APPLICATION OF GENERALISED PREDICTIVE CONTROL TO A MILK PASTEURISATION PROCESS

摘要

In this paper, Generalised Predictive Control (GPC) is applied to a milk pasteurisation process, in order to improve pasteurisation temperature control. The controller robustness to plant/model mismatch is investigated, as the process is simulated by a nonlinear Artificial Neural Network (ANN) based model, while the embedded internal model is given by a linear First Principles (FP) model. The GPC controller is furthermore compared to an optimally tuned PID controller.