煤田地质与勘探2024，Vol.52Issue(11)：132-140,9.DOI:10.12363/issn.1001-1986.24.02.0136
嵌入傅里叶神经算子的卷积自编码声波速度反演方法
Acoustic velocity inversion based on convolutional autoencoder embedded with Fourier neural operator
摘要
Abstract
[Background]Seismic wave inversion serves as an effective method for obtaining the characteristics of struc-tures,lithology,and physical properties of subsurface media using the arrival times,amplitude,and waveforms of seis-mic waves.Seismic inversion methods based on wave equations iteratively update model parameters using forward mod-eling.This generally involves extensive numerical simulations and optimization calculations,requiring large quantities of computational resources and time.In recent years,neural operators for deep learning,represented by the Fourier neur-al operator(FNO),have gained widespread attention.However,the original FNO structure fails to effectively learn the wavefield information with sharp changes in geological structures in the seismic wave inversion of complex media,lead-ing to low accuracy of inversion results.[Objective and Methods]To enhance the accuracy and generalization perform-ance of FNO in learning seismic wavefield information under complex geological models,this study developed a novel acoustic velocity inversion method-Convolutional autoencoder embedded with Fourier neural operator(CAE-FNO),which utilized an encoder for feature extraction and performed efficient training based on FNO to effectively capture the fine features of the seismic wavefield and improved prediction accuracy.During the network training,the CAE-FNO method progressively reduced the size of the Fourier mode,thus effectively reducing the number of network parameters while enhancing the generalization capability of the network.[Results and Conclusions]The numerical experiments on homogeneous,heterogeneous,layered,and Marmousi2 models demonstrate that the CAE-FNO method exhibited signi-ficantly higher inversion accuracy than FNO and its variants UFNO and UNO.The experiments on the homogeneous model revealed that the velocity inversion results of the CAE-FNO method had a relative error of 1.3%and those of UFNO,UNO,and FNO exhibited relative errors of 1.7%,2.3%,and up to 10.1%,respectively.In the experiments on the heterogeneous model,CAE-FNO yielded accurate inversion results of geological structures and velocity change posi-tions,whereas UFNO and UNO exhibited higher errors for zones with sharp velocity fluctuations.During the experi-ments on the layered model,CAE-FNO clearly distinguished minor velocity changes between layers,while FNO failed.For both smooth zones and zones with abrupt changes in the Marmousi2 model,CAE-FNO exhibited higher accuracy in capturing irregular interfaces with velocity changes than UFNO and UNO,while FNO failed to effectively handle the ab-rupt changes in velocity and detail changes in these zones.Therefore,the CAE-FNO method,demonstrating small loss functions and high accuracy,enjoys advantages in the inversion of complex media,providing a novel research philo-sophy for seismic inversion.关键词
地震波反演/傅里叶神经算子/卷积自编码器/深度学习/数据驱动Key words
seismic wave inversion/Fourier neural operator/convolutional autoencoder/deep learning/data-driven分类
天文与地球科学引用本文复制引用
李谌,赵海霞,白钊蔚,郝禹帆..嵌入傅里叶神经算子的卷积自编码声波速度反演方法[J].煤田地质与勘探,2024,52(11):132-140,9.基金项目
国家重点研发计划"变革性技术关键科学问题"重点专项项目(2021YFA0716901) （2021YFA0716901）
国家自然科学基金面上项目(41974132) （41974132）
中央高校基本科研业务费专项资金项目(xzy012023050) （xzy012023050）
