异步电机低开关频率的模型预测直接电流控制

针对二极管箝位型(neutral point clamped, NPC) 三电平逆变器驱动异步电机的低开关频率控制, 提出了一种新颖的模型预测直接电流控制(model predictive direct current control, MPDCC) 方法. 基于推导的传动系统离散时间内部预测模型, 控制器预测了每个容许开关位置对应的电流输出轨迹、外推输出轨迹, 并根据评估平均开关频率的价值函数在线滚动优化实时求取最优开关矢量, 使得逆变器平均开关频率最小化, 且保持电流轨迹在给定的滞环范围内. 相对于已有的单步预测电流控制方法, 该方法在保持基本相同谐波性能的同时显著降低了开关频率. 仿真结果表明, 低开关频率模型预测电流控制算法可将三电平逆变器的平均开关频率降低至500 Hz 以下, 并且具有较理想的电流谐波畸变和动态响应性能.

Model Predictive Direct Current Control of Induction Motor at Low Switching Frequency

This paper proposes a model predictive direct current control (MPDCC) method for low switching frequency control of induction motors fed by a neutral point clamped (NPC) three-level inverter. Based on a discrete-time internal predictive model of the drives, the controller predicts future trajectories of stator current
for each admissible switching position, extrapolates the output trajectories and selects the optimal switch vector according to a receding horizon policy of the cost function that evaluates the average switching frequency of inverter. This makes the switching frequency minimized and the predicted current trajectories within its specified hysteresis bounds. Compared with the established one-step predictive current control, the proposed scheme can reduce switching frequency significantly, and keeps similar harmonic characteristics. Simulation results show that the switching frequency of the three-level inverter can be dropped below 500 Hz. A low harmonic distortion rate of three-phase current is achieved, and fast transient response can be realized by using the proposed scheme.