中文摘要：

针对煤矿安全生产中的综放工作面煤矸界面探测问题,提出利用煤矸下落冲击钢板的振动特征来探测煤矸界面的方法。煤矸振动信号表现出非平稳特征,采用经验模态分解方法将复杂矿井环境下的煤矸振动信号分解成固有模态分量。选择包含煤矸振动特征的前7个本征模函数（IMF）分量,通过Hilbert变换得到Hilbert谱。分析不同放煤状态下钢板振动信号的Hilbert谱发现,顶煤下落时的Hilbert谱分布较均匀,而煤矸混放时的Hilbert谱呈现不均匀分布。根据信息熵理论,提出了基于Hilbert谱信息熵的煤矸振动特征提取方法。试验结果表明,顶煤下落时的Hilbert谱信息熵要大于煤矸混放时的Hilbert谱信息熵,因此,煤矸振动的Hilbert谱信息熵特征能够准确地反映放煤状态。

英文摘要：

Targeting coal-gangue interface detection on fully mechanized mining face for coal mine safety,a new method to detect coal-gangue interface by utilizing vibration signals of coal and gangue was presented.Because of non-stationary characteristics contained in response signals under complicated environment,EMD method was used to decompose the original vibration signals into intrinsic mode components（IMFs）.Hilbert spectrum was obtained through Hilbert transformation with the first seven IMFs.It was found that the distribution of Hilbert spectrum of top coal caving was more uniform than that of coal-gangue caving.Thus the method of vibrational feature extraction based on information entropy of Hilbert spectrum was proposed.Experimental results show that information entropy of Hilbert spectrum of top coal caving is always greater than that of coal-gangue caving.Thus the vibrational feature for coal and gangue based on information entropy of Hilbert spectrum can reflect the coal caving condition accurately.