ARTIFICIAL NEURAL NETWORKS IN OPTIMIZING METHANE PRODUCTION FROM DOMESTIC WASTE DIGESTION

Objective: The viability of methane production depends on a suitable combination of various physical and chemical parameters. This study systematically analyses the process of biogas generation, modelling and optimization and prediction of methane fraction. It further makes an attempt to optimize the parameters for effective biogas production. The required data and information for the optimization process are collected from a biogas plant in Pammal, Chennai, Tamil Nadu operated and maintained by Exnora. Methods: Artificial Neural Networks (ANNs) are one of the latest tools that assist to resolve complex issues that could not be addressed by conventional methods. The present study utilizes the ANN as a tool for simulating and optimizing biogas production process from the floating dome digester of Khadi and Village Industries Commission (KVIC) model biogas plant. Operational data of the plant for duration of 100 days were collected and utilized in the analysis. Findings: This study analyses the effect of Total Solids (TS), Total Volatile Solids (TVS), Organic Loading Rate (OLR), and pH which are considered to be the operational parameters on the biogas yield. A multi-layer ANN model was considered in simulating the digester operation and predicting the methane production.