LFT formulation of an anaerobic digestion model for parameters identification

Anaerobic digestion includes numerous biological and physical-chemical processes, whose complexity and potential instability is well known. The Anaerobic Digestion Model by IWA (ADM1) is the state of the art model to describe anaerobic digestion, nevertheless simpler models, such as the AMOCO, had been proposed which are easier to identify and more suitable for process control purposes. In this paper, the application of the Linear Fractional Transformation procedure to identify a subset of model parameters of an AMOCO model, applied to waste activated sludge digestion modelling, is presented. The ADM1 was used to produce synthetic data-sets for parameter identification and further validation. We assumed that volatile fatty acids, methane production and overall volatile solids were the available experimental measurements. Those sole data-sets were sufficient to identify the most uncertain stoichiometric and kinetic parameters. Validation results show a satisfactory matching between the ADM1 synthetic data and those predicted by the identified AMOCO, thus opening up interesting perspectives for its use in process control design.