Low-carbon retrofit of rural dwellings in the dabie mountain region of China based on life-cycle assessment

[Display omitted] •LCA quantifies carbon reduction potential in rural dwelling retrofitting.•A MOO framework optimizes for rural  residential retrofit solutions.•Two low-carbon retrofitting options for Dabie Mountain rural dwellings. This study proposes a multi-objective optimization (MOO) framework...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Energy and buildings Jg. 344; S. 115991
Hauptverfasser: Wang, Bo, Xi, Hui, Hou, Wanjun, Li, Yueyao
Format: Journal Article
Sprache:Englisch
Veröffentlicht: Elsevier B.V 01.10.2025
Schlagworte:
Life-cycle carbon emissions
Low-carbon retrofit
Machine learning
Multi-objective optimization
Rural dwellings
Multi-objective optimization
Life-cycle carbon emissions
Low-carbon retrofit
Machine learning
Rural dwellings
ISSN:0378-7788
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:[Display omitted] •LCA quantifies carbon reduction potential in rural dwelling retrofitting.•A MOO framework optimizes for rural  residential retrofit solutions.•Two low-carbon retrofitting options for Dabie Mountain rural dwellings. This study proposes a multi-objective optimization (MOO) framework to mitigate the high carbon emissions associated with traditional energy dependence and poor building envelopes in rural mountainous dwellings within China’s hot-summer and cold-winter (HSCW) climate zone, using the Dabie Mountain region as a case study. The framework integrates active and passive low-carbon technologies to simultaneously: 1) reduce life-cycle carbon emissions (LCCE) based on life-cycle assessment (LCA), 2) lower building energy use intensity (EUI), 3) improve thermal comfort percentage (TCP), and 4) enhance retrofit net present value (NPV). The optimization process employs Latin hypercube sampling (LHS) to generate combinations of 17 retrofit variables. Building performance simulation is then used to create a dataset for training high-accuracy machine learning (ML) surrogate models, ensuring computational efficiency and predictive reliability. These ML models are subsequently integrated with advanced multi-objective optimization algorithms (MOOAs) to address the high-dimensional MOO problem and obtain Pareto-optimal solutions. Given the reliance on traditional energy sources in the region, two retrofit options are proposed using multi-criteria decision-making (MCDM) methods. Option 1: A passive retrofit approach, maintaining reliance on traditional energy sources. This option achieves a 5.1 % reduction in LCCE, a 17.6 % decrease in EUI, and a 35.6 % improvement in TCP, though with an NPV of − 22,730 CNY. Option 2: A renewable energy alternative, reducing LCCE by 36 %, cutting EUI by 24 %, improving TCP by 81.6 % and yielding a positive NPV of 48,070 CNY. This phased strategy provides region-specific solutions for decarbonizing rural housing while addressing the area’s socioeconomic constraints.
ISSN:0378-7788
DOI:10.1016/j.enbuild.2025.115991