Application of sparse S transform network with knowledge distillation in seismic attenuation delineation  

作　　者：Nai-Hao Liu Yu-Xin Zhang Yang Yang Rong-Chang Liu Jing-Huai Gao Nan Zhang 

机构地区：[1]School of Information and Communications Engineering,Xi'an Jiaotong University,Xi'an,710049,Shaanxi,China [2]School of Software Engineering,Xi'an Jiaotong University,Xi'an,710049,Shaanxi,China [3]PetroChina Research Institute of Petroleum Exploration and Development(RIPED),CNPC,Beijing,100083,China [4]Research Institute of Exploration and Development,Yumen Oilfield Company,CNPC,Jiuquan,735019,Gansu,China

出　　处：《Petroleum Science》2024年第4期2345-2355,共11页石油科学（英文版）

基　　金：supported by the National Natural Science Foundation of China (42274144,42304122,and 41974155);the Key Research and Development Program of Shaanxi (2023-YBGY-076);the National Key R&D Program of China (2020YFA0713404);the China Uranium Industry and East China University of Technology Joint Innovation Fund (NRE202107)。

摘　　要：Time-frequency analysis is a successfully used tool for analyzing the local features of seismic data.However,it suffers from several inevitable limitations,such as the restricted time-frequency resolution,the difficulty in selecting parameters,and the low computational efficiency.Inspired by deep learning,we suggest a deep learning-based workflow for seismic time-frequency analysis.The sparse S transform network(SSTNet)is first built to map the relationship between synthetic traces and sparse S transform spectra,which can be easily pre-trained by using synthetic traces and training labels.Next,we introduce knowledge distillation(KD)based transfer learning to re-train SSTNet by using a field data set without training labels,which is named the sparse S transform network with knowledge distillation(KD-SSTNet).In this way,we can effectively calculate the sparse time-frequency spectra of field data and avoid the use of field training labels.To test the availability of the suggested KD-SSTNet,we apply it to field data to estimate seismic attenuation for reservoir characterization and make detailed comparisons with the traditional time-frequency analysis methods.

关 键 词：S transform Deep learning Knowledge distillation Transfer learning Seismic attenuation delineation 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程] P631.4[自动化与计算机技术—控制科学与工程]