模型预测控制(MPC)技术近年来在高动态性能电机驱动系统中应用广泛。为了克服传统MPC技术中有限控制集(FCS)造成的稳态电流脉动问题,提出了一种基于混合控制集(MCS)预测控制的永磁同步电机(PMSM)电流脉动抑制方法。分析建立PMSM预测控制系统离散数学模型,并分析电压矢量精度与电流脉动之间的关联性；在此基础上,MCSMPC将电压源型逆变器有限的有效电压矢量数,扩展为多个以占空比形式存在的虚拟电压矢量,并基于上述虚拟电压矢量完成MPC优化问题在线求解；此外,考虑到MCSMPC系统的参数敏感性问题,对MCSMPC系统反馈噪声问题进行分析讨论。最后,搭建双15 kW PMSM对拖样机测试平台进行试验分析,分析内容包括MCS方法动态跟踪特性、电流脉动稳态效果。试验结果表明所提出的MCSMPC方法在保留了传统预测控制技术高动态响应的基础上,可有效降低PMSM稳态电流脉动幅度和运行噪声。

In recent years, model predictive control technology (MPC) widely used in the high dynamic performance of motor drive system, in order to overcome the traditional MPC technology limited control set (FCS) steady current pulsation problem, a kind of based on mixing control set (MCS) predictive control method of alldigital fuzzy current ripple suppression was put forward. First of all, the analysis of discrete mathematics model alldigital fuzzy predictive control system, and the precision of voltage vector and the correlation between current ripple was analyzed. On this basis, the MCSMPC would voltage source inverter limited number of effective voltage vector, extended to multiple compared with the form of virtual voltage vector, and based on the above online virtual voltage vector to complete the MPC optimization problem; In addition, considering the MCSMPC system parameter sensitivity problem, analyzed the MCSMPC system feedback noise problems were discussed. Finally, set up double 15 kW permanent magnet synchronous motor for prototype test platform for experimental analysis, content analysis including the MCS method of dynamic characteristics, current ripple effect of the steady state. The experimental results showed that the proposed MCSMPC method in retained the traditional predictive control technology, on the basis of high dynamic response, which could effectively reduce the noise of alldigital fuzzy steady current pulsation and operation.

TM 351

国家自然科学基金资助项目(50274052)