Utility-scale solar farms utilise large single-axis trackers to follow the sun and maximise energy capture. However, these mechanical systems are subjected to extreme weather and constant wear. When components degrade, the resulting downtime causes significant financial and energy losses. Currently, the industry struggles to identify these faults until catastrophic failure occurs.
This project offers a hands-on introduction to predictive maintenance. Instead of relying on physical inspections, you will use real-world SCADA telemetry data (such as motor angles, current draw, and wind speed) to identify failing trackers remotely. You will build a simple dynamic model of how a healthy tracker should behave, and write a script to flag when the real-world data deviates from that baseline.
This is an excellent opportunity to work with proprietary, utility-scale data and develop highly transferable skills in time-series analysis and fault detection. We will provide the foundational physics and guide you through the data processing pipeline.
School
Photovoltaic and Renewable Energy Engineering
Research Area
Photovoltaic systems | Data science | Data driven engineering
Suitable for recognition of Work Integrated Learning (industrial training)?
No
- Research environment
- Expected outcomes
- Supervisory team
- Reference material/links
The project will be conducted within the ACDC research group at UNSW, providing access to a collaborative research environment and support from peers for discussion, problem-solving, and knowledge exchange.
You can use Python (Pandas, NumPy, SciPy, pvlib) or MATLAB. We will supply a small proprietary utility-scale SCADA telemetry dataset for you to work on.
- A simple mathematical model of a single-axis solar tracker.
- An algorithm capable of identifying mechanical degradation from raw telemetry data.
- Overview of single-axis solar tracker mechanics (https://www.mdpi.com/2076-3417/12/19/9682)