基于小波包分形的矿山微震波形特征分析  

The wavelet fractal characteristic of micro-seismic waveinmining

作　　者：于正兴[1] 何学秋[2] 朱权洁[3] 吕垒[4] 

机构地区：[1]中国安全生产科学研究院,北京100012 [2]华北科技学院,北京101601 [3]北京科技大学土木与环境工程学院,北京100083 [4]中钢集团武汉安全环保研究院,湖北武汉430081

出　　处：《中国安全生产科学技术》2014年第6期27-32,共6页Journal of Safety Science and Technology

基　　金：国家"十二五"科技支撑计划项目(2012BAK09B01);国家科技重大专项项目(2011ZX05040-001)

摘　　要：矿山微震波形具有非平稳、非高斯、突发瞬态等特点,使用常规方法很难寻求其特征。通过确立微震波形的分形特性,编制了相应的MATLAB程序求解计算,对矿山微震波形的小波分形特征进行了研究。建立了矿山微震波形的分形计算算法及合理标度;利用小波包方法将微震波形分解为多个频带,选择了（0～64）Hz、（64～96）Hz、（96～128）Hz和（128～256）Hz四个主频带进行研究;建立了四个频带对应的logNk与logk关系拟合曲线,并提取了相应的特征值。结果表明：微震波形具有良好的分形特性,其盒维数的大小在一定程度上与频率的高低相对应;波形主频带S4,2与整个波形的盒维数近似相等。微震波形主频带上的分形特性可以作为该波形的特征,用于不同类型微震波形的分类识别。The microseismic signal inmine is a kind of discrete linear signal,which is difficult to find its characteristics using conventional methods. In this paper,based on previous studies,the fractal characteristics of the microseismic signals inmine was verified and researched. The microseismic signal was analyzed firstly by the establishing of scale-free interval and the calculating of the box-dimension,and then decomposed by using the wavelet packet theory according to its power spectrum characteristics; Four main frequency bands were selected for further analysis： extracting the reconstructed signal of the bands to calculate the box-counting dimension. Those bands were S3,0（ 0-64Hz）,S4,2（ 64-96Hz）,and this is the frequency band where the peak of the power spectrum located）,S4,3（ 96-128Hz）,and S3,2（ 128-256Hz）; The fitting curve of reconstruction signal corresponding relation between logNkand logk was been drawn,and the corresponding box dimension was extracted. The results showed that the microseismic signal has a good fractal characteristics,the value of the box-counting dimension to some extent reflects the value of the frequency; and the box-counting dimension of the main frequency band,S4,2,and the wholesignal are approximately equal to each other. In this case,the box-counting dimension can be used to represent the characteristic of the signal for quantitative recognition of different microseismic signals.

关 键 词：微地震波形 小波分形 分形盒维数 MATLAB 波形特征 

分 类 号：X936[环境科学与工程—安全科学]