Understanding the relationship between rural morphology and photovoltaic (PV) potential in traditional and non-traditional building clusters using shapley additive exPlanations (SHAP) values

Jiang Liu, Changhai Peng and Junxue Zhang

Applied Energy, 2025, vol. 380, issue C, No S0306261924024759

Abstract: Rural areas have a large quantity of rooftops and facades appropriate for installing PV panels. However, the unclear impact of rural morphology on PV potential hinders their effective utilization. To address this challenge, this study examined 300 clusters of traditional and non-traditional rural buildings in Nanjing. 17 morphological indicators were identified, representing plot shape, built density, building form, and terrain variation. The annual PV power generation and Levelized cost of electricity (LCOE) were simulated. Using an explainable machine learning framework (XGBoost algorithm combined with SHAP values), we explored the relationship between rural building morphology and PV potential. The results revealed that mean building height (BH) and floor area ratio (FAR) are key factors for PV power generation, while only BH is crucial for LCOE. As BH and FAR increase, PV generation declines, while LCOE rises. Particularly, BH has a stronger influence on technical potential in traditional clusters, whereas FAR plays a comparable role in non-traditional ones. Using these indicators, rural clusters can be categorized into three typologies for technical potential: low BH-low FAR, high BH-low FAR, and high BH-high FAR, and two for economic potential: low BH and high BH, with mean values being 176.1, 134, 121.5 kWh/m2/y, and 0.5, 0.53 CHY/kWh, respectively. A demonstration conducted outside Nanjing showed that our findings can be applied to the broader Yangtze River Delta region with a maximum error of less than 15 %. This study provides insights to inform rural PV policy-making and system planning, which are essential for China's low-carbon energy transition.

Keywords: Rural morphology; PV potential; XGBoost algorithm; SHAP value; Impact analysis (search for similar items in EconPapers)
Date: 2025
References: View references in EconPapers View complete reference list from CitEc
Citations:

Downloads: (external link)
http://www.sciencedirect.com/science/article/pii/S0306261924024759
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:appene:v:380:y:2025:i:c:s0306261924024759

Ordering information: This journal article can be ordered from
http://www.elsevier.com/wps/find/journaldescription.cws_home/405891/bibliographic
http://www.elsevier. ... 405891/bibliographic

DOI: 10.1016/j.apenergy.2024.125091

Access Statistics for this article

Applied Energy is currently edited by J. Yan

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