基于支持向量机的非天然地震识别方法在广东及周边区域的应用

Application of Non-Natural Earthquakes Identification Method Based on Support Vector Machine in Guangdong and Surrounding Areas

ES评分 0

DOI 10.14075/j.jgg.2024.08.399
刊名
Journal of Geodesy and Geodynamics
年,卷(期) 2025, 45(8)
作者
作者单位

1 广东省地震局,广州市先烈中路81号,510070;
2 中国地震台网中心,北京市三里河南横街5号,100045

摘要
选取广东及周边区域天然和非天然地震事件作为研究对象,提取P波和S波在频率域的典型物理特征作为特征值,并采用特征组合最优解算法将最佳特征参数输入支持向量机模型用于样本训练,实现对广东及周边区域非天然地震的识别。测试结果表明,地震事件整体识别准确率为99.3%。另外,开发基于支持向量机的非天然地震识别模块,并将其对接广东台网日常编目系统,实现了地震事件类型准实时自动识别。应用结果显示,人工编目的130个地震事件识别准确率为96.9%。
Abstract
In this study, natural and non-natural seismic events in Guangdong and surrounding areas were selected as the research objects, and the typical physical characteristics of P waves and S waves in frequency domain were extracted from seismic waveform data. We used the feature combination optimal solution algorithm to input the best feature parameters into the support vector machine model for sample training, and identified the non-natural earthquakes in Guangdong and surrounding areas. The test results showed that the accuracy of earthquake classification is 99.3%. In addition, the non-natural seismic identification module based on support vector machine was developed and connected to the daily cataloging system of Guangdong seismic network, which realized the quasi-real-time automatic identification of seismic event types. The application results showed that the recognition accuracy of 130 manually cataloged seismic events was 96.9%.
关键词
支持向量机; 非天然地震; 广东及周边区域; 物理特征; 机器学习
KeyWord
support vector machine; non-natural earthquakes; Guangdong and surrounding areas; physical characteristics; machine learning
基金项目
页码 875-880
  • 参考文献
  • 相关文献
  • 引用本文

姜喜姣, 林庆西, 梁皓, 田平, 蒋策, 梁明, 朱腾. 基于支持向量机的非天然地震识别方法在广东及周边区域的应用 [J]. 大地测量与地球动力学. 2025; 45; (8). 875 - 880.

  • 文献评论

相关学者

相关机构