IoT Fire Detection System

This project implements a Smart IoT Fire Detection System using an ESP32 microcontroller. It reads data from environmental sensors (temperature, humidity, smoke, flame) and sends it to an MQTT broker. The data is stored in InfluxDB and visualized in real time with Grafana. All services are containerized using Docker Compose.

Feature

• Real-time monitoring of temperature, humidity, smoke and flame.
• Wireless communication via Wi-Fi.
• Local alarm via LEDs and buzzer when danger is detected.
• MQTT communication with Mosquitto broker.
• Data stored as time-series in InfluxDB.
• Interactive Grafana dashboards.
• Fully containerized setup using Docker Compose.

Hardware Requirements

• ESP32 board
• DHT22 sensor (Temperature & Humidity)
• MQ-2 smoke sensor
• Flame sensor
• Buzzer
• 3 LEDs

Software Requirements

• Arduino IDE
• Libraries: WiFiManager, PubSubClient, DHT
• Docker & Docker Compose

MQTT Topics

• sensors/temperature
• sensors/humidity
• sensors/smoke
• sensors/flame

System Architecture

1. ESP32 reads sensor values.
2. Sends values to MQTT broker (Mosquitto).
3. Telegraf subscribes to MQTT and pushes data to InfluxDB.
4. Grafana fetches and visualizes the data from InfluxDB.

Setup Instructure

1. Flash ESP32

• Open fire_detector.ino in Arduino IDE.
• Install required libraries (DHT, WiFiManager, PubSubClient).
• Upload the code to the ESP32.
• On first boot, the ESP32 starts in Wi-Fi Access Point mode.
• Connect to the ESP32's Wi-Fi network (e.g., FireDetector-Setup) and open the configuration portal.
• Choose your local Wi-Fi network and enter your Wi-Fi credentials.
• The ESP32 saves the settings and will automatically reconnect to the same Wi-Fi and MQTT broker in future restarts.
• Εnter the MQTT broker IP address in the code (const char* mqtt_server = " ";).

2. Run the Docker Stack

docker-compose up -d

This will start:

• Mosquitto MQTT broker
• Telegraf collector
• InfluxDB time-series database
• Grafana dashboard server

3. InfluxDB Configuration

When you run InfluxDB for the first time, it will initialize a fresh setup.

1. Open your browser and go to: http://localhost:8086
2. Create the following:
   • Organization: myorg
   • Bucket: fire_data
   • Token: mytoken

4. Access Services

• InfluxDB: http://localhost:8086
• Grafana: http://localhost:3000
  (Default login: admin / admin)

Visualization

Grafana dashboards display:

• Temperature and Humidity
• Smoke levels
• Flame levels

Screenshots