Robust model-predictive-control for a constrained nonlinear system with bounded disturbance
一类干扰有界约束非线性系统的鲁棒模型预测控制

For a constrained nonlinear system with bounded disturbance, we develop a new robust model predictive controller(MPC) based on the switching strategy of control-invariant set. Firstly, we propose for the nominal nonlinear system an algorithm for calculating the control-invariant set through linearizing the nonlinear system around the equilibrium point. Secondly, considering the effect of linearization errors and additive bounded disturbances, we determine the minimal robust positively invariant set around the equilibrium point. The robust MPC algorithm is then developed for the constrained nonlinear system with bounded disturbances, based on the switching strategy of the control-invariant set and the control strategy of the Tube-invariant set. The method is tested in a typical nonlinear continuously stirring tank reactor(CSTR); the results show the feasibility of the proposed method.