A systematic approach for developing phenomena models from experimental data is presented. The approach is based on integrated application of stochastic differential equation (SDE) modelling and multivariate nonparametric regression, and it is shown how these techniques can be used to uncover unknown functionality behind various phenomena in first engineering principles models using experimental data. The proposed modelling approach has significant application potential, e.g. for determining unknown reaction kinetics in both chemical and biological processes. To illustrate the performance of the approach, a case study is presented, which shows how an appropriate phenomena model for the growth rate of biomass in a fed-batch bioreactor can be inferred from data.
Computer-aided Chemical Engineering: 36th European Symposium of the Working Party on Computer Aided Process, 2003, p. 1091-1096
Main Research Area:
Computer - Aided Chemical Engineering
13th European Symposium on Computer Aided Process Engineering, 2003