【作者】 张艳丽； 张守祥； 王永强；

【Author】 Zhang Yanli,Zhang Shouxiang,Wang Yongqiang (School of Information & Electronics Engineering,Shandong Institute of Business and Technology,Yantai,Shandong,264005)

【机构】 山东工商学院信息与电子工程学院；

【摘要】 本文简要介绍了独立分量分析的基本思想及基于负熵极大的FastICA算法,并对放顶煤过程中产生的声波信号进行ICA分析和频谱分析,结果表明,FastICA算法可以较好地分离煤和矸石的混合声音信号,分离出的两独立信号的频谱能量分别集中在2000～3000Hz之间和1000Hz左右。根据独立信号频谱的不同,能够初步判断下落的是煤还是矸石,实现煤岩界面的识别。更多还原

【Abstract】 The basic theory of independent component analysis and a FastICA algorithm based on maximum negentropy was briefly introduced.Then the FastICA and Fourier frequency spectrum analysis were used for the processing of the sound signal brought in the top coal caving.The results show that FastICA algorithm has better performance and efficiency in mixed sound signal separation, the spectrum energy of independent components separated respectively focus on between 2000～3000Hz and about 1000Hz.So the Coal and rock can be judged based on the spectrum difference of independent components and the identification of coal and rock interface can be realized.更多还原