Your solar plant's early-warning doctor.
SolarDoctor analyses data from photovoltaic systems with an ensemble of machine-learning models to detect faults early, protect energy output and plan maintenance before small issues become big losses.
The problem
Undetected faults quietly drain solar output.
Malfunctions in panels and components reduce energy production and raise maintenance costs. Traditional methods often miss the many different situations a PV system can face.
fault detection accuracy of our ensemble model.
algorithms combined: random forest, k-nearest neighbours and gradient boosting.
continuous monitoring for instant fault detection and maintenance scheduling.
Source: the founding research paper (International Journal of Reconfigurable and Embedded Systems, 2025).
How it works
How it works
A clear path from data to decision.
Connect
Link SolarDoctor to inverter and monitoring data.
Learn
The ensemble model learns normal operating patterns.
Detect
Faults are flagged as soon as patterns deviate.
Fix
Maintenance teams receive prioritised alerts.
Product
What makes it different.
Every feature is grounded in the founding research.
Ensemble intelligence
Combines GB's handling of complex patterns, KNN's local structure and RF's robustness to noise.
95% accuracy
Substantial improvement over individual models.
Early warning
Catch faults before they cut production.
Smarter maintenance
Schedule repairs based on real system condition.
Who it's for
Who it's for.
Solar farm operators
Protect output across large plants.
O&M contractors
Prioritise crews where they matter most.
Commercial rooftops
Keep self-generation running at full capacity.
The research
Built on peer-reviewed research.
- Proposes an ensemble model for fault detection in PV systems.
- Combines random forest, k-nearest neighbours and gradient boosting.
- Ensemble outperforms individual models in handling diverse PV conditions.
- Achieved 95% fault detection accuracy.
Ibrahim, M., et al. (2025). Enhanced fault detection in photovoltaic systems using an ensemble machine learning approach. International Journal of Reconfigurable and Embedded Systems, 14(2), 507–517.
Team
The team behind it.
Founded in 2025 by a student team from the faculty below.
Faculty of Engineering Technologies
Solar systems, data acquisition and machine learning.
FAQ
Questions we hear often.
What data does it need?
Standard electrical data from inverters and monitoring systems.
Does it work for small systems?
Yes. It can be applied to rooftop and utility-scale plants.
Is this a finished commercial product?
It is a research-based student startup moving from validated prototype toward pilots with real users.
Can we partner or run a pilot?
Yes. We are looking for organisations willing to test the solution in real conditions and share feedback.
Bring SolarDoctor to your organisation.
We are onboarding pilot partners now. Read the research and get in touch.