<![CDATA[Fire detection systems commonly rely on a single variable, such as smoke or flame, to detect fires. However, using a single sensor or threshold-based method for early detection is prone to false alarms. To address this issue, this study proposes an early fire detection system using an artificial neural network based on the Radial Basis Function Network (RBFN) architecture. The aim of this research is to minimize false alarms by implementing an early fire detection system that not only detects flames but also hazardous gases, temperature, and humidity as potential sources of fire. A multisensor system comprising an IR flame sensor, gas sensors MQ-9, MQ-2, and MQ-4, as well as a DHT-11 temperature and humidity sensor, was integrated and processed using an RBFN-based ANN on a Raspberry Pi 3. The ANN processes a series of datasets trained to generate a model that determines fire conditions. Testing results showed that the proposed method did not produce any false alarms, with a response time of 2.1 seconds from the ignition of a fire source to the issuance of a warning.]]>
Copyrights © 2025
