基于XGBoost算法的工程场地实测和人工地震波时频特征分析与判别

曹子龙; 黄杜若

doi:10.16511/j.cnki.qhdxxb.2022.25.036

PDF(12273 KB)

清华大学学报（自然科学版） ›› 2022, Vol. 62 ›› Issue (8) : 1330-1340. DOI: 10.16511/j.cnki.qhdxxb.2022.25.036

智能预测反馈

基于XGBoost算法的工程场地实测和人工地震波时频特征分析与判别

曹子龙, 黄杜若

作者信息 +

Time-frequency characteristic analyses of measured and artificial seismic waves using the XGBoost algorithm

CAO Zilong, HUANG Duruo

Author information +

文章历史 +

摘要

近年来，人工智能方法快速发展，在许多工程问题中，逐渐引入具有良好预测能力和泛化能力的机器学习算法。该文考虑采用XGBoost人工智能方法，对工程场地实测和人工模拟地震波的时频规律特征进行深入探索，旨在解决地震波研究目前存在的资料缺乏与认识匮乏两大问题。采用的XGBoost算法优势在于人工智能方法的高性能计算能够完成传统计算方法难以实现的对大量数据的分析，从而发掘、重现地震波的时域和频域特征。在对实测和SIMQKE人工地震波的判别研究中，发现本算法对于二者的判别准确率能达到91%，进一步研究发现人工地震波与实测波差别主要体现在时频域特征的相关性上。该文有助于进一步认识地震波的时频特征，同时也对人工地震波模拟方法的发展具有重要意义。

Abstract

The rapid development of artificial intelligence and machine learning techniques that can predict and generalize data sets are gradually being introduced into many engineering projects. The XGBoost artificial intelligence method was used here to analyze the time-frequency characteristics of measured and artificially simulated seismic waves to resolve the two problems of the lack of data and the lack of understanding seismic waves. The advantage of the XGBoost method is that high-performance artificial intelligence methods can analyze large amounts of data that would be difficult by traditional calculational methods. The method can then be used to analyze the time-frequency domain characteristics of seismic waves. The algorithm accurately discriminated 91% of the measured and artificial SIMQKE seismic waves. Further research showed that the difference between the artificial seismic waves and the measured waves was mainly reflected in the correlation of the time-frequency domain characteristics. This study reveals the time-frequency characteristics of seismic waves that will facilitate the development of artificial seismic wave simulation methods.

导出引用

曹子龙, 黄杜若. 基于XGBoost算法的工程场地实测和人工地震波时频特征分析与判别[J]. 清华大学学报（自然科学版）. 2022, 62(8): 1330-1340 https://doi.org/10.16511/j.cnki.qhdxxb.2022.25.036

CAO Zilong, HUANG Duruo. Time-frequency characteristic analyses of measured and artificial seismic waves using the XGBoost algorithm[J]. Journal of Tsinghua University(Science and Technology). 2022, 62(8): 1330-1340 https://doi.org/10.16511/j.cnki.qhdxxb.2022.25.036

参考文献

[1] YOUNGS R R, CHIOU S J, SILVA W J, et al. Strong ground motion attenuation relationships for subduction zone earthquakes[J]. Seismological Research Letters, 1997, 68(1):58-73.
[2] YEH C H, WEN Y K. Modeling of nonstationary ground motion and analysis of inelastic structural response[J]. Structural Safety, 1990, 8(1-4):281-298.
[3] CONTE J P, PENG B F. Fully nonstationary analytical earthquake ground-motion model[J]. Journal of Engineering Mechanics, 1997, 123(1):15-24.
[4] URSINO A, LANGER H, SCARFI L, et al. Discrimination of quarry blasts from tectonic microearthquakes in the Hyblean Plateau (Southeastern Sicily)[J]. Annals of Geophysics, 2001, 44(4):703-722.
[5] WISZNIOWSKI J, PLESIEWICZ B M, TROJANOWSKI J. Application of real time recurrent neural network for detection of small natural earthquakes in Poland[J]. Acta Geophysica, 2014, 62(3):469-485.
[6] PEROL T, GHARBI M, DENOLLE M. Convolutional neural network for earthquake detection and location[J]. Science Advances, 2018, 4(2):e1700578.
[7] ANCHETA T D, DARRAGH R B, STEWART J P, et al. NGA-West2 database[J]. Earthquake Spectra, 2014, 30(3):989-1005.
[8] YAMAMOTO Y, BAKER J W. Stochastic model for earthquake ground motion using wavelet packets[J]. Bulletin of the Seismological Society of America, 2013, 103(6):3044-3056.
[9] HUANG D R, WANG G. Stochastic simulation of regionalized ground motions using wavelet packets and cokriging analysis[J]. Earthquake Engineering & Structural Dynamics, 2015, 44(5):775-794.
[10] HUANG D R, WANG G. Region-specific spatial cross-correlation model for stochastic simulation of regionalized ground-motion time histories[J]. Bulletin of the Seismological Society of America, 2015, 105(1):272-284.
[11] 黄杜若, 王刚, 盛志刚. 基于小波包和空间相关性分析的人工地震波仿真技术[J]. 华南地震, 2014, 34(3):82-90. HUANG D R, WANG G, SHENG Z G. Simulation technology of artificial seismic waves based on wavelet packets and spatial correlation analysis[J]. South China Journal of Seismology, 2014, 34(3):82-90. (in Chinese)
[12] WANG G, YOUNGS R, POWER M, et al. Design ground motion library:An interactive tool for selecting earthquake ground motions[J]. Earthquake Spectra, 2015, 31(2):617-635.
[13] JENNINGS P C, HOUSNER G W, TSAI N C. Simulated earthquake motions[R]. USA:California Institute of Technology, EERL, 1968.
[14] WANG M X, HUANG D R, WANG G, et al. SS-XGBoost:A machine learning framework for predicting newmark sliding displacements of slopes[J]. Journal of Geotechnical and Geoenvironmental Engineering, 2020, 146(9):04020074.
[15] CHEN T Q, GUESTRIN C. XGBoost:A scalable tree boosting system[C]//Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. San Francisco, CA, USA:ACM, 2016:785-794.

基金

国家自然科学基金资助项目（52179134）

PDF(12273 KB)

Accesses

Citation

Detail

段落导航

收稿日期	出版日期
2021-10-28	2022-08-15
录用日期	发布日期
2022-03-31	2022-03-31

选择文件类型/文献管理软件名称

选择包含的内容

摘要

Abstract

关键词

Key words

引用本文

{{custom_sec.title}}

{{custom_sec.title}}

参考文献

基金

访问统计

模态框（Modal）标题

选择文件类型/文献管理软件名称

选择包含的内容

摘要

Abstract

关键词

Key words

引用本文

{{custom_sec.title}}

{{custom_sec.title}}

参考文献

基金

访问统计