{"title":"Acoustic Impedance Prediction Using an Attention-Based Dual-Branch Double-Inversion Network","authors":"Wen Feng, Yong Li, Yingtian Liu, Huating Li","doi":"arxiv-2408.02524","DOIUrl":null,"url":null,"abstract":"Seismic impedance inversion is a widely used technique for reservoir\ncharacterization. Accurate, high-resolution seismic impedance data form the\nfoundation for subsequent reservoir interpretation. Deep learning methods have\ndemonstrated significant potential in seismic impedance inversion. Traditional\nsingle semi-supervised networks, which directly input original seismic logging\ndata, struggle to capture high-frequency weak signals. This limitation leads to\nlow-resolution inversion results with inadequate accuracy and stability.\nMoreover, seismic wavelet uncertainty further constrains the application of\nthese methods to real seismic data. To address these challenges, we propose\nADDIN-I: an Attention-based Dual-branch Double-Inversion Network for Impedance\nprediction. ADDIN-I's dual-branch architecture overcomes the limitations of\nsingle-branch semi-supervised networks and improves the extraction of\nhigh-frequency weak signal features in sequence modeling. The network\nincorporates an attention mechanism to further enhance its feature extraction\ncapabilities. To adapt the method for real seismic data applications, a deep\nlearning forward operator is employed to fit the wavelet adaptively. ADDIN-I\ndemonstrates excellent performance in both synthetic and real data\napplications.","PeriodicalId":501270,"journal":{"name":"arXiv - PHYS - Geophysics","volume":"7 1","pages":""},"PeriodicalIF":0.0000,"publicationDate":"2024-08-05","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-2408.02524","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0
Abstract
Seismic impedance inversion is a widely used technique for reservoir
characterization. Accurate, high-resolution seismic impedance data form the
foundation for subsequent reservoir interpretation. Deep learning methods have
demonstrated significant potential in seismic impedance inversion. Traditional
single semi-supervised networks, which directly input original seismic logging
data, struggle to capture high-frequency weak signals. This limitation leads to
low-resolution inversion results with inadequate accuracy and stability.
Moreover, seismic wavelet uncertainty further constrains the application of
these methods to real seismic data. To address these challenges, we propose
ADDIN-I: an Attention-based Dual-branch Double-Inversion Network for Impedance
prediction. ADDIN-I's dual-branch architecture overcomes the limitations of
single-branch semi-supervised networks and improves the extraction of
high-frequency weak signal features in sequence modeling. The network
incorporates an attention mechanism to further enhance its feature extraction
capabilities. To adapt the method for real seismic data applications, a deep
learning forward operator is employed to fit the wavelet adaptively. ADDIN-I
demonstrates excellent performance in both synthetic and real data
applications.