TY - JOUR A1 - Beyer, Marc-André A1 - Grote, Wolfgang A1 - Reinig, Gunter T1 - Adaptive exact linearization control of batch polymerization reactors using a Sigma-Point Kalman Filter JF - Journal of Process Control N2 - The Chylla–Haase polymerization reactor is widely accepted as a benchmark process for the evaluation of control strategies for batch reactors. In this contribution a control concept based on Exact I/O-Linearization is proposed and compared to a conventional cascade control structure. In order to adapt the exact linearization control strategy to various polymerization products and batch conditions, an advanced probabilistic inference algorithm (Sigma-Point Kalman Filter) is applied and investigated. Sigma-Point Kalman Filters have the major improvement of simplified implementation compared to local linearization methods (i.e. Extended Kalman Filter) because no analytical Jacobians are required. Stochastic simulation studies are introduced and show the effectiveness, accuracy and benefit of the control concept. Within several scenarios a satisfying robustness against structural errors in the underlying model equations for the nonlinear control law and the inference algorithm is demonstrated. Furthermore it is pointed out, that with little effort in reassembling the plant design, control performance can be improved significantly. KW - Batch process operation KW - Exact I/O-Linearization control KW - Online parameter estimation KW - Sigma-Point Kalman Filter Y1 - 2008 UR - 1873-2771 U6 - https://doi.org/10.1016/j.jprocont.2007.12.002 VL - 18 IS - 7-8 SP - 663 EP - 675 PB - Elsevier ER -