-42%

Smart Post-Harvest Storage Chamber

Original price was: ₹6,500.Current price is: ₹3,799.

Post-harvest losses in fruits, grains, and vegetables are primarily caused by excessive heat, humidity, and gas accumulation—particularly methane (CH₄) and ethylene, which accelerate decay. This project develops an IoT-based intelligent storage chamber that continuously monitors temperature, humidity, and methane concentration inside the chamber. A microcontroller (ESP32 or Raspberry Pi) processes sensor data and dynamically activates ventilation fans or Peltier cooling units to maintain safe thresholds. Real-time data are transmitted to a cloud dashboard via Wi-Fi or LoRa for remote monitoring and predictive analytics. An embedded machine-learning model forecasts spoilage probability based on time-series gas and humidity patterns, optimizing ventilation cycles. The system improves storage life, reduces wastage, and provides actionable insights for warehouse and cold-chain operators.

 

Key Hardware Components

  • Gas Sensor: MQ-4 / MQ-135 (Methane detection, 200–10000 ppm)

  • Temperature & Humidity Sensor: DHT22 / SHT31

  • MCU: ESP32 (Wi-Fi + BLE + ADC)

  • Fan/Exhaust Control: Relay driver (5 V or 12 V)

  • Optional: Peltier cooler, DC exhaust fan, solar power module

  • Display: OLED 0.96” or LCD 1602

  • Communication: LoRa (SX1278) for long-range warehouse setups

🔹 Software & Data Flow

  1. Sensors sample data every few seconds.

  2. Node computes real-time gas concentration trend (Δppm/min).

  3. A trained LightGBM/Decision-Tree model classifies the state as Safe / Warning / Spoilage Risk.

  4. If thresholds exceed limits, the system triggers a relay-controlled exhaust fan or activates cooling.

  5. Data are pushed to Firebase / ThingsBoard / InfluxDB dashboards with time-series plots and alerts.

  6. LoRa nodes can send aggregated data to a central gateway for large warehouses.

🔹 Innovations

  • Methane-driven predictive ventilation using real-time ML on microcontroller.

  • Multi-sensor fusion for humidity-gas-temperature correlation.

  • Self-calibrating thresholds via cloud learning loop.

  • Optional solar-powered controller for rural cold-chain setups.