1. Institute of Control and Computation Engineering, University of Zielona Góra, ul. Podgórna 50, 65–246 Zielona Góra, Poland;2. School of Mathematical Sciences, Queen Mary, University of London, Mile End Road, London E1 4NS, UK
Abstract:
We present the theoretical background and the numerical procedure for calculating optimum experimental designs for non-linear model discrimination in the presence of constraints. The design support points consist of two kinds of factors: a continuous function of time and discrete levels of other quantitative factors. That is, some of the experimental conditions are allowed to continually vary during the experimental run. We implement the theory in a chemical kinetic model discrimination problem.