A method of control performance diagnosis to discriminate between model-plant mismatch and unmeasured disturbance

Control performance assessment and fault diagnosis is important to ensure long-term effective operation of control systems. In this paper, for the poor performance case, the methodology which is based on the analysis of closed-loop data from the process is proposed to discriminate between model-plant mismatch and unmeasured disturbance, which are the principal reasons for control performance deterioration. In the comprehensive method for fault diagnosis, neither the priori knowledge of process model nor external excitation is necessary. A simulation example was given to verify the effectiveness of the proposed method.