The IoTricity Solar Panel Monitor is a smart IoT + ML project that optimizes the energy efficiency of solar panels.
Using Light Dependent Resistors (LDRs) and servo motors, the system tracks the sun’s movement to adjust panel orientation.
Unlike traditional trackers, this system features a dual approach:
- Rule-Based Local Control → Works on Arduino logic for real-time tracking, even without internet.
- Cloud-Integrated ML Control → Sends sensor data to a FastAPI backend deployed on the cloud, where trained ML models predict the optimal horizontal and vertical tilt of the solar panels.
The project ensures maximum energy capture under varying light conditions, while also providing IoT dashboard integration for monitoring.
This makes it ideal for hackathon prototyping, student projects, and a foundation for scalable smart energy solutions.
IoTricity Solar Panel Monitor is an IoT + ML powered project designed for smart solar tracking.
It monitors light intensity, panel orientation, and environmental parameters to automatically optimize the tilt of solar panels for maximum efficiency.
We implemented two approaches:
- 🔹 Rule-Based Local Tracking → Uses LDR sensors and servo logic (Arduino).
- 🔹 Cloud-Integrated ML Tracking → Uses ESP32 + FastAPI ML API to predict the best panel angles in real time.
- Real-time LDR sensor data acquisition.
- Dual-mode: Rule-based fallback & ML-based cloud tracking.
- Servo-based alignment of solar panels.
- Cloud-hosted FastAPI model inference.
- IoT dashboard ready (ThingSpeak integration possible).
- Sweeping fallback in low-light conditions.
train_model.ipynb: Trains the XGBoost model for predicting optimal panel angles, saves to.joblib.preprocessing.ipynb: Processes LDR sensor data and visualizes panel alignment performance.
Hardware:
- ESP32 / Arduino UNO + Wi-Fi module
- LDR Sensors × 4
- Servo Motors × 2
- Breadboard, Jumper Wires
Software:
- Arduino IDE (C++)
- FastAPI (Python)
- Scikit-Learn (ML models)
- Joblib (model storage)
- Cloud Deployment → Render
- IoT Dashboard → ThingSpeak
Strengths
- Dual tracking logic (rule-based + ML-based).
- Low-cost prototype (~₹1500–2200).
- Cloud integration for remote monitoring.
Weaknesses
- Relies on Wi-Fi availability for ML mode.
- Prototype scale, not yet industrial grade.
- Accuracy limited by LDRs (could use better sensors).
Opportunities
- Scale for large solar farms.
- Integrate with predictive weather APIs.
- Use AI-based energy optimization for smart grids.
Threats
- Market competition with commercial trackers.
- Harsh weather can damage low-cost sensors.
- Dependency on cloud costs for scaling.
- Upgrade LDRs → Pyranometer / Lux sensors for higher accuracy.
- Add battery storage monitoring and MPPT logic.
- Integrate edge ML models (run directly on ESP32).
- Develop mobile app dashboard for users.
- Scale system for smart city solar farms.
- LDR sensors detect light levels from four quadrants.
- Data is sent to:
- Local Logic (Arduino) → compares differences → moves servos.
- Cloud ML API (ESP32) → predicts best angles → servos align.
- IoT dashboard shows real-time panel data.
📢 Developed for IoTricity-S2 Hackathon by Team Loading…