Are you eager to make a meaningful contribution to the field of renewable energy while gaining invaluable data analytics skills for your future career? We invite you to embark on a transformative journey with us as we delve into the fascinating realm of solar power systems powered by Machine Learning.

Automatic detection of faults in solar power systems allows for maximised energy production, reduced downtime and maintenance costs, and grid stability. Our project focuses on advancing the capabilities of fault detection and diagnostics in utility-scale PV systems through the utilisation of multivariate time series analysis and state-of-the-art Deep Learning approaches such as generative AI models. This research aims to enhance the monitoring and maintenance of such systems by developing an automated approach to identifying potential faults or anomalies using a combination of various time series signals.

In this project, you will collaborate with world-class Computer Scientists, and gain hands-on data analytics skills on real-world industrial datasets. Candidates with Computer Science, Electrical Engineering, or Statistics background, who possess a burning curiosity and wish to push boundaries, are highly desirable. Don't miss out on this incredible opportunity to pioneer advancements in renewable energy research while acquiring skills that will define your future success.

School

Photovoltaic and Renewable Energy Engineering

Research Area

Time-series | Machine learning | Deep learning | Signal processing | Python programming

  • Office environment with access to GPU server
  • Collaboration with the School of Computer Science and Engineering
  • Investigate correlations between various time series signals including inverter power, irradiance and temperature to identify patterns that indicate potential faults in power generation or sensor measurements in utility-scale PV systems.
  • Investigate the potential of applying semi-/unsupervised learning approaches (such as Autoencoders) to time-series data to capture the outliers presented in the latent space.
  • Gain hands-on data analytics and programming skills on real-world datasets that are highly applicable to industrial settings.
Dr. Tien-Chun Wu is research fellow in Machine Learning for photovoltaic Tien-Chun Wu
Dr. Tien-Chun Wu is research fellow in Machine Learning for photovoltaic