Scholarlink

<p>Due to uncertainties in seismic pipeline damage and post-earthquake recovery processes, probabilistic characteristics such as mean value, standard deviation, probability density function, and cumulative distribution function provide valuable information. In this study, a simulation-based framework to evaluate these probabilistic characteristics in water distribution systems (WDSs) during post-earthquake recovery is developed. The framework first calculates pipeline failure probabilities using seismic fragility models and then generates damage samples through quasi-Monte Carlo simulations with Sobol’s sequence for faster convergence. System performance is assessed using a hydraulic model, and recovery simulations produce time-varying performance curves, where the dynamic importance of unrepaired damage determines repair sequences. Finally, the probabilistic characteristics of seismic performance indicators, resilience index, resilience loss, and recovery time are evaluated. The framework is applied in two benchmark WDSs with different layouts to investigate the probabilistic characteristics of their seismic performance and resilience. Application results show that the cumulative distribution function reveals the variations in resilience indicators for different exceedance probabilities, and there are dramatic differences among the recovery times corresponding to the system performance recovery targets of 80%, 90%, and 100%.</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

Probabilistic characteristic analysis of seismic performance of water distribution system based on quasi-Monte Carlo simulation

  • 期刊论文

  • 发表日期：

    2025年4月1日

查看全文
  • Benwei Hou,
  • Minghao Yuan,
  • Kegong Diao,
  • Xitao Ma,
  • Baojin Zhou,
  • Xiuli Du

摘要

Due to uncertainties in seismic pipeline damage and post-earthquake recovery processes, probabilistic characteristics such as mean value, standard deviation, probability density function, and cumulative distribution function provide valuable information. In this study, a simulation-based framework to evaluate these probabilistic characteristics in water distribution systems (WDSs) during post-earthquake recovery is developed. The framework first calculates pipeline failure probabilities using seismic fragility models and then generates damage samples through quasi-Monte Carlo simulations with Sobol’s sequence for faster convergence. System performance is assessed using a hydraulic model, and recovery simulations produce time-varying performance curves, where the dynamic importance of unrepaired damage determines repair sequences. Finally, the probabilistic characteristics of seismic performance indicators, resilience index, resilience loss, and recovery time are evaluated. The framework is applied in two benchmark WDSs with different layouts to investigate the probabilistic characteristics of their seismic performance and resilience. Application results show that the cumulative distribution function reveals the variations in resilience indicators for different exceedance probabilities, and there are dramatic differences among the recovery times corresponding to the system performance recovery targets of 80%, 90%, and 100%.