中国海域及邻区基于BP神经网络的矩震级转换模型
CSTR:
作者:
中图分类号:

P315

基金项目:

国家重点研发项目(2017YFC1500402)、中国地震局地质研究所基本科研业务专项(IGCEA1823、IGCEA1901)共同资助


Moment Magnitude Conversion Model Based on BP Neural Network for China's Seas and Adjacent Regions
Author:
  • 摘要
  • | |
  • 访问统计
  • |
  • 参考文献 [53]
  • |
  • 相似文献 [20]
  • | | |
  • 文章评论
    摘要:

    海域地震对我国海洋资源开发和沿海地区的经济建设形成严重威胁,开展相关地震活动性研究的重要前提是编译我国海域及邻区的地震目录。我国常用的震级标度为面波震级(MS),而国际上最新的地震活动模型多采用矩震级(MW),因此在应用这些模型时需要拟合面波震级与矩震级之间的转换关系。本文以中国海域及邻区为研究区,收集了1988—2020年中国地震台网的面波震级和全球矩心矩张量(GCMT)项目的矩震级数据,从中提取年份、深度、经度、纬度、面波震级作为影响因子,以实际记录的矩震级值作为标记,训练BP神经网络建立以GCMT的矩震级为目标的震级转换模型。同时,使用最小二乘回归和正交回归建立线性模型作为对比。结果显示,最小二乘回归和正交回归的平均绝对误差和均方根误差比BP神经网络高40%左右。此外,BP神经网络的残差绝对值更小、分布更集中。

    Abstract:

    Earthquakes in sea areas pose a serious threat to the development of China's marine resources and the economic construction of coastal areas. A crucial prerequisite for conducting relevant seismicityresearch is to compile an earthquake catalog for China's seas and adjacent regions. The commonly used magnitude scale in China is surface wave magnitude(MS), however moment magnitude(MW) has beencommonly used in latest seismic activity models andrecommended as a recognized magnitude standard worldwide. Therefore, it is necessary to fit the conversion relationship between surface wave magnitude of China and moment magnitude. Taking China's seas and neighboring regions as the study area, we collected the earthquake records from 1988 to 2020 with surface wave magnitudes from the China Earthquake Networks Center(CENC) and moment magnitudes from the Global Centroid Moment Tensor(GCMT). Based on the dataset, the year, depth, longitude, latitude, and MS of CENC were extracted as impact factors, and the corresponding MW of GCMT were used as markers. Then aBP neural network was trained to establish a moment magnitude conversion model. Meanwhile, least square regression and orthogonal regression were also applied to buildtraditionallinear models for comparisons. The results show that mean absolute error and root mean square error of least square regression and orthogonal regression are about 40% higher than that of BP neural network. Furthermore, the residual error of our proposed method has smaller absolute value and more concentrated distribution.

    参考文献
    蔡润, 武震, 云欢, 等, 2018. 基于BP和SOM神经网络相结合的地震预测研究. 四川大学学报(自然科学版), 55(2):307~315.
    陈宏峰, 袁菲, 徐志国, 等, 2014. 使用中国地震台网资料快速测定中强地震矩震级. 地震地磁观测与研究, 35(5~6):51~57.
    陈运泰, 刘瑞丰, 2004. 地震的震级. 地震地磁观测与研究, 25(6):1~12.
    陈运泰, 刘瑞丰, 2018. 矩震级及其计算. 地震地磁观测与研究, 39(2):1~9.
    丁海平, 刘中良, 李燕杰, 2011. 台湾地震对福建地区基岩地震动的影响. 地震工程与工程振动, 31(4):26~32.
    冯利华, 2000. 基于人工神经网络的地震活动性研究. 西北地震学报, 22(4):402~406.
    蒋淳, 冯德益, 汪德馨, 等, 1994. 神经网络模型在地震预报中的某些应用. 中国地震, 10(3):262~269.
    李小军, 陈苏, 任治坤, 等, 2020. 海域地震区划关键技术研究项目及研究进展. 地震科学进展, 50(1):2~19.
    林彬华, 金星, 陈惠芳, 等, 2019. 基于反向传播神经网络的闽台ML震级偏差分析与修正. 地震学报, 41(6):723~734.
    林趾祥, 晁洪太, 1999. 加强海域地震研究及其意义. 见:中国地震学会成立20周年纪念文集. 北京:中国地震学会, 94~98.
    刘光鼎, 1992. 中国海地球物理场和地球动力学特征. 地质学报, 66(4):300~314.
    刘瑞丰, 陈运泰, Bormann P, 等, 2005. 中国地震台网与美国地震台网测定震级的对比(Ⅰ)——体波震级. 地震学报, 27(6):583~587.
    刘瑞丰, 陈运泰, Bormann P, 等, 2006. 中国地震台网与美国地震台网测定震级的对比(Ⅱ)——面波震级. 地震学报, 28(1):1~7.
    刘瑞丰, 陈运泰, 任枭, 等, 2007. 中国地震台网震级的对比. 地震学报, 29(5):467~476.
    刘瑞丰, 陈运泰, 薛峰, 2018. 测定的震级之间不应相互换算. 地震地磁观测与研究, 39(3):1~9.
    潘华, 高孟潭, 谢富仁, 2013. 新版地震区划图地震活动性模型与参数确定. 震灾防御技术, 8(1):11~23.
    彭艳菊, 孟小红, 吕悦军, 等, 2008. 我国近海地震活动特征及其与地球物理场的关系. 地球物理学进展, 23(5):1377~1388.
    沙海军, 吕悦军, 2018. 中国地震台网面波震级与矩震级统计关系. 地震地磁观测与研究, 39(6):31~36.
    孙印, 潘素珍, 刘明军, 2018. 天然地震识别与震相自动拾取技术进展. 中国地震, 34(4):606~620.
    汪素云, 俞言祥, 2009. 震级转换关系及其对地震活动性参数的影响研究. 震灾防御技术, 4(2):141~149.
    王炜, 宋先月, 2000. 人工神经网络在地震中短期预报中的应用. 中国地震, 16(2):149~157.
    王钰清, 陆文凯, 刘金林, 等, 2019. 基于数据增广和CNN的地震随机噪声压制. 地球物理学报, 62(1):421~433.
    吴果, 周庆, 冉洪流, 2014. 中亚地震目录震级转换及其完整性分析. 震灾防御技术, 9(3):368~383.
    项月文, 饶泓, 汤兰荣, 等, 2015. 基于SOM和BP神经网络的地震预报技术. 地震地磁观测与研究, 36(4):139~144.
    肖亮, 2011. 水平向基岩强地面运动参数衰减关系研究. 博士学位论文. 北京:中国地震局地球物理研究所.
    谢卓娟, 李山有, 吕悦军, 等, 2020. 中国海域及邻区统一地震目录及其完整性分析. 地震地质, 42(4):993~1019.
    袁爱璟, 王伟君, 彭菲, 等, 2021. 机器学习在地震预测中的应用进展. 地震, 41(1):51~66.
    张彭达, 戴志阳, 查显杰, 2020. 基于深度卷积神经网络的剪切波分裂质量检测. 中国地震, 36(3):539~549.
    Abrahamson N, Gregor N, Addo K, 2016. BC Hydro ground motion prediction equations for subduction earthquakes. Earthq Spectra, 32(1):23~44.
    Abrahamson N, Kuehn N, Gulerce Z, et al, 2018. Update of the BC hydro subduction ground-motion model using the NGA-subduction dataset. Berkeley:Pacific Earthquake Engineering Research Center.
    Asim K M, Javed F, Hainzl S, et al, 2019. Fault parameters-based earthquake magnitude estimation using artificial neural networks. Seismol Res Lett, 90(4):1544~1551.
    Bormann P, Liu R F, Ren X, et al, 2007. Chinese national network magnitudes, their relation to NEIC magnitudes, and recommendations for new IASPEI magnitude standards. Bull SeismolSoc Am, 97(1B):114~127.
    Bormann P, Liu R F, Xu Z G, et al, 2009. First application of the new IASPEI teleseismic magnitude standards to data of the China National Seismographic Network. Bull SeismolSoc Am, 99(3):1868~1891.
    Castellaro S, Bormann P, 2007. Performance of different regression procedures on the magnitude conversion problem. Bull SeismolSoc Am, 97(4):1167~1175.
    Castellaro S, Mulargia F, Kagan YY, 2006. Regression problems for magnitudes. Geophys J Int, 165(3):913~930.
    Cheng J, Rong Y F, Magistrale H, et al, 2017. An MW-based historical earthquake catalog for Mainland China. Bull SeismolSoc Am, 107(5):2490~2500.
    Das R, Meneses C, 2021. A unified moment magnitude earthquake catalog for Northeast India. Geomatics, Nat Hazards Risk, 12(1):167~180.
    Di Giacomo D, Harris J, Storchak D A, 2021. Complementing regional moment magnitudes to GCMT:a perspective from the rebuilt International Seismological Centre Bulletin. Earth SystSci Data, 13(5):1957~1985.
    Kishida T, Contreras V, Bozorgnia Y, et al, 2018a. NGA-Sub ground motion database. In:Eleventh U.S. National Conference on Earthquake Engineering. Los Angeles:UCLA.
    Kishida T, Derakhshan S, Muin S, et al, 2018b. Multivariate conversion of moment magnitude for small-to-moderate-magnitude earthquakes in Iran. Earthq Spectra, 34(1):313~326.
    Konstantinou K I, Melis N S, 2018. The relationship between local and moment magnitude in Greece during the period 2008-2016.Pure ApplGeophys, 175(3):731~740.
    Kros J F, Lin M, Brown M L, 2006. Effects of the neural network s-Sigmoid function on KDD in the presence of imprecise data. ComputOper Res, 33(11):3136~3149.
    KumarR, YadavRBS, CastellaroS, 2020. Regional earthquake magnitude conversion relations for the Himalayan seismic belt. Seismol Res Lett, 91(6):3195~3207.
    Lolli B, Gasperini P, Rebez A, 2018. Homogenization in terms of MW of local magnitudes of Italian earthquakes that occurred before 1981.Bull SeismolSoc Am, 108(1):481~492.
    Lolli B, Randazzo D, Vannucci G, et al, 2020. The homogenized instrumental seismic catalog(HORUS) of Italy from 1960 to present. SeismolRes Lett, 91(6):3208~3222.
    Manzunzu B, Brandt M B C, Midzi V, et al, 2021.Towards a homogeneous moment magnitude determination for earthquakes in South Africa:reduction of associated uncertainties. JAfr Earth Sci, 173:104051.
    Pandey A K, Chingtham P, Roy P N S, 2017.Homogeneous earthquake catalogue for Northeast region of India using robust statistical approaches. Geomatics, Nat Hazards Risk, 8(2):1477~1491.
    Tian YY, Xu C, Hong H Y, et al, 2019. Mapping earthquake-triggered landslide susceptibility by use of artificial neural network(ANN) models:an example of the 2013 Minxian(China) MW5.9 event. Geomatics, Nat HazardsRisk, 10(1):1~25.
    Wason H R, Das R, Sharma M L, 2012. Magnitude conversion problem using general orthogonal regression. Geophys J Int, 190(2):1091~1096.
    Wu G, Zhou Q, Ran H L, et al, 2019.Long-term probabilistic forecast for M ≥ 5.0 earthquakes in the Eastern Tibetan Plateau from adaptively smoothed seismicity. Bull SeismolSoc Am, 109(3):1110~1124.
    Xie Z J, Li S Y, Lyu Y J, et al, 2021. Empirical relations for conversion of surface- and body-wave magnitudes to moment magnitudes in China's seas and adjacent areas. J Seismol, 25(1):213~233.
    Xu W J, 2019. Probabilistic seismic hazard assessment using spatially smoothed seismicity in North China seismic zone. J Seismol, 23(3):613~622.
    Xu W J, Gao M T, Zuo H Q, 2021. Generation of a stochastic seismic event set based on a new seismicity model in China's earthquake catastrophe model. Seismol Res Lett, 92(4):2308~2320.
    引证文献
    网友评论
    网友评论
    分享到微博
    发 布
引用本文

吴果,李建强,冉洪流,周庆,谢卓娟.中国海域及邻区基于BP神经网络的矩震级转换模型[J].中国地震,2021,37(3):659-670

复制
分享
文章指标
  • 点击次数:454
  • 下载次数: 922
  • HTML阅读次数: 339
  • 引用次数: 0
历史
  • 收稿日期:2021-04-27
  • 最后修改日期:2021-05-26
  • 在线发布日期: 2021-12-24
文章二维码
您是第2856891位访问者
中国地震 ® 2025 版权所有
技术支持:北京勤云科技发展有限公司
请使用 Firefox、Chrome、IE10、IE11、360极速模式、搜狗极速模式、QQ极速模式等浏览器,其他浏览器不建议使用!