NOTE CONCERNING THE ONLINE KINETICS IDENTIFICATION BY USING PARALLEL SHORT-CUT RECURSIVE EXACT ESTIMATORS AND A KNOWLEDGE KINETIC MODEL-DATA BANK - EXEMPLIFICATION FOR A CATALYTIC BATCH PROCESS

On-line concentration measurements from an isothermal semi-batch syste m, i.e. the liquid-phase acetoacetylation of pyrrole with diketene, ar e used for identification of a kinetic model. An integrated expert sys tem, containing coupled estimation procedures and statistical analysis tools as well as a kinetic model data bank, chooses the most suitable structure for the kinetic model and subsequently identifies model par ameters. With the increased availability of data during the run, refin ements or structural changes in the kinetic model are performed automa tically and parameter estimates are updated. A recently developed shor t-cut procedure (the Modified Integral transformation Procedure, MIP. Maria and Rippin, 1994a) is compared against classical direct methods. Interactive use of a kinetic model-data bank through a numerical tran sfer of information strategy allows adaption of the mechanism-based ki netic model structure. Multiple nonlinear stare estimators of the Exte nded Kalman Filter (EKF) type are operated in parallel, while the MIP/ kinetic data-bank allow information for switching between the EKFs and concerning mechanism and model structure. This on-line use of paralle l direct (short-cut) estimation procedures is particularly suitable fo r real-time process supervision, control and optimisation purposes.