基于微分包含的约束非线性系统显示预测控制

Explicit Predictive Control of Constrained Nonlinear Systems via Linear Differential Inclusion

提出了一种非线性预测控制的新方法。首先基于线性微分包含理论,利用泰勒级数对系统进行线性化,通过对偏导数取最大和最小的方法构造多面体描述的线性时变系统包裹原非线性系统,然后对于多面体描述的线性不确定系统,采用多参数规划的方法建立显示模型预测控制系统。对该方法进行了仿真计算,仿真结果表明采用这种方法可以更好的描述非线性系统,改善控制器的控制效果。

A new method of nonlinear predictive control is proposed in this paper. In our method,the coefficients is linearized by the Taylor series based on the theorem of linear differential inclusion, the original nonlinear system is wrapped by the linear time-varying system which is described by the polyhedron and constructed by the method of maximize and minimize the partial derivative. Then ,the explicit model predictive control system is developed based on the multi-parameter programming for the linear uncertain system described by polyhedron. A simulation example is presented to show that our method can describe the nonlinear system well and improve the performance of controller.