Monitoring System for Early Detection of Fire in Wetlands based Internet of Things (IoT) using Fuzzy Methods

Abstract

In Indonesia, especially the province of South Kalimantan is a province with a high frequency of forest and wetland fires. These fires seem to be a routine annual natural disaster. BNPB (National Disaster Management Agency) and BPBD (Regional Disaster Management Agency) Central Kalimantan recorded that during 2014 to early 2019, the largest area of forest and wetlands fires in Central Kalimantan occurred in 2015 with a total burned area of 583,833.44 ha. The fire early detection monitoring system that will be made based on Arduino Uno uses the MQ-2, DHT-22 sensor, and flame detector sensor and is integrated with the Thingspeak IoT, which functions to receive data sent from the Arduino Uno microcontroller using the ESP8266 module, then Thingspeak will send the data to an Android smartphone in the form of a real-time display and also a pop-up notification when a potential fire is detected. Finally, the fuzzy method is also added so that later the reading data from the MQ-2, DHT-22, and flame detector sensors will be more accurate before being sent to Thingspeak. The level of accuracy obtained from the readings of the three sensors against the tests that have been carried out is 90%.