Scholarlink

<p>The displacement of dams results from interactions between environmental factors and structural responses over time. This study uses neural networks with logarithmically distributed time delays to model the Enguri Dam's surface displacement, leveraging Copernicus meteorological data and GB-SAR measurements. Findings highlight that long-term memory (100-1 000 hours) significantly enhances predictive accuracy. Water levels exhibit medium- and long-term effects, while environmental influences refine short-term predictions, demonstrating the importance of time-aware AI for improving dam safety.</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

Mehrskalige Staumauer-Analyse: KI-basierter Ansatz mit zeitlichem Gedächtnis

  • 期刊论文

  • 发表日期：

    2026年4月27日

查看全文
  • Raphael Dujardin,
  • Sadeeb Simon Ottenburger

摘要

The displacement of dams results from interactions between environmental factors and structural responses over time. This study uses neural networks with logarithmically distributed time delays to model the Enguri Dam's surface displacement, leveraging Copernicus meteorological data and GB-SAR measurements. Findings highlight that long-term memory (100-1 000 hours) significantly enhances predictive accuracy. Water levels exhibit medium- and long-term effects, while environmental influences refine short-term predictions, demonstrating the importance of time-aware AI for improving dam safety.