State Estimation of the Vinyl Acetate Reactor Using Unscented Kalman Filters (UKF)

Abstract

The main objective of our research is to develop several unscented transform techniques (UTT) to estimate the state of the nonlinear processes such as an improvement of an Extended Kalman Filter (EKF) approach. The Extended Kalman Filter (EKF) has become a standard nonlinear estimation technique in control systems and parameter estimation for nonlinear system identification. The Unscented Kalman Filter (UKF) developed in this paper is a superior alternative to the Exdended Kalman Filter for the most of estimation and control applications. We will try to figure out in this paper that the UKF algorithm performs slightly superior compared to EKF algorithm based on the state estimation of the nonlinear vinyl acetate reactor developed well in [1]. These results are encouraging for us and we want to explore the possibility of extension of its applicability to the other possible applications such as state and parameter estimation, neural network identification, and Fault Detection, Diagnosis and Isolation (FDDI) of the nonlinear control systems.