SmartLPG: AI-Powered Real-Time LPG Monitoring, Prediction & Safety System

Inspiration

LPG cylinders are used daily in homes, restaurants, and businesses, yet users often have no reliable way to know how much gas remains. Unexpected gas depletion causes inconvenience, while gas leaks pose serious safety risks. We wanted to create a smart solution that not only monitors LPG levels but also predicts future consumption and enhances safety through real-time leak detection.

What It Does

SmartLPG is an AI-powered IoT system that continuously monitors LPG cylinder weight using a load cell and HX711 module. An MQ2 gas sensor detects LPG leaks, while an ESP32 microcontroller sends real-time data to Firebase. Users can monitor gas levels, usage history, leak status, and refill predictions through a web dashboard.

The system estimates the remaining gas percentage using:

$$ Gas\ Percentage = \frac{Current\ Weight - Empty\ Weight}{Full\ Weight - Empty\ Weight} \times 100 $$

Machine Learning algorithms analyze historical usage data to predict future consumption and estimate refill dates.

How We Built It

Hardware

  • ESP32 Development Board
  • Load Cell
  • HX711 Amplifier Module
  • MQ2 Gas Sensor
  • Solenoid Valve
  • Buzzer
  • LEDs
  • 3D Printed LPG Cylinder Model

Software

  • Arduino IDE
  • Firebase Realtime Database
  • HTML, CSS, JavaScript
  • Machine Learning (Linear Regression)
  • EmailJS and Twilio for notifications

The ESP32 collects sensor data and uploads it to Firebase. The dashboard retrieves this information and displays real-time analytics, refill predictions, and safety alerts.

Challenges We Faced

One of the biggest challenges was accurately calibrating the load cell to provide stable weight measurements. Integrating multiple hardware modules with ESP32 and ensuring reliable cloud communication was another challenge. We also worked on improving prediction accuracy by cleaning and analyzing sensor data for machine learning.

What We Learned

Through this project, we gained hands-on experience in IoT system design, sensor integration, cloud databases, web development, and machine learning. We also learned how predictive analytics can be combined with real-time monitoring to solve practical problems and improve user safety.

Future Scope

  • Automatic LPG refill booking
  • Mobile application support
  • Smart home integration
  • Voice assistant compatibility
  • Advanced AI models for improved prediction accuracy

Impact

SmartLPG helps users monitor LPG consumption, prevents unexpected gas shortages, improves safety through leak detection, and reduces gas wastage. By combining AI, IoT, and cloud technologies, it creates a smarter, safer, and more efficient LPG management ecosystem.

Built With

  • analytics
  • arduino
  • artificial
  • c++
  • chart.js
  • cloud
  • computing
  • css
  • data
  • database
  • esp32
  • firebase
  • html
  • hx711
  • intelligence
  • internet
  • iot
  • javascript
  • learning
  • libraries
  • library
  • machine
  • of
  • predictive
  • pythonfirebase
  • real-time
  • realtime
  • scikit-learnfirebase
  • sdk
  • things
Share this project:

Updates