Modifier Adaptation for Constrained Closed-Loop Systems

S. COSTELLO; G. FRANÇOIS; D. BONVIN; A. MARCHETTI

Cape Town

Congreso; 19th IFAC World Congress; 2014

IFAC

The steady advance of computational methods makes model-based optimization an increasingly attractive method for process improvement. Unfortunately, the available models are often inaccurate, which results in significant plant-model mismatch. An iterative optimization method called ?modifier adaptation? overcomes this obstacle by incorporating plant information into the optimization framework. A particular situation is when the plant operates under feedback control and only a model of the open-loop system is available. This paper extends modifier adaptation for constrained optimization problems to that particular situation. The inputs of the controlled plant are the setpoints, whereas the model inputs are the manipulated variables. Using this open-loop model and process measurements, the proposed algorithm guarantees both optimality and constraint satisfaction for the controlled plant upon convergence. A simulated CSTR example with constraints illustrates the method.