Assessment of BIPV power generation potential at the city scale based on local climate zones: Combining physical simulation, machine learning and 3D building models

Haida Tang, Xingkang Chai, Jiayu Chen, Yang Wan, Yuqin Wang, Wei Wan and Chunying Li

Renewable Energy, 2025, vol. 244, issue C

Abstract: The adoption of distributed photovoltaic (PV) in cities can alleviate energy shortages, and building integrated photovoltaic (BIPV) has multiple advantages including building material saving and space saving. Predicting the potential of annual BIPV power generation (BIPVPG) and exploring the influencing factors in built environment is of great significance. This study created 50 simplified LCZ models based on the indicator ranges of LCZ categories 1–10. These models assume uniform building height and distribution to facilitate the simulation of annual BIPVPG. The annual BIPVPG (with PV materials applied on roofs and vertical facades) and the urban heat island (UHI) effects of the LCZ models in 15 cities were simulated. The UHI effects of the LCZ models were used to account for the reduction in BIPV efficiency caused by urban heat islands. Following that, the impact of climate and urban morphological factors on BIPVPG was studied using a multiple linear regression (MLR) and random forest (RF) model. According to the results, the RF model performed better in BIPV power generation prediction. Among urban morphological factors, average building height (ABH), aspect ratio (AR), and sky view factor (SVF) have dominant impact on urban BIPVPG. The impact on BIPVPG is minimal when ABH is between 10 and 15 m, and AR is between 1.5 and 2. When SVF is less than 0.8, the BIPVPG per unit building roof and wall area can be improved with building surface fraction (BSF) below 80. Among climate factors, solar radiation has significant impact on urban BIPVPG over other factors. Finally, based on the Guangzhou building data, the validated RF model was applied to predict Guangzhou's annual BIPV power potential. This study provides insights for machine learning models in BIPVPG assessment and offers quantitative recommendations for decision-makers and urban planners in developing BIPV cities with high energy resilience and sustainable level.

Keywords: BIPV power generation; Local climate zone; Machine learning; Random forest (search for similar items in EconPapers)
Date: 2025
References: Add references at CitEc
Citations:

Downloads: (external link)
http://www.sciencedirect.com/science/article/pii/S0960148125003507
Full text for ScienceDirect subscribers only

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:eee:renene:v:244:y:2025:i:c:s0960148125003507

DOI: 10.1016/j.renene.2025.122688

Access Statistics for this article

Renewable Energy is currently edited by Soteris A. Kalogirou and Paul Christodoulides

More articles in Renewable Energy from Elsevier
Bibliographic data for series maintained by Catherine Liu ().