A unified framework for improving the quality of continuous time models of dynamic systems based on experimental data is presented. The framework is based on an interplay between stochastic differential equation (SDE) modelling, statistical tests and multivariate nonparametric regression. This combination provides systematic methods for pinpointing and repairing model deficiencies by uncovering their structural origin. The potential of the proposed framework in terms of modelling complex dynamic phenomena such as reaction kinetics is illustrated with a case study involving a model of a fed-batch bioreactor, where it is illustrated how an incorrectly modelled biomass growth rate can be pinpointed and an estimate provided of the functional relation needed to properly describe it.
Computer-aided Chemical Engineering: 8th International Symposium on Process Systems Engineering, 2003, p. 1292-1297
Main Research Area:
Computer - Aided Chemical Engineering
8th International Symposium on Process Systems Engineering, 2003