Intelligent Detection and Energy-Driven Repair of Building Envelope Defects for Improved Thermal and Energy Performance
Open Access
Abstract
This study addresses the challenge of rapid identification and assessment of localized damage to building envelopes under resource-constrained conditions—specifically, the absence of specialized inspection equipment—with a particular focus on the detrimental effects of such damage on thermal performance and energy efficiency. An efficient detection methodology tailored to small-scale maintenance scenarios is proposed, leveraging the YOLOv11 object detection architecture to develop an intelligent system capable of recognizing common envelope defects in contemporary residential buildings, including cracks, spalling, and sealant failure. The system prioritizes the detection of anomalies that may induce thermal bridging, reduced airtightness, or insulation degradation. Defects are classified according to severity and their potential impact on thermal behavior, enabling a graded, integrated repair strategy that holistically balances structural safety, thermal restoration, and façade aesthetics. By explicitly incorporating energy performance recovery as a core objective, the proposed approach not only enhances the automation of spatial data processing but also actively supports the green operation and low-carbon retrofitting of existing urban building stock. Characterized by low cost, high efficiency, and ease of deployment, this method offers a practical and scalable technical pathway for the intelligent diagnosis of thermal anomalies and the enhancement of building energy performance. It aligns with the principles of high-quality architectural development and sustainable building governance, while concretely advancing operational energy reduction in the built environment and contributing meaningfully to energy conservation goals.Keywords
Funding Information
- Postgraduate Research & Practice Innovation Program of Suzhou University of Science and Technology (SJCX24_1858)
- Jiangsu Provincial Engineering Research Center for Building Carbon Neutrality Technology, Su-zhou University of Science and Technology and the National Natural Science Foundation of China (No.51908393)
- National Natural Science Foundation of China (No.52508006)
This publication has 35 references indexed in Scilit:
- Extreme small object detection with transformer-based local and global context attentionInfrared Physics & Technology, 2025
- Hollowing defect detection and risk assessment of facade tile shedding in high-rise residential buildings using the UAV–mounted infrared thermography methodJournal of Building Engineering, 2025
- An Improved YOLOv11 architecture with multi-scale attention and spatial fusion for fine-grained residual detectionResults in Engineering, 2025
- Insights into the influence of reinforcing iron bars corrosion on historical buildings: Towards optimizing limestone performanceJournal of Building Engineering, 2025
- Deep learning-based automated tile defect detection system for Portuguese cultural heritage buildingsJournal of Cultural Heritage, 2024
- Thermal performance of flat roof insulation materials: A review of temperature, moisture and aging effectsJournal of Building Engineering, 2023
- A rapid review of the impact of increasing airtightness on indoor air qualityJournal of Building Engineering, 2022
- Infrared thermography in the built environment: A multi-scale reviewRenewable and Sustainable Energy Reviews, 2022
- Mid-term and long-term changes in building airtightness: A field study on low-energy housesEnergy and Buildings, 2021
- A hybrid ANN-based imperial competitive algorithm methodology for structural damage identification of slab-on-girder bridge using data miningApplied Soft Computing, 2020