下可取得更优的电流分配，铜耗平均降低28.6%，有效减少了电机能量损失；相较于集总参数搜索法MTPA控制策略，所提IS-MTPA-CLTC控制策略在负载突变工况下可显著加快搜索法自寻优过程，d、q轴电流收敛时间缩短70%，有效改善了动态跟踪性能，增强了系统的动态响应性能。【结论】所提IS-MTPA-CLTC 控制策略能够兼顾稳态高效与动态快速寻优需求，适用于电动汽车真实运行场景下的IPMSM驱动控制，具有良好的工程应用潜力。

[Objective] Aiming at the difficulty of directly detecting the current in permanent magnet synchronous motor (PMSM) equipped with inductor-capacitor-transformer (LCT) high-order filter, this paper proposes a current cascaded observation method based on an extended state observer (ESO) and a Luenberger state observer (LSO). [Methods] The novel ESO cascaded method proposed in this paper required only the inverter-side current signal. The intermediate system state was obtained through the LSO, and disturbances were estimated and compensated in real-time by the ESO, achieving high-precision observation of the motor current. [Results] The experimental results showed that, under rated conditions, the motor current observation error was reduced by 3.2% with the novel ESO cascaded method proposed in this paper compared to the traditional LSO. [Conclusion] The proposed method achieves high-precision observation of high-order filtered PMSM current without requiring additional sensors or hardware modifications, effectively reducing system complexity and cost.

国家自然科学基金(52477070);山东省自然科学基金(ZR2025MS782)