Robust Predictive Current Control of Permanent-Magnet Synchronous Motors With Newly Designed Cost Function
Xicai Liu, Libing Zhou, Jin Wang, Xiaonan Gao, Zhixiong Li, Zhiwei Zhang
Abstract
In model predictive control, mathematical model of the system is used to predict the value of state variables. Control performances of model predictive control suffer from parameter mismatches and model uncertainties. Steady-state errors will exist due to inaccurate predictions. This article presents a simple predictive current control strategy for robustness improvement of finite control set-model predictive control, with which steady-state errors under parameter mismatches can be eliminated. Neither disturbance observer nor explicit solution of compensation voltage is needed in the proposed control strategy. A cost function is newly designed, which is in proportional-integral form. Moreover, the integral action is only activated in a predefined range, which facilitates the design of the integral coefficients. The accumulated error is not simply included but weighted with the sampling time. In this way, a careful selection of the number of integral terms to be included in the cost function is not required. Experimental results demonstrate superior robustness of the proposed current control strategy to that of conventional predictive current control against parameter uncertainties.