Predictive monitoring for abnormal situation management

A novel process monitoring method is proposed that uses predictions from a dynamic model to predict whether process variables will violate an emergency limit in the future. The predictions are based on a Kalman ®lter and disturbance estimation. A critical feature of the proposed method is the evaluation of a T2 statistic as a ``reality checkʹʹ for deciding if the future predictions are reliable and thus can be used for making control decisions. Several simulation examples demonstrate the e€ectiveness of the proposed tech- nique for both linear and nonlinear processes, and for a variety of disturbances.