An Artificial intelligence-powered  Internet of Things-based environmental monitoring system for hazardous gas detection and fire prevention

Leakage of dangerous combustible gas, resulting in numerous fatalities and extensive property damage, is a significant problem worldwide. Identifying the concentration of gaseous composites at the workplace is essential for ensuring safety during the gas spill. In this paper, we propose an adaptive IoT-based environmental monitoring system for this purpose. This research presents a low-cost IoT-based system for harmful and flammable gas detection to prevent diseases and damage caused by the leakage of harmful gaseous compounds. The suggested system comprises a remote terminal server and the wireless sensor node where the sensor nodes continuously detect the presence of flames, temperature, humidity, and gas concentration and transmit uniquely sensed data to the intended server. The data is subsequently processed, stored, and uploaded to the remote server through accessing web applications, to achieve real-time alerts and visualization.The system makes use of the Random Forest algorithm to improve precision accuracy and reduce overfitting. Effectiveness of the overall system is done through deep investigation in both the real time and virtual environment. The performance of the projected system is measured in terms of PDR (Packet Delivery Ratio), network efficiency, and lower latency, demonstrating the viability of the proposed system.