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

This dataset includes measured photovoltaic (PV) power generation data and on-site weather data collected from 60 grid-connected rooftop PV stations in Hong Kong over a three-year period (2021-2023). The PV power generation data was collected at 5-minute intervals. The meteorological data was collected at 1-minute intervals from an on-site weather station. The metadata was represented using Brick schema was developed, which simplifies the data comprehension and the development of smart analytics applications. The detailed Brick model is stored in the .ttl file format, which can be accessed for retrieving metadata through the use of SPARQL queries.This dataset can be used in various applications - PV generation benchmarking, PV degradation analysis, PV fault detection, solar radiation and PV power generation forecasting, and the simulation and design of PV systems.

Methods
The data was collected from 60 grid-connected rooftop PV stations and 1 weather station. These stations are located within the Hong Kong University of Science and Technology campus. The university is located in the rural coastal area of Sai Kung District, Hong Kong (22.3363°N 114.2634°E) and covers an area of 60 hectares. For PV stations without panel level optimizers (comprising 23 stations, accounting for 38.3% of the total), the data was individually measured and transferred by the inverter. For PV stations equipped with panel level optimizers (comprising 37 stations, accounting for 61.7% of the total), the PV generation data was measured and transferred by both the inverter and the panel level optimizer. Meteorological data is collected from the weather station located on the eastern side of the campus. The station comprises a 10-meter-high automatic weather tower and an outdoor plinth area that housing 6 samplers that measure meteorological data at 1-minute intervals. The collected PV generation data was transferred to wireless gateway using dedicated Wi-Fi. Meteorological data is initially sensed, transmitted, and stored in the data logger using RS232 or RS485 communication protocols, and subsequently transferred to the server via LAN cables. All streams of PV generation and meteorological data were extracted and consolidated into a centralized database.