{"title":"高效三维贝叶斯全波形反演和先验假设分析","authors":"Xuebin Zhao, Andrew Curtis","doi":"arxiv-2409.09746","DOIUrl":null,"url":null,"abstract":"Spatially 3-dimensional seismic full waveform inversion (3D FWI) is a highly\nnonlinear and computationally demanding inverse problem that constructs 3D\nsubsurface seismic velocity structures using seismic waveform data. To\ncharacterise non-uniqueness in the solutions we demonstrate Bayesian 3D FWI\nusing an efficient method called physically structured variational inference\napplied to 3D acoustic Bayesian FWI. The results provide reasonable posterior\nuncertainty estimates, at a computational cost that is only an order of\nmagnitude greater than that of standard, deterministic FWI. Furthermore, we\ndeploy variational prior replacement to calculate Bayesian solutions\ncorresponding to different classes of prior information at low additional cost,\nand analyse those prior hypotheses by constructing Bayesian L-curves. This\nreveals the sensitivity of the inversion process to different prior\nassumptions. Thus we show that fully probabilistic 3D FWI can be performed at a\ncost that may be practical in small FWI problems, and can be used to test\ndifferent prior hypotheses.","PeriodicalId":501270,"journal":{"name":"arXiv - PHYS - Geophysics","volume":"6 1","pages":""},"PeriodicalIF":0.0000,"publicationDate":"2024-09-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Efficient 3D Bayesian Full Waveform Inversion and Analysis of Prior Hypotheses\",\"authors\":\"Xuebin Zhao, Andrew Curtis\",\"doi\":\"arxiv-2409.09746\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Spatially 3-dimensional seismic full waveform inversion (3D FWI) is a highly\\nnonlinear and computationally demanding inverse problem that constructs 3D\\nsubsurface seismic velocity structures using seismic waveform data. To\\ncharacterise non-uniqueness in the solutions we demonstrate Bayesian 3D FWI\\nusing an efficient method called physically structured variational inference\\napplied to 3D acoustic Bayesian FWI. The results provide reasonable posterior\\nuncertainty estimates, at a computational cost that is only an order of\\nmagnitude greater than that of standard, deterministic FWI. Furthermore, we\\ndeploy variational prior replacement to calculate Bayesian solutions\\ncorresponding to different classes of prior information at low additional cost,\\nand analyse those prior hypotheses by constructing Bayesian L-curves. This\\nreveals the sensitivity of the inversion process to different prior\\nassumptions. Thus we show that fully probabilistic 3D FWI can be performed at a\\ncost that may be practical in small FWI problems, and can be used to test\\ndifferent prior hypotheses.\",\"PeriodicalId\":501270,\"journal\":{\"name\":\"arXiv - PHYS - Geophysics\",\"volume\":\"6 1\",\"pages\":\"\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2024-09-15\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"arXiv - PHYS - Geophysics\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/arxiv-2409.09746\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"arXiv - PHYS - Geophysics","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/arxiv-2409.09746","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0
摘要
空间三维地震全波形反演(3D FWI)是一个高度非线性和计算要求极高的反演问题,它利用地震波形数据构建三维次表层地震速度结构。为了描述解的非唯一性,我们演示了贝叶斯三维 FWI,将一种称为物理结构变异推理的高效方法应用于三维声学贝叶斯 FWI。结果提供了合理的后验不确定性估计,计算成本仅比标准的确定性 FWI 高一个数量级。此外,我们还采用变异先验替换法,以较低的额外成本计算出与不同类别先验信息相对应的贝叶斯解,并通过构建贝叶斯 L 曲线对这些先验假设进行分析。这揭示了反演过程对不同先验假设的敏感性。因此,我们证明了全概率三维全维反演可以在小型全维反演问题中以实用的成本进行,并可用于测试不同的先验假设。
Efficient 3D Bayesian Full Waveform Inversion and Analysis of Prior Hypotheses
Spatially 3-dimensional seismic full waveform inversion (3D FWI) is a highly
nonlinear and computationally demanding inverse problem that constructs 3D
subsurface seismic velocity structures using seismic waveform data. To
characterise non-uniqueness in the solutions we demonstrate Bayesian 3D FWI
using an efficient method called physically structured variational inference
applied to 3D acoustic Bayesian FWI. The results provide reasonable posterior
uncertainty estimates, at a computational cost that is only an order of
magnitude greater than that of standard, deterministic FWI. Furthermore, we
deploy variational prior replacement to calculate Bayesian solutions
corresponding to different classes of prior information at low additional cost,
and analyse those prior hypotheses by constructing Bayesian L-curves. This
reveals the sensitivity of the inversion process to different prior
assumptions. Thus we show that fully probabilistic 3D FWI can be performed at a
cost that may be practical in small FWI problems, and can be used to test
different prior hypotheses.