PhD Scholarship in AI-Assisted Monitoring and Inspection of Solar Photovoltaic Power Plants Using Aerial Imagery

This research contributes to computer vision, machine learning, robotics and optimization methods for automating fault/anomaly detection and cleanliness checks for PV systems using visual/thermal images.

Autonomous methods for fault or anomaly detection and classification of PV plants with high accuracy are necessary for the monitoring of large-size PV power plants. Objectives also include other types of inspection, such as detecting hot spots, glass breakage, soiling and cleanliness check.

This co-funded scholarship provides a stipend of $32,841 per annum pro rata (full-time) for 3.5 years, and the successful applicant will also be awarded a Tuition Fee Scholarship if required and comply with Scholarship terms and conditions.

Open now.

Applications will close once the candidate is appointed.

One scholarship is available. 

Eligibility requirements:

  • have a first-class Honours or Master's or equivalent degree in a relevant discipline of engineering/science
  • be an Australian citizen, New Zealand Citizen or an Australian permanent resident or an international student -meeting the minimum English language requirements
  • provide evidence of adequate oral and written communication skills
  • demonstrate the ability to work as part of a multi-disciplinary research team
  • meet RMIT’s entry requirements for the PhD by research degree

Potential candidates should contact Dr Ehsan Asadi (ehsan.asadi@rmit.edu.au) and provide:

  • a cover letter (including research statement and reasons for applying for this scholarship and project)
  • a copy of electronic transcripts
  • a CV that includes any publications/awards, experience relevant to the project and the contact details for 2 referees

PV’s monitoring process can be done intelligently thanks to drone. Remotely operated drones can capture visual and thermal imaging at a very high resolution, suitable for detecting photovoltaic modules and the cleanliness of solar panels. These images and other data can be processed by computer vision and machine learning methods to interpret the photovoltaic (PV) solar farm's condition and perform various inspections and anomaly detection.  The research will draw from state-of-art computer vision and machine learning methods and adapt new algorithms to automate inspection procedures of PV plants.

Given the data captured by a remotely operated drone, we first investigate the required techniques to obtain an overall status of an operational PV array and identify specific PV strings or modules for further detailed analysis and monitoring tasks. Then other methods will be studied and developed for fault or anomaly detection and classification of PV plants with high accuracy to enable comprehensive monitoring of large PV power plants. The rest of the research focuses on automating other inspection types, such as detecting soiling over the panels and evaluating the level of soiling or cleanliness conditions for planning an efficient cleaning cycle. Other inspection methods could include detecting any damage to solar panels like hot spots and large cracks.

Please contact Dr Ehsan Asadi 

ehsan.asadi@rmit.edu.au

aboriginal flag
torres strait flag

Acknowledgement of Country

RMIT University acknowledges the people of the Woi wurrung and Boon wurrung language groups of the eastern Kulin Nation on whose unceded lands we conduct the business of the University. RMIT University respectfully acknowledges their Ancestors and Elders, past and present. RMIT also acknowledges the Traditional Custodians and their Ancestors of the lands and waters across Australia where we conduct our business - Artwork 'Luwaytini' by Mark Cleaver, Palawa.