• 欢迎访问ERC!
引用本文:
【打印本页】   【HTML】   【下载PDF全文】   查看/发表评论  【EndNote】   【RefMan】   【BibTex】
过刊浏览    高级检索
本文已被:浏览 210次   下载 0  
分享到: 微信 更多
基于PhaseNet的地震波和相位检测系统
赵明1, 马嘉卉2, 常昊3, 陈石4
1.中国地震局地球物理研究所;2.中国科学院国家空间科学中心;3.中国科学院微电子研究所;4.Institute of Geophysics, China Earthquake Administration
摘要:
我们开发了基于PhaseNet的地震波和相位自动检测系统,这是一种高效且高度通用的基于深度学习神经网络的地震P波和S波相位选择器,并结合了数据可视化,滑动窗口,ssh协议数据传输和其他辅助模块。该系统使用非常方便:用户只需添加三分量地震数据作为输入文件,在软件界面中定义用户参数,然后系统将自动处理并返回P波和S波的结果。系统可以处理sac,mseed以及numpy格式的连续或截断波形数据。该系统有望解决人工处理大量地震数据的效率低下和主观性较强问题,为区域网络监测人员和研究人员在研究地球内部结构方面提供了便利。
关键词:  深度神经网络,地震波,检测,.Net Framework 4.0,Docker
DOI:
分类号:
基金项目:国家自然科学基金项目(面上项目,重点项目,重大项目),中国地震局地球物理研究所基本科研业务专项
PhaseNet-based Seismic Wave and Phase Detection System
Zhao Ming,Ma Jiahui,Chang Hao,chenshi
1.Institute of Geophysics, China Earthquake Administration;2.National Space Science Center, Chinese Academy of Sciences;3.Institute of Microelectronics, Chinese Academy of Science
Abstract:
We developed an automatic seismic wave and phase detection system based on PhaseNet, an efficient and highly generalized deep learning neural network P and S phase picker, and combined with data visualization, docker image, ssh protocol data transmission and other auxiliary modules. The system is very convenient to use: the user only needs to prepare three-component seismic data as input, customize some user parameters in the software interface, then the system will automatically process and return P, S picks, regardless of whether the waveform is continuous or truncated, and whether the datatype is sac, mseed or numpy array. The system is expected to solve the inefficiency and subjectivity in the manual processing of large amounts of seismic data, provide convenience to regional network monitoring staff and researchers in the study of the internal structure of the earth.
Key words:  Deep neural network,Seismic wave,Detection,.Net Framework 4.0,Docker