ESTIMATION OF PARAMETERS IN BIOREACTOR DYNAMICS MODELS

Abstract

Bioreactors are applied in the production of various products and the analysis of parameters that describe the production/consumption kinetics of the species becomes important to be able to design the bioreactors, as well as through mathematical models to be able to carry out simulations that make it possible to infer the concentration of species in scenarios where there is no experimental data. In this context, this article shows the application of Bayesian techniques (Monte Carlo Via Markov Chain-MCMC) to estimate both parameters and state variables in which there are no experimental measurements. The application was carried out using a model that has as state variables substrate (S), product (P) and biomass (X) using the Monod model as the kinetic model. The estimates obtained had good accuracy and precision in the evaluated scenario.