Scholarlink

<p>Built cultural heritage is a vital testimony to human technological evolution and civilization transmission, yet its preservation increasingly faces challenges from environmental change and intrinsic material degradation. Traditional restoration approaches are often constrained by delayed responses and high costs. To address this, the study proposes a comprehensive risk assessment framework that integrates deep learning-based meteorological forecasting with dynamic vulnerability and exposure analysis. First, high-precision time-series predictions of meteorological factors are achieved using data science and deep learning techniques. Based on these predictions, a multidimensional assessment method is constructed to evaluate environmental variability, heritage vulnerability, and exposure. Dynamic risk maps are then generated to visually present evolving risk trends, highlighting temporal sensitivity and spatial heterogeneity. Application in the Yangtze River Delta demonstrates strong performance. The framework provides spatiotemporal coupling analysis, offering a scientific basis for differentiated protection strategies, optimized allocation of restoration resources, advancing heritage conservation toward data-driven and precise management.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

A deep learning-based meteorological factors forecast and dynamic risk assessment approach for built cultural heritage

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2025年11月5日

查看全文
  • Shuyi Li,
  • Jiaxing Yan,
  • Qian Wang,
  • Yuanhang Zhou,
  • Ying Cao,
  • Wei Xu,
  • Ke Du

摘要

Built cultural heritage is a vital testimony to human technological evolution and civilization transmission, yet its preservation increasingly faces challenges from environmental change and intrinsic material degradation. Traditional restoration approaches are often constrained by delayed responses and high costs. To address this, the study proposes a comprehensive risk assessment framework that integrates deep learning-based meteorological forecasting with dynamic vulnerability and exposure analysis. First, high-precision time-series predictions of meteorological factors are achieved using data science and deep learning techniques. Based on these predictions, a multidimensional assessment method is constructed to evaluate environmental variability, heritage vulnerability, and exposure. Dynamic risk maps are then generated to visually present evolving risk trends, highlighting temporal sensitivity and spatial heterogeneity. Application in the Yangtze River Delta demonstrates strong performance. The framework provides spatiotemporal coupling analysis, offering a scientific basis for differentiated protection strategies, optimized allocation of restoration resources, advancing heritage conservation toward data-driven and precise management.