PV performance and technology choice
Which modules, mounting structures, and trackers give the lowest cost of electricity at a given site, and how climate, albedo, and dust change the answer.
Module datasheets are rated at standard test conditions that rarely occur in the field. Temperature, spectrum, ground reflectance (albedo), and dust all change how much energy a system produces, so the cheapest technology on paper is not always the best choice for a site. My work compares technologies on their energy yield and cost of electricity at real locations.
What I have worked on
- Bifacial modules and trackers worldwide. A global analysis of fixed-tilt, single-axis, and dual-axis systems with monofacial and bifacial modules found that bifacial modules on single-axis trackers increase energy yield by 35% on average and give the lowest levelised cost of electricity (LCOE) on 93.1% of the world's land area. Dual-axis trackers produce the most energy, but their cost is still too high for them to be cost-effective. An earlier study of 55 locations showed that whether bifacial modules pay off on fixed structures depends on latitude and albedo: below 40° latitude, they need a minimum albedo of about 0.12 to 0.30, depending on the location.
- How trackers should track. Following the best orientation rather than the sun gained less than 1.8% in irradiance below 60° latitude, so the extra sensors may be hard to justify at the latitudes of Southeast Asia and the Middle East.
- Test conditions that match the climate. With colleagues at SERIS, I derived climate-specific reference conditions from long-term weather data at four test sites, in Singapore, Australia, Germany, and China. They differ noticeably from standard test conditions, and the technology that yields the most energy depends on the local climate.
- Soiling. With colleagues at NUS, I worked on estimating soiling losses from surveillance-camera images using deep learning, including a version that runs on low-cost edge devices at the site. Dust and other deposits can reduce output by up to 50%.
- Agrivoltaics in arid climates. A review with colleagues at the Qatar Environment and Energy Research Institute (QEERI) estimated the agrivoltaic potential of cultivated arid and semi-arid land at 139 to 231 TWp; deployment on even 1% of that land could deliver 1.3 to 2.3 TWp.
- Cell and module design. Optimisation of the metal contact design of solar cells and modules to raise efficiency, reduce silver use, and suit the conditions where the modules will be installed.
- Performance in desert conditions. At QEERI, I work on PV, battery storage, and grid technologies in desert conditions, and I contributed to the IEA PVPS Task 13 report on optimising PV systems for different climates (2025).
Where I have applied it
Technology selection and energy yield assessment are part of every feasibility study I have worked on, from floating PV in the Philippines, Sri Lanka, and Suriname to ground-mounted and rooftop PV for developers and investors. I have also analysed the operating performance of more than 6 MWp of floating and ground-mounted PV systems with storage. More on my experience.
Questions I help answer
- Monofacial or bifacial modules, fixed tilt or trackers: which gives the lowest LCOE at this site?
- How far will real output depart from datasheet ratings in this climate?
- How much energy does soiling cost at this site, and how can it be monitored?
- Which yield assumptions should a lender accept for a technology with little local track record?
Publications
- 2026Real Testing Conditions (RTC) for photovoltaic systems: A climate-responsive framework based on the TruePower™ AllianceSolar Energy 316, 114944 · doi:10.1016/j.solener.2026.114944
- 2026Agrivoltaics for arid climates: Developments, technological trends, and global data-driven technoeconomic assessmentsRenewable and Sustainable Energy Reviews 238, 117056 · doi:10.1016/j.rser.2026.117056
- 2025Optimisation of Photovoltaic Systems for Different ClimatesIEA PVPS Task 13, Report IEA-PVPS T13-32:2025 · doi:10.69766/QSYC8858
- 2025Floating Photovoltaic Power Plants: A Review of Energy Yield, Reliability, and MaintenanceIEA PVPS Task 13, Report IEA-PVPS T13-31:2025
- 2023SoilingEdge: PV soiling power loss estimation at the edge using surveillance camerasIEEE Transactions on Sustainable Energy 15(1), 556–566 · doi:10.1109/TSTE.2023.3320690
- 2021Deep-learning-based probabilistic estimation of solar PV soiling lossIEEE Transactions on Sustainable Energy 12(4), 2436–2444 · doi:10.1109/TSTE.2021.3098677
- 2021Techno-economic performance modelling of bifacial and tracking PV systems worldwideIEEE 48th Photovoltaic Specialists Conference (PVSC) · doi:10.1109/PVSC43889.2021.9518852
- 2020Global techno-economic performance of bifacial and tracking photovoltaic systemsJoule 4(7), 1514–1541 · doi:10.1016/j.joule.2020.05.005
- 2020On the PV tracker performance: tracking the sun versus tracking the best orientationIEEE Journal of Photovoltaics 10(5), 1474–1480 · doi:10.1109/JPHOTOV.2020.3006994
- 2019PV-GO: A multiobjective and robust optimization approach for the grid metallization design of Si-based solar cells and modulesProgress in Photovoltaics: Research and Applications 27(2), 113–135 · doi:10.1002/pip.3036
- 2018Monofacial vs bifacial Si-based PV modules: Which one is more cost-effective?Solar Energy 176, 412–438 · doi:10.1016/j.solener.2018.10.012
- 2018On the grid metallization optimization design for monofacial and bifacial Si-based PV modules for real-world conditionsIEEE Journal of Photovoltaics 9(1), 112–118 · doi:10.1109/JPHOTOV.2018.2882188