基于太赫兹时域光谱技术的煤岩界面识别  

Coal-rock interface recognition based on terahertz time-domain spectroscopy

作　　者：王昕[1,2,3,4] 胡克想[1,2,3,4] 俞啸[1,2,3,4] 丁恩杰[1,2,3,4] 

机构地区：[1]中国矿业大学物联网(感知矿山)研究中心,江苏徐州221008 [2]中国矿业大学信息与电气工程学院,江苏徐州221008 [3]矿山互联网应用技术国家地方联合工程实验室,江苏徐州221008 [4]江苏省感知矿山物联网工程实验室,江苏徐州221008

出　　处：《工矿自动化》2017年第1期29-34,共6页Journal Of Mine Automation

基　　金：国家重点基础研究发展计划(973计划)资助项目(2014CB046300)

摘　　要：为了研究基于太赫兹时域光谱的煤岩界面识别可行性,首先应用太赫兹时域光谱技术,以气煤和砂岩粉末压片作为实验样本进行了相关实验,得到了太赫兹脉冲穿过不同样本时所产生的时间延迟和衰减幅度。然后采用Hilbert-Huang变换对煤岩样的太赫兹光谱进行了时频域分析,并与由传统快速傅里叶变换方法获得的样品频谱进行了对比,结果表明,在时频域内太赫兹光谱信号具有更高的分辨率,可以更好、更直观地分析太赫兹脉冲在任意时刻的频率、能量成分,研究任意频率在不同时刻对吸收谱线的作用。最后介绍了基于太赫兹时域光谱技术的煤岩界面识别过程,通过实验验证了采用Hilbert谱和边际谱提取出的光谱特征值可以很好地区分煤岩介质。In order to study feasibility of coal-rock interface recognition based on terahertz time-domain spectroscopy,some experiments were taken out by using terahertz time-domain spectroscopy,which used gas coal and sandstone powder tablet as samples.Time delay and attenuation amplitude that caused by terahertz pulse transmission between different samples were obtained.Hilbert-Huang transform was used to make time-frequency analysis of the samples＇ terahertz spectrum,and the analysis results were compared with the ones by traditional fast Fourier transform method.The comparison results show that resolution of terahertz spectrum is higher in time-frequency domain,and frequency and energy components of terahertz pulse can be analyzed better and more intuitively at any time as well as absorption effect of any frequency atdifferent time.Coal-rock interface recognition process based on terahertz time-domain spectroscopy was introduced.The experimental results show that coal-rock medium can be well distinguished by characteristic value extracted by Hilbert spectrum and marginal spectrum.

关 键 词：煤炭开采 煤岩界面 煤岩识别 太赫兹时域光谱 HILBERT-HUANG变换 

分 类 号：TD67[矿业工程—矿山机电]