Graphical MPC for Fast Dynamic Systems

Software configuration and engineering costs have limited the application of model predictive control (MPC) for small but fast dynamic systems. This work illustrates the benefits of using a graphical programming framework for the configuration and implementation of MPC controllers. Graphical programming facilitates the understanding and configuration of advanced applications so that engineers in industry can be responsible for the installation and maintenance of advanced controllers. Costs reduction and minimal specialized labor opens the possibilities of applying MPC to small systems with fast dynamics. Fast MPC execution is achieved by including the optimization constraints as penalty terms in the cost function. An air-heater pilot system is successfully used to demonstrate the advantages of a graphical framework for process modeling, design, and real-time implementation of MPC controllers in systems with fast dynamics.