Abstract
In this research study, the impact of the feedstock to the inoculum (F/I) amount ratio in the dry anaerobic digestion (DAD) of Hura crepitans leaves was evaluated. Measured biogas volumes, as well as the chemical kinetic predictions for exponential, logistic and Gompertz model, depicting the agreement of the simulations over time, were also determined. From the F/I equivalents 2, 4 and 6 at 22 % of packed total solids, which were considered in analysis test procedure, the DAD digester with F/I number 4 was the most promising in terms of biogas’ production rate. Its daily methane/carbon dioxide was 690 mL, while cumulative generation productivity was greater than 2 L/sample, respectively. On the other hand, the DAD reaction with F/I contained 6, recorded the lowest related expressed primary matter of < 1 L. An associated early commencement of the organic material breakdown in all bio vessels was indicative of a good start-up phase, which is one of the challenges, often encountered in DAD process. Furthermore, applied experimental methods revealed the direct correlation phenomena between biodegradability physical constants, measured molecular CH4/CO2 synthesis and simulations. Hura crepitans being an invasive plant species makes its lignocellulosic fractions desired in terms of valorisation, as it is not competing with agricultural crop products. Modelling can, moreover, contribute to consecutive operation optimisation, scaling and integrating, also taking dynamics under consideration. As opposed to bio-refining wood residues, where individual cellulose, hemicellulose or lignin biopolymers can be attained, degradation to yield CH4 is robust, as well as compatible in combustion.
Prediction of biogas production rate from dry anaerobic digestion of food waste: Process-based approach vs. recurrent neural network black-box model
