Adaptive-predictive Control with Intelligent Virtual Sensor

In many industrial plants, some key variables cannot always be measured on-line and for the purpose of control, a virtual sensing system is often required. This paper is concerned with a development of an alternative intelligent control strategy, which is an integration between adaptive-predictive based controller and intelligent virtual sensing system. This allows an immeasurable variable to be inferred and used for control. The neuro-fuzzy approach is used for modelling the process as it has learning capability from the numerical data obtained from the measurements and subsequently used as process model in the generalized predictive control scheme. The intelligent virtual sensor is composed of the Diagonal Recurrent Neural Network (DRNN) and the Extended Kalman Filter (EKF) as the estimator with inputs from DRNN. The integration between virtual sensor and the controller enables the development of an on-line control scheme involving the immeasurable variable. Experimental results show some potential benefits on applying the proposed technique using the real-world process plant. Copyright ©2005 IFAC