🌍 AI-Based Real-Time Air Pollution Monitoring System
ESP32 + IoT + n8n Automation + Telegram Alerts + ThingSpeak + AI Prediction
1. 🚀 Project Overview
This system monitors air quality using ESP32 sensors and sends real-time data to cloud platforms.
It uses AI to predict pollution levels and sends Telegram alerts with voice notifications.
2. 🧰 Components List
- ESP32 Development Board
- MQ-135 Air Quality Sensor
- DHT11 Temperature & Humidity Sensor
- Jumper Wires
- WiFi Connection
- ThingSpeak + Google Sheets + Telegram Bot
3. 🏗 System Architecture
MQ Sensors → ESP32 → WiFi → ThingSpeak
↓
n8n Automation
↓
Google Sheets + Telegram Alerts + AI Prediction
4. ⚡ Circuit Diagram (Connections)
- MQ135 → GPIO 34
- DHT11 → GPIO 4
- VCC → 5V
- GND → GND
5. 💻 ESP32 Code
#include <WiFi.h>
#include "DHT.h"
#define DHTPIN 4
#define DHTTYPE DHT11
DHT dht(DHTPIN, DHTTYPE);
const char* ssid = "YOUR_WIFI";
const char* password = "YOUR_PASSWORD";
const char* server = "api.thingspeak.com";
String apiKey = "YOUR_API_KEY";
int mq135Pin = 34;
void setup() {
Serial.begin(115200);
dht.begin();
WiFi.begin(ssid, password);
while (WiFi.status() != WL_CONNECTED) {
delay(500);
}
}
void loop() {
float h = dht.readHumidity();
float t = dht.readTemperature();
int airValue = analogRead(mq135Pin);
float AQI = map(airValue, 0, 4095, 0, 500);
WiFiClient client;
if (client.connect(server, 80)) {
String url = "/update?api_key=" + apiKey +
"&field1=" + String(AQI) +
"&field2=" + String(t) +
"&field3=" + String(h);
client.print("GET " + url + " HTTP/1.1\r\nHost: api.thingspeak.com\r\nConnection: close\r\n\r\n");
}
delay(15000);
}
6. 🔁 n8n Workflow Logic
Webhook → Google Sheets → IF AQI Check → Telegram Alert
7. 🤖 Telegram Bot Setup
- Open Telegram → BotFather
- Create bot using /newbot
- Copy API Token
8. 📊 Google Sheets Setup
- Create sheet with columns: Time, AQI, Temperature, Humidity
- Enable Google Sheets API
- Connect in n8n
9. ☁️ ThingSpeak Setup
- Create channel
- Fields: AQI, Temp, Humidity
- Use API Key in ESP32 code
10. 🧠 AI Prediction Logic
from sklearn.linear_model import LinearRegression
import pandas as pd
data = pd.read_csv("aqi.csv")
X = data[["temperature", "humidity", "aqi"]]
y = data["future_aqi"]
model = LinearRegression()
model.fit(X, y)
11. 🔊 Voice Alerts
- Use Google TTS API or ElevenLabs API
- Convert alert text → voice
- Send via Telegram Bot
12. 🚨 AQI Alert Levels
- 0–50 → Good
- 51–100 → Moderate
- 101–150 → Unhealthy for sensitive people
- 151–200 → Unhealthy
- 200+ → Hazardous
13. 🚀 Future Enhancements
- Mobile App Dashboard
- LSTM AI Forecasting
- GPS Pollution Mapping
- Edge AI on ESP32-S3
✔ Project Completed: AI Agentic IoT Air Pollution System
