High precision models, whether physical or AI based, are widely used for detecting underperformance in utility scale solar plants. However, much of the data for modern plants is not effectively utilised due to the operation of the inverters. Most systems are "oversized", such that the inverter cannot effectively use the maximum power of all connected panels during periods of high irradiance, which results in "clipping". This is where the system output is fixed at the inverter capacity. In some cases this can occur for the majority of a clear-sky day.
In this project the student will join a team looking at the development of high precision modelling methods that allow this data to be used for fault detection by accurately modelling the expected current and voltage characteristics at the inverter. It is expected this can be leveraged to create pseudo I_V curves that will not only aid in detection of underperformance but aid in diagnosing the underlying causes.
Students with backgrounds in renewable energy, electrical engineering, computing science or software engineering are encouraged to apply.
School
Photovoltaic and Renewable Energy Engineering
Research Area
Renewable energy | Data analysis | Electrical modelling
- Research environment
- Expected outcomes
- Supervisory team
- Reference material/links
The student will work as part of the modelling group at SPREE, under the direction of Professor Bram Hoex. They will be supported by a team of 4 postdocs and HDR students. The group works with data from commercial partners, as well as publically available datasets. The work will consist of data analysis, programming and physical modelling, with support provided for area's outside the student's core competence.
Demonstration of the ability to predict I_V characteristics of a PV system under inverter clipping and re-creation of a pseudo I_V curve. Quantification of the uncertainty in the resulting model parameters and suggestions for improvement. Students will experience a research environment including interaction with commercial partners, as well as broaden their existing skills to cover new areas.
- 2024 Solar Risk Assessment, kWh Analytics, Available at https://kwhanalytics.com/industry-report/
- Balfour, John, et al. "Masking of photovoltaic system performance problems by inverter clipping and other design and operational practices." Renewable and Sustainable Energy Reviews 145 (2021): 111067.
- Micheli, Leonardo, et al. "Quantifying the impact of inverter clipping on photovoltaic performance and soiling losses." Renewable Energy 225 (2024): 120317.