Mapping Solar PV Potential for Existing Building Stocks in the UK Using Deep Learning of Satellite and Aerial Image Data Kingston University in United Kingdom

This PhD opportunity at Kingston University invites applications for a research project focused on mapping the solar photovoltaic (PV) potential of existing building stocks across the UK. The project leverages advanced deep learning techniques applied to satellite and aerial imagery, aiming to provide actionable insights for the construction industry and support the UK's net-zero carbon targets by 2050.

Solar PV technology is a leading solution in the renewable energy sector, with rapid growth in UK installations over the past two decades. The research addresses a critical gap: quantifying the extent to which rooftop solar PV panels can be installed on existing buildings in specific UK regions, thereby supporting decarbonisation and renewable electricity generation. The project will assess available rooftop space, estimate regional PV capacity, and produce GIS-based maps to inform strategic planning and policy development.

Key research activities include collecting GIS data, applying deep learning and semantic segmentation to identify suitable rooftop areas from satellite and aerial images, and using GIS analysis software (such as QGIS or ArcGIS) to statistically estimate and map PV installation capacity. The outcomes will guide the construction industry, government, and stakeholders in promoting solar PV installations and developing effective incentive policies, ultimately fostering sustainable development and economic growth in the solar PV sector.

The ideal candidate will have a strong academic background, holding an Honours Degree above 2:1 (or equivalent) in renewable energy, architectural engineering, GIS, computer science, or related engineering fields. Experience with deep learning, big data analysis, and programming languages (Python, MATLAB) is highly desirable.

This project is part of the Kingston University Graduate School studentships competition for October 2026 entry, offering potential funding for tuition and stipend. For application details, candidates should consult the PhD Studentships and Faculty research webpages. The application deadline is March 4, 2026.

References supporting the research include recent advances in AI and smart vision for construction, as well as studies on estimating solar PV installation capacity using satellite and aerial images.