Hybrid neural modelling of an anaerobic digester with respect to biological constraints

A hybrid model for an anaerobic digestion process is proposed. The fermenta tion is assumed to be performed in two steps, acidogenesis and methanogenes is, by two bacterial populations. The model is based on mass balance equati ons, and the bacterial growth rates are represented by neural networks. In order to guarantee the biological meaning of the hybrid model (positivity o f the concentrations, boundedness, saturation or inhibition of the growth r ates) outside the training data set, a method that imposes constraints in t he neural network is proposed. The method is applied to experimental data f rom a fixed bed reactor.