【目的】为解决永磁同步电机(PMSM)传统三矢量模型预测电流控制(TV-MPCC)的计算复杂性问题，提出了一种基于改进电压矢量选择和计算方法的低复杂度三矢量模型预测电流控制(LCTV-MPCC)方法。【方法】首先，基于前向欧拉法建立PMSM离散化电流预测模型，通过相邻有效电压矢量筛选方法确定三个候选矢量，缩小电压矢量选择范围。然后，计算价值函数以确定最优电压矢量组合，并计算出各自对应的虚拟作用时间。通过两个虚拟电压矢量及其作用时间合成期望电压矢量，根据虚拟电压矢量作用时间的大小关系，进一步确定合成期望电压矢量所需要的实际作用电压矢量及其作用时间。最后，按最小开关频率原则生成逆变器开关序列输出。【结果】为验证所提方法的有效性，通过仿真和试验进行对比分析。结果表明，所提LCTV-MPCC方法与TV-MPCC方法在稳态性能方面近似，两种控制方法的转速波动和电流波动基本一致；在计算效率方面，相较于TV-MPCC方法，所提LCTV-MPCC方法的单个周期平均计算时间由54.3 μs缩短至37.6 μs，减小了30.76%；电压矢量的迭代次数由6次缩减为3次，减少了50%，所提方法能够有效降低了传统控制方法的计算负担。【结论】本文所提LCTV-MPCC方法通过改变候选电压矢量集和实际电压矢量作用时间计算方法，有效解决了TV-MPCC方法计算复杂的问题，能够减少系统周期平均运行时间，同时固定开关频率，保持电机稳态控制性能不变。

[Objective] To address computational complexity issues in traditional three-vector model predictive current control (TV-MPCC) for permanent magnet synchronous motor (PMSM), a low-complexity TV-MPCC (LCTV-MPCC) method based on improved voltage vector selection and calculation is proposed. [Methods] Firstly, a discrete current prediction model for PMSM was established based on the forward Euler method, and three candidate vectors were selected using an adjacent effective voltage vector screening method to narrow down the voltage vector selection range. Then, the value function was calculated to determine the optimal voltage vector combinations, and their corresponding virtual action time was calculated. The desired voltage vector was synthesized using two virtual voltage vectors and their action time. Based on the duration of virtual voltage vector action time, the actual voltage vectors and their action time required for synthesizing the desired voltage vector were further determined. Finally, the inverter switching sequence output was generated according to the principle of minimum switching frequency. [Results] To verify the effectiveness of the proposed method, a comparative analysis was conducted through simulations and experiments. The results showed that the proposed LCTV-MPCC method achieved comparable steady-state performance to the TV-MPCC method, with nearly identical speed and current fluctuations under both control methods. In terms of calculation efficiency, the proposed LCTV-MPCC method reduced the average calculation time per cycle from 54.3 μs to 37.6 μs, which was a reduction of 30.76% compared to TV-MPCC. Additionally, the number of voltage vector iterations decreased from 6 to 3, representing a 50% reduction. The proposed method effectively reduced the computational burden compared to traditional control methods. [Conclusion] The LCTV-MPCC method proposed in this study effectively addresses the computational complexity issue of the TV-MPCC method by modifying the candidate voltage vector set and the calculation method for actual voltage vector action time. This method reduces average cycle runtime of the system, maintains fixed switching frequency, and ensures unchanged steady-state control performance of motors.

辽宁省“揭榜挂帅”科技计划项目(2023JH1/11100010)