A high-resolution three-year dataset supporting rooftop photovoltaics (PV) generation analytics

Area of Science:

• Renewable Energy Systems
• Urban Climatology
• Data Science

Background:

• Accurate data is crucial for optimizing photovoltaic (PV) power generation in urban settings.
• Existing datasets may lack the granularity or specific urban context needed for advanced analysis.
• Smart analytics require standardized metadata for efficient integration and application development.

Purpose of the Study:

• To present a novel, open-source dataset for photovoltaic power generation in urban environments.
• To provide high-resolution, multi-year data encompassing both PV output and meteorological conditions.
• To facilitate research in PV performance analysis, forecasting, and system design.

Main Methods:

• Collected 5-minute inverter-level PV power data from 60 rooftop stations in Hong Kong (2021-2023).
• Gathered 1-minute on-site meteorological data using dedicated weather stations.
• Utilized the Brick schema for metadata representation to enhance data interoperability and analytics.

Main Results:

• A comprehensive dataset covering three years of urban PV generation and weather data is now available.
• The dataset includes detailed site specifications, collection methodologies, data records, and validation information.
• Metadata is standardized using the Brick schema, enabling easier integration with smart building and energy analytics platforms.

Conclusions:

• The released dataset is a valuable resource for advancing urban solar energy research and development.
• It supports diverse applications including PV benchmarking, degradation studies, fault detection, and forecasting.
• The open-source nature and standardized metadata promote wider adoption and innovation in the solar energy sector.