Null Space Method for Integrating Steady State Optimization and Model Predictive Control
【摘要】：In the process industries, steady state operation consumes the most part of the operating cost. The overall process system is typically implemented by a hierarchy model with various layers. The steady state optimization layer obtains the optimal setpoints based on a nonlinear model of the process. The optimal setpoints are implemented by means of the model predictive control layer. In this paper, the null space method is presented to integrate steady state optimization and model predictive control in the presence of expected and unexpected disturbances. Through using null space method to select the controlled variables as a combination of the measurements, the main advantage is to reduce the need for frequent setpoint changes as well as reject both expected and unexpected disturbances. An example is presented to illustrate the superiority of the proposed method.