Scholarlink

<p>In many cities worldwide, residential buildings are designed with a uniform architectural layout, with variations in the number of floors and the spacing between frames. In Djelfa, Algeria, the government constructs residential housing following this standardized approach, with most buildings consisting of four to five floors and a fixed structural configuration. This uniformity simplifies design and construction. However, it also necessitates an efficient method for evaluating seismic performance, especially in seismically active regions. This study proposes an alternative approach to rapidly estimate the seismic response of such buildings. A nonlinear static pushover analysis was carried out, considering variations in frame spacing, material uncertainties, and seismic loading within a probabilistic framework. The resulting database is used to train an artificial neural network (ANN) model, which provides a predictive function for seismic response. Additionally, a correlation between key response parameters is established, leading to a simplified expression for direct seismic response computation. The proposed methodology is applied to four- and five-story buildings in Djelfa, demonstrating promising accuracy and efficiency, making it a valuable tool for seismic assessment and design optimization in similar urban developments.</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

Artificial neural network-based prediction of seismic response for residential buildings with uniform architectural design

  • 期刊论文

  • 发表日期：

    2026年3月14日

查看全文
  • Abdallah Yacine Rahmani,
  • Mohamed Badaoui

摘要

In many cities worldwide, residential buildings are designed with a uniform architectural layout, with variations in the number of floors and the spacing between frames. In Djelfa, Algeria, the government constructs residential housing following this standardized approach, with most buildings consisting of four to five floors and a fixed structural configuration. This uniformity simplifies design and construction. However, it also necessitates an efficient method for evaluating seismic performance, especially in seismically active regions. This study proposes an alternative approach to rapidly estimate the seismic response of such buildings. A nonlinear static pushover analysis was carried out, considering variations in frame spacing, material uncertainties, and seismic loading within a probabilistic framework. The resulting database is used to train an artificial neural network (ANN) model, which provides a predictive function for seismic response. Additionally, a correlation between key response parameters is established, leading to a simplified expression for direct seismic response computation. The proposed methodology is applied to four- and five-story buildings in Djelfa, demonstrating promising accuracy and efficiency, making it a valuable tool for seismic assessment and design optimization in similar urban developments.