Development of Artificial Neural Network (ANN)-Based Model to Predict Electricity Production from Food Waste Biogas System

Abstract

This paper has been carried out to investigate the electrical production from biogas of food waste in Batu Pahat area, Johor, Malaysia. The objectives of this research are to identify the potential of food waste in Batu Pahat, Johor for electricity production using Artificial Neural Network (ANN) technique. The data of food waste was collected in 2016 and 2017. The prediction using ANN are divided into two model that are Model 1 for prediction of biogas production (m3) while Model 2 for prediction of electricity production (MWh). The input for Model 1 were the average food waste, daily feedstock volume, feedstock retention time, temperature, yield factor, and digester volume and for input in Model 2 are gas leaking, energy for plant operation, efficiency in electricity production and biogas production. As the results, the highest electricity was produced in 2016 that was produced maximum that is 613.87 MWh per month. It shows that, the higher the volume of average food waste, the highest the electricity in MWh.