Performance diagnosis of MPC with model-plant mismatch

Model predictive controllers have been widely applied in refinery and chemical plant. The process model plays an important role in the performance assessment of model predictive control. In this work, a methodology is adopted for the diagnosis of model-plant mismatch (MPM) based on closed-loop operating data in the DMC system. The problem of MPM is transformed into the effectiveness of the model. Also apply statistic inference to detect the occurrence of the mismatch. This measure is successfully illustrated through the implementation in the Wood-Berry tower process with three cases of mismatch and show the feasibility and effectiveness of the method.