Predicting Biogas Quality for Renewable Energy System: An Implementation of Gaussian Naïve Bayes Classifier

The utilization of organic household waste for biogas production presents a promising alternative energy solution; however, it is often limited by the lack of intelligent monitoring and control systems. This study proposes the design and implementation of an Internet of Things (IoT)-based monitoring and control system integrated with a Gaussian Naïve Bayes (GNB) algorithm to classify biogas quality in real time. The system employs an ESP32 microcontroller combined with a K-Type thermocouple sensor, MQ-4 and MQ-135 gas sensors, and an MPX5700 pressure sensor to collect environmental data during the biogas production process. A dataset consisting of 120 samples was collected and divided into training (80%) and testing (20%) sets. The GNB model classifies biogas into three categories: Good, Moderate, and Poor. Experimental results show that the model achieved an accuracy of 95.83%, with high precision and recall across all classes. The system also demonstrated an energy conversion efficiency of 33.3% when converting biogas into electrical energy to power a fan. These results indicate that the proposed system effectively integrates IoT and machine learning to automate and optimize biogas utilization, providing a scalable and cost-effective solution for renewable energy applications in community-based waste management systems.