地震事件分类识别软件  

Seismic Event Recognition Software

作　　者：王婷婷[1] 边银菊[1] 任梦依 杨千里[1] 侯晓琳 WANG Ting-ting;BIAN Yin-ju;REN Meng-yi;YANG Qian-li;HOU Xiao-lin(Institute of Geophysics,China Earthquake Administration,Beijing 100081,China)

机构地区：[1]中国地震局地球物理研究所,北京100081

出　　处：《地震》2024年第2期104-119,共16页Earthquake

基　　金：北京市自然科学基金项目(8234066)。

摘　　要：非天然地震事件分类是地震监测业务部门的日常工作之一。本研究主要针对地震、爆炸和矿震的分类问题,在地震波数据处理、特征提取和人工智能综合分类的研究基础上,基于Qt开发框架,结合Python、 Matlab等多种编程语言,开发了一个具有良好的可移植性和可扩展性、具有自主知识产权的地震分类识别软件。该软件可以部署在不同操作系统上,由七个模块组成:地震数据导入模块、数据处理模块、特征提取模块、综合分类模块、特征分析模块、当量估算模块和结果分析模块。软件集成了多种时频特征提取技术和人工智能分类方法,形成了较为完整的地震类型判定流程。软件内置的地震事件分类模型准确率高于90%,适用范围较广,已推广应用于多个地震监测部门,并取得了较好的应用成果,提高了对非天然地震的快速分析能力。Classification of non-natural seismic events is one of the daily tasks of the seismic monitoring business. This research is mainly aimed at the classification of earthquakes, explosions and mining-induced earthquakes. On the basis of the research results of seismic wave data processing, feature extraction and artificial intelligence comprehensive classification, a seismic event recognition software(SERS) with good portability, expansibility and independent intellectual property rights is developed based on the Qt development framework and combined with Python, Matlab and other programming languages. The software can be deployed on different operating systems and consists of seven modules: seismic data import module, data processing module, feature extraction module, comprehensive classification module, feature analysis module, yield estimation module, and result analysis module. The software integrates various time-frequency feature extraction techniques and artificial intelligence classification methods, to form a comprehensive process for classifying the seismic events. The built-in classification models in the software have an accuracy rate exceeding 90% and a wide range of applications. It has been applied in a number of earthquake monitoring departments, achieved favorable outcomes and enhanced the capability for rapid analysis of non-natural earthquakes.

关 键 词：非天然地震事件分类 Qt开发框架 特征提取 人工智能方法 

分 类 号：P315.7[天文地球—地震学]