北山花岗岩区微震事件的Fisher判别模型

1.中国矿业大学(北京)力学与建筑工程学院,北京 100083; 2.核工业北京地质研究院 环境工程研究所,北京 100029;3.四川大学 水利水电学院,四川 成都 610065

北山花岗岩区; 微震事件; 震源参数; Fisher判别模型

Fisher discriminant analysis model for microseismic events of Beishan granite area
HE Shu-sheng1,ZHOU Hong-wei1,WANG Chao-sheng1,WANG Zi-hui1,CHEN Liang2,LIU Jian-feng3

(1.School of Mechanics & Civil Engineering,China University of Mining and Technology,Beijing 100083,China;2.Division of Environment Engineering,Beijing Research Institute of Uranium Geology,Beijing 100029,China; 3.School of Water Resources and Hydropower,Sichuan University,Chengdu 610065,China)

Beishan granite area; microseismic events; seismic parameters; FDA model

DOI: 10.13800/j.cnki.xakjdxxb.2017.0410

备注

应用微震监测系统,对北山坑探设施开挖过程进行实时监测。为研究北山花岗岩区微震事件的判别方法,采用Fisher判别分析法,选取地震力矩、微震能量、微震体变势、静应力降和视应力等5个震源参数作为判别因子,综合考虑各震源参数的差异性,建立爆破和微震事件Fisher判别模型,并对该模型进行检验。结果 表明:总体判别正确率为86.3%,其中爆破组监测数据正确判别率达100%,微震组监测数据的正确判别率为76.4%,判别结果与事件实际基本相符。该模型简便易行,正确率较高,能够对北山花岗岩区爆破和微震事件进行有效地识别。

Based on the Beishan Exploration Tunnel(BET),the excavation progress was real-time monitored by using microseismic monitoring system.In order to research the discriminant method of microseismic events in Beishan granite area,the Fisher discriminant analysis(FDA)method was conducted.By selecting 5 hypocenter parameters as discriminant factors,such as seismic moment,energy,potency,statistical stress drop and apparent stress,and considering their differences comprehensively,the FDA model was established and tested.It shown that,the accuracy of total samples is 86.3%,in which the accuracy of blasting samples is 100%,and the accuracy of microseismic samples is 76.4%.So,owing to the simple and high correct rate of this model,the blasting and microseismic events in Beishan granite area can be indentified effectively.