Probabilistic simulation of biogas production from anaerobic co-digestion using Anaerobic Digestion Model No. 1: A case study on agricultural residue

Abstract

Listen

Translate article

Anaerobic co-digestion is a well-established approach to overcome drawbacks associated with the digestion of individual substrates, but the identification of suitable substrate combinations requires extensive experimental work. We develop a framework for the probabilistic estimation of methane production potential from mixed substrates using Anaerobic Digestion Model No. 1 and validate the proposed framework using agricultural residues as a case study. The validity of anaerobic digestion simulations is often questioned due to the inherent difficulty in parameterizing such complex models. Here, an extensive literature survey was combined with Monte Carlo analysis and a Gaussian Mixture Model approach to account for parameter variability, leading to a probabilistic estimate of steady-state biogas production. Predicted values of biogas yield and methane content were validated using literature data for mono-substrates before extending the simulations to new substrate combinations. Results indicated that co-digestion increases mean methane content by up to 8 % compared to mono-digestion of individual substrates. However, the predicted probability distributions are multimodal: traditional point estimates for methane production rates may be sensitive to small variations in parameter values, yielding misleading results. The results emphasize the necessity of a probabilistic approach in anaerobic digestion modeling for informed decision making, especially in subsequent techno-economic feasibility assessments.