Problemas inversos aplicados em caracterização hidrodinâmica de reator anaeróbio de leito fixo

Abstract

Anaerobic fixed-bed reactors (AFRBs) are an excellent alternative for the treatment of water, sewage and industrial waste. As a result, many engineering areas have begun to perform hydrodynamic modeling of these equipment using the Residence Time Distribution (RTD) technique. The application of Inverse Problems becomes interesting in this context because it allows the estimation of hydrodynamic parameters that cannot be obtained experimentally and that provide important information about the best design and operating conditions. In this work, we propose the application of the Markov Chain Monte Carlo Method (MCMC) to estimate the parameters of the Piston, Dispersion and Mass Transfer (PDE) hydrodynamic model, adjusted to evaluate the variation of physical properties along the length of the reactor. Experimental data from pulse-type stimulus-response tests were obtained with the tracers eosin γ, bromophenol blue and bromocresol green at high and low flow rates. The results showed that bromophenol blue was the best tracer and that the best operating conditions are at high flow rates, since low flow rates favor an increase in short circuits and decrease the hydraulic efficiency of the reactor. The values of hydraulic efficiency (λH > 0,5), variance (0 < < 1) and short-circuits (ѰCC > 0,3) showed that the AFRB is well designed. Therefore, it is concluded that the proposed adaptation to the PDE model adjusted well to the experimental data and that the 2 xi application of Inverse Problems to perform the hydrodynamic characterization of reactors is a good alternative.