Sustainable methane energy from bagasse treated via bokashi technology: comparative between neural network and mathematical modeling

Abstract

Bagasse is the major leftover material from the sugarcane industry, and it has significant untapped energy. Biogas production from bagasse is employed as eco-friendly energy but its intricate composition makes it resistant to degradation. This study endeavors to explore the impact of bokashi technology, a technique that applies effective microorganisms on the potential methane production from bagasse. According to findings, bagasse had the ability to produce biogas but applying bokashi technology to bagasse led to getting more methane production. The methane production from treated bagasse for one month via bokashi bran was 243.80 LCH4/kgVS compared to 106.84 LCH4/kgVS only from fresh bagasse which is often attributed to improved fibrous carbohydrates degradation by the pre-treatment process. The reduction of total solids and chemical oxygen demand were more with treated bagasse. Two-dimensional mathematical modeling (TDMM) and artificial neural network (ANN) were utilized to forecast the production of methane through the anaerobic co-digestion process. The main advantage of ANN model is its ability to be constructed and trained for any experiment, regardless of the availability of a pre-existing study or understanding of the underlying phenomena. On the other hand, existence of a mathematical model that accurately describes the behavior of the current experiment is a fundamental requirement for constructing the TDMM model. The TDMM model remains stable in each run, as it relies on the established mathematical equations. On the other hand, ANN model may exhibit variations in each run due to the random initialization of weights.Graphical abstract