石油地球物理勘探2026，Vol.61Issue(1)：34-45,12.DOI:10.13810/j.cnki.issn.1000-7210.20250268
应用注意力增强特征UNet的地震速度建模
Seismic velocity modeling based on attention-enhanced UNet
摘要
Abstract
Accurate underground velocity information is crucial for seismic imaging in complex area.While ex-isting seismic waveform inversion techniques are highly accurate,they have shortcomings such as high computa-tional amount and reliance on initial models.Currently,deep learning technology experiences rapid advance-ments in various fields and has successfully been applied to nonlinear seismic inversion.However,conven-tional end-to-end deep learning networks struggle to establish a multi-scale physical coupling relationship be-tween velocity parameters and seismic records.To this end,this paper proposes a hybrid network AER-UNet,which reorganizes the encode and decoder structures and adds an attention mechanism-based jumping connection module on this basis.This approach effectively obtains key spatial information from seismic records and en-hances the representation of the subtle structures in velocity fields,thus accurately capturing the characteristics of underground medium velocity parameters.An appropriate number of random velocity models should be built in the network training phase to simulate the true structure of the underground medium and thus obtain the accu-rate mapping relationship between velocity models and seismic records.Additionally,developing new loss func-tions can help improve the computational accuracy of velocity modeling.By carrying out numerical experiments using the SEG/EAGE thrust model,the effectiveness of the hybrid network for velocity modeling is evaluated.Compared to FWI and other deep learning networks,this method can more efficiently and accurately rebuild un-derground velocity models.关键词
速度建模/深度学习/注意力机制/损失函数Key words
velocity modeling/deep learning/attention mechanism/loss function分类
天文与地球科学引用本文复制引用
刘文革,谢雨柔,杜增利,李浩,熊鹏超..应用注意力增强特征UNet的地震速度建模[J].石油地球物理勘探,2026,61(1):34-45,12.基金项目
本项研究受中石油塔里木油田公司"揭榜挂帅"科技项目"塔西南山前地震处理成像攻关"(671023060002)和中国石油—西南石油大学创新联合体科技合作项目"地震自聚焦成像、多信息约束波形反演与地质解释一体化关键技术"(2020CX010202)联合资助. （671023060002）
