基于支持向量机的地震体波震相自动识别及到时自动拾取

面对海量地震资料,自动准确地拾取震相并确定其到时的需求非常迫切.基于支持向量机技术,本文提出了使用两个分类器SSD和SPS自动识别地震体波震相并自动拾取其到时的方法.相比于传统的自动拾取方法,本文方法能够更准确地识别震相并区分P波和S波.进一步地,我们提出了利用台阵资料辅助识别震相的方案,有效地提高了地震震相拾取的准确率.

Automatic detection of seismic body-wave phases and determination of their arrival times based on support vector machine

Facing massive seismic data, it is urgent to automatically detect earthquakes and determine their arrival times accurately. Based on the support vector machine technology, we developed a method by using two classifiers SSD and SPS to automatically identify seismic body-wave phases and automatically determine their arrival times. Compared with the traditional automatic phase-picking methods, our method can more accurately identify both the seismic phases from noises, and the S phases from P phases. Moreover, we employ the array strategy to further effectively improve the accuracy of phase-detection.