Scholarlink

<p>Rock extraction via blasting poses numerous risks, including ground vibration, air overpressure, and fly rock, among others. Predicting ground vibration accurately is vital for mitigating environmental impacts. Various empirical and statistical methods have been employed, along with machine learning (ML) and soft computing (SC) techniques, to enhance prediction accuracy. This study focuses on the prediction of Peak Particle Velocity (PPV) resulting from blasting activities, particularly in a lead–zinc mine. Using a vulnerability index (VI) based on RES methodology, risk levels associated with PPV are assessed and predictive models are developed. The developed models are evaluated based on performance metrics, highlighting the superior predictive capabilities of the Z-based RES model. In addition, sensitivity analysis reveals the influential parameters affecting PPV. The prediction and sensitivity analysis results indicate a medium–high risk level with an overall PPV of 55.64&#xa0;mm/s under the specific blasting conditions of the studied lead–zinc mine; caution is advised when generalizing these values to other sites or blasting designs. The ZRES model demonstrates superior performance compared to the RES model, with an <i>R</i><sup>2</sup> value of 0.941, RMSE of 1.733, VAF of 94.123%, MAPE of 22.640, and <i>B</i><sub>f</sub> of 1.412. Sensitivity analysis identifies <i>S</i>/<i>B</i> ratio as the most influential parameter and <i>H</i>/<i>B</i> ratio as the least influential.</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

Advancing Peak Particle Velocity Prediction in Rock Blasting: A Z-Based Rock Engineering System

  • 期刊论文

  • 发表日期：

    2025年9月1日

查看全文
  • Qiang Wang,
  • Jun Wang,
  • Pengfei Yue,
  • Shihua Zhang,
  • Yijun Lu,
  • Xinzhe Lv,
  • Yongsen Zhou,
  • Nikolai Ivanovich Vatin,
  • Jiandong Huang

摘要

Rock extraction via blasting poses numerous risks, including ground vibration, air overpressure, and fly rock, among others. Predicting ground vibration accurately is vital for mitigating environmental impacts. Various empirical and statistical methods have been employed, along with machine learning (ML) and soft computing (SC) techniques, to enhance prediction accuracy. This study focuses on the prediction of Peak Particle Velocity (PPV) resulting from blasting activities, particularly in a lead–zinc mine. Using a vulnerability index (VI) based on RES methodology, risk levels associated with PPV are assessed and predictive models are developed. The developed models are evaluated based on performance metrics, highlighting the superior predictive capabilities of the Z-based RES model. In addition, sensitivity analysis reveals the influential parameters affecting PPV. The prediction and sensitivity analysis results indicate a medium–high risk level with an overall PPV of 55.64 mm/s under the specific blasting conditions of the studied lead–zinc mine; caution is advised when generalizing these values to other sites or blasting designs. The ZRES model demonstrates superior performance compared to the RES model, with an R2 value of 0.941, RMSE of 1.733, VAF of 94.123%, MAPE of 22.640, and Bf of 1.412. Sensitivity analysis identifies S/B ratio as the most influential parameter and H/B ratio as the least influential.