LPG Safety Monitoring System Using Artificial Neural Network with Back Propagation Method Based on The Internet of Things (IoT)

Abidin, Ramadhani
Supriyanto, Amir
Surtono, Arif
Suciyati, Sri Wahyu

Journal of Energy, Material, and Instrumentation Technology
Vol 5 No 3 (2024): Journal of Energy, Material, and Instrumentation Technology

This study aims to create an Artificial Neural Network (ANN) to determine the results of LPG gas monitoring based on gas leakage levels, smoke levels, fires, and ambient temperatures and apply Internet of Things technology in the monitoring system. Variations in the number of nodes in the hidden layer indicate that the ANN performance will be maximal, with seven nodes in the hidden layer with an accuracy value of 99.63%, a precision of 100%, and a loss function of 0.423%. The microcontroller used is NodeMCU ESP32S, with input from the MQ6 sensor to detect LPG gas leaks, an infrared sensor to detect flames, an MQ-2 sensor to detect smoke, and a DHT-22 sensor to measure the ambient temperature. The resulting system output is a monitoring display using the Blynk platform, fans and Buzzers controlling, and WhatsApp notifications. The system will turn on the fan when the detected LPG level exceeds 250 ppm.