Aplicação de técnicas bayesianas no processo debiolixiviação de urânio

Abstract

This work aims to study the application of Bayesian techniques, Monte Carlo methodvia Markov chain and the importance sampling filter and sequential resampling, for the estimation of kinetic parameters and state variables of the uranium bioleaching processby Acidithiobacillus ferrooxidans. To make the estimates, experimental measures of cell concentration, uranium, ferrous and ferric ion, available in the literature, were used,which were evaluated using a mathematical model composed of a system of ordinary differential equation. Different case studies were conducted involving the measurement and model deviation (1 and 5%), as well as the kinetic model of the specific speed of cell growth. The results of the estimates showed a good agreement with the experimental measurements, presenting coefficient of determination values ranging from 0.95 to 0.99. The mathematical models evaluated were validated by reproducing the experimental measures and the use of Bayesian techniques proved to be adequate.