Beyond generation: Spatiotemporal dynamics of rooftop photovoltaic electricity accommodation in China's urban building stock

Rooftop photovoltaic deployment is crucial for urban decarbonization, yet the spatiotemporal variability of buildings’ potential to accommodate photovoltaic electricity remains insufficiently quantified. A bottom-up framework is developed for 30 Chinese cities to model hourly photovoltaic supply and building electricity demand, evaluating accommodation ratios under building-level self-accommodation and citywide sharing modes. Demand is decomposed into base loads and space heating and cooling loads, and representative cities, typical days, and typical buildings are selected to explore seasonal and intraday matching mechanisms. Results show that annual accommodation ratios range from 32.4% to 65.3% for self-accommodation and 60.1% to 94.9% for citywide sharing, revealing a “south-high, north-low” spatial gradient, driven by climate and solar resources. Monthly accommodation ratio exhibits “W-shaped” or “inverted U-shaped” profiles governed by a “competitive–dominant” relationship between base and space heating and cooling loads. Typical day analysis identifies an “exporter–absorber” pattern: residential buildings are generally photovoltaic surplus-prone at midday, while non-residential buildings can largely accommodate local photovoltaic electricity. An “accommodation paradox” is revealed: energy conservation measures may reduce photovoltaic accommodation potential. Accommodated electricity could reduce city-scale carbon dioxide emissions by 0.95%–24.64%, underscoring the need for flexibility-oriented operations and policies, electric vehicle–building interaction, and mixed building type microgrids.