Neuronal Modeling of a Two Stages Anaerobic Digestion Process for Biofuels Production

Abstract

The biological processes for biofuels generation are highly nonlinear systems submitted to external disturbances and parameter uncertainties which require estimation, control and optimization strategies to maintain stability and optimal production. In this work, a nonlinear neural network for unknown nonlinear systems in the presence of external disturbances and parameter uncertainties is proposed. The objective is to estimate unmeasurable complex variables in a two continuous stages anaerobic digestion process for hydrogen and methane production. Simulation results are presented, where it is demonstrated that the neural model is efficient to calculate complex dynamics of the process in presence of disturbances. As future work, control and optimization algorithms based on the neural model can be developed to biofuel production optimization.