Scholarlink

<p>A reliability-based design optimisation framework is proposed to mitigate rib spalling hazards in longwall faces. Coupled coal-mass–support interactions are first examined to isolate the governing parameters. Three-dimensional physical similarity experiments quantify the influence of support geometry on rib failure, yielding linear response models. The stochastic behaviour of rib stability is then captured via response-surface-guided Monte-Carlo simulation, delineating the dominant probability ranges of spalling areas. A hybrid predictor, integrating nonlinear principal-component analysis with a genetic-algorithm-optimised back-propagation neural network, attains prediction errors of 0–5% in 90.1% of cases and 5–15% in 9.9% of cases. The methodology offers both a theoretical basis and a practical tool for safe extraction of ultra-thick hard coal seams.</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

Optimized reliability prediction model for rib spalling in ultra-thick hard coal seams with high cutting height

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2025年10月6日

查看全文
  • Feitian Zhang,
  • Bo Xue,
  • Yan Shang,
  • Yongfeng Jia,
  • Jiafei Zhang

摘要

A reliability-based design optimisation framework is proposed to mitigate rib spalling hazards in longwall faces. Coupled coal-mass–support interactions are first examined to isolate the governing parameters. Three-dimensional physical similarity experiments quantify the influence of support geometry on rib failure, yielding linear response models. The stochastic behaviour of rib stability is then captured via response-surface-guided Monte-Carlo simulation, delineating the dominant probability ranges of spalling areas. A hybrid predictor, integrating nonlinear principal-component analysis with a genetic-algorithm-optimised back-propagation neural network, attains prediction errors of 0–5% in 90.1% of cases and 5–15% in 9.9% of cases. The methodology offers both a theoretical basis and a practical tool for safe extraction of ultra-thick hard coal seams.