石灰岩声发射频谱特性演化及破裂阶段识别
DOI:
作者:
作者单位:

作者简介:

通讯作者:

中图分类号:

TU45

基金项目:

国家自然科学基金(51464036),内蒙古自治区自然科学基金(2018MS05037)


Evolution of Acoustic Emission Spectrum Characteristics of Limestone and Identification of Fracture Stage
Author:
Affiliation:

Fund Project:

  • 摘要
  • |
  • 图/表
  • |
  • 访问统计
  • |
  • 参考文献
  • |
  • 相似文献
  • |
  • 引证文献
  • |
  • 资源附件
  • |
  • 文章评论
    摘要:

    为获得石灰岩受压破裂阶段声发射识别信息,开展石灰岩单轴压缩声发射试验,获得加载全过程的声发射特征参数及其波形信息,首先结合石灰岩应力与累积能量走势的特征选取 5 个岩石加载破坏过程中的关键点。通过频谱分析与小波包分析,找到石灰岩声发射频谱特征变化规律,获得适合的破裂阶段识别信息;其次利用小波包分解所提取的频带能量占比训练 BP 神经网络,为岩石破裂阶段预警提供双 重保证。研究结果表明:主频值整体呈现先上升后下降的趋势,频谱宽度变化为 200 kHz-300 kHz-200 kHz,频率由单一低频转化为低高频并存后又恢复单一低频;低频高幅为频谱特征的一种主要形态存在;小波包分解将信号频率分为 8 个频段,可将 1、2 低频段之和小于 30%的信号作为特征点,进行岩石破裂预测;以该特征点为界将石灰岩声发射信号划分为两类,训练后的 BP 神经网络能够实现这两类信号的快速准确识别与分类,对于预测岩石破裂阶段有一定价值。

    Abstract:

    To obtain the acoustic emission identification information of limestone compression rupture stage The acoustic emission characteristic parameters and waveform information of the whole process of loading were obtained by carrying out the uniaxial compression acoustic emission experiment of limestone. First of all, combined with characteristics of limestone stress and cumulative energy trend, five key points were selected during the process of rock loading to failure. Through spectrum analysis and wavelet packet analysis, obtaining the change law of acoustic emission spectrum characteristics of limestone and seeking suitable identification information of rupture stage .Secondly, BP neural network was trained by using the frequency band energy ratio extracted from wavelet packet decomposition. It provide a double guarantee for warning of rock failure stage. The results show that: the main frequency values generally show a tendency to rise first and then fall;The spectrum width changes from 200 kHz to 300 kHz and back to 200 kHz. Signal frequency changes from a low frequency to a frequency with low and high ,and then returns to a low frequency. Low frequency and high amplitude are the main forms of spectrum characteristics. Wavelet packet decomposition divides signal frequency into 8 frequency bands, the signal with the sum of 1 and 2 low frequency bands less than 30% can be taken as the characteristic point for predicting rock failure. Limestone acoustic emission signals are divided into two categories by the characteristic point. The trained BP neural network can identify and classify these two kinds of signals quickly and accurately, which is of certain value for predicting rock failure stage.

    参考文献
    相似文献
    引证文献
引用本文

王创业,刘沂琳,常新科. 石灰岩声发射频谱特性演化及破裂阶段识别[J]. 科学技术与工程, 2020, 20(35): 14646-14652.
Wang Chuangye, Liu Yilin, Chang Xinke. Evolution of Acoustic Emission Spectrum Characteristics of Limestone and Identification of Fracture Stage[J]. Science Technology and Engineering,2020,20(35):14646-14652.

复制
分享
文章指标
  • 点击次数:
  • 下载次数:
  • HTML阅读次数:
  • 引用次数:
历史
  • 收稿日期:2020-04-10
  • 最后修改日期:2020-09-02
  • 录用日期:2020-06-21
  • 在线发布日期: 2021-01-06
  • 出版日期:
×
律回春渐,新元肇启|《科学技术与工程》编辑部恭祝新岁!
亟待确认版面费归属稿件,敬请作者关注