Fei Li, Qiang Mao, Juan Chen, Yan Huang, Jianping Huang
The anisotropy and attenuation properties of real earth media can lead to amplitude reduction and phase dispersion as seismic waves propagate through it. Ignoring these effects will degrade the resolution of seismic imaging profiles, thereby affecting the accuracy of geological interpretation. To characterize the impacts of viscosity and anisotropy, we formulate a modified pure-viscoacoustic (PU-V) wave equation including the decoupled fractional Laplacian (DFL) for tilted transversely isotropic (TTI) media, which enables the generation of stable wavefields that are resilient to noise interference. Numerical tests show that the newly derived PU-V wave equation is capable of accurately simulating the viscoacoustic wavefields in anisotropic media with strong attenuation. Building on our TTI PU-V wave equation, we implement stable reverse time migration technique with attenuation compensation (Q-TTI RTM), effectively migrating the impacts of anisotropy and compensates for attenuation. In the Q-TTI RTM workflow, to remove the unstable high-frequency components in attenuation compensated wavefields, we construct a stable attenuation compensated wavefield modeling (ACWM) operator. The proposed stable ACWM operator consists of velocity anisotropic and attenuation anisotropic parameters, effectively suppressing the high-frequency artifacts in the attenuation compensated wavefield. Synthetic examples demonstrate that our stable Q-TTI RTM technique can simultaneously and accurately correct for the influences of anisotropy and attenuation, resulting in the high-quality imaging results.
{"title":"Stable Q-compensated reverse time migration in TTI media based on a modified fractional Laplacian pure-viscoacoustic wave equation","authors":"Fei Li, Qiang Mao, Juan Chen, Yan Huang, Jianping Huang","doi":"10.1093/jge/gxae066","DOIUrl":"https://doi.org/10.1093/jge/gxae066","url":null,"abstract":"\u0000 The anisotropy and attenuation properties of real earth media can lead to amplitude reduction and phase dispersion as seismic waves propagate through it. Ignoring these effects will degrade the resolution of seismic imaging profiles, thereby affecting the accuracy of geological interpretation. To characterize the impacts of viscosity and anisotropy, we formulate a modified pure-viscoacoustic (PU-V) wave equation including the decoupled fractional Laplacian (DFL) for tilted transversely isotropic (TTI) media, which enables the generation of stable wavefields that are resilient to noise interference. Numerical tests show that the newly derived PU-V wave equation is capable of accurately simulating the viscoacoustic wavefields in anisotropic media with strong attenuation. Building on our TTI PU-V wave equation, we implement stable reverse time migration technique with attenuation compensation (Q-TTI RTM), effectively migrating the impacts of anisotropy and compensates for attenuation. In the Q-TTI RTM workflow, to remove the unstable high-frequency components in attenuation compensated wavefields, we construct a stable attenuation compensated wavefield modeling (ACWM) operator. The proposed stable ACWM operator consists of velocity anisotropic and attenuation anisotropic parameters, effectively suppressing the high-frequency artifacts in the attenuation compensated wavefield. Synthetic examples demonstrate that our stable Q-TTI RTM technique can simultaneously and accurately correct for the influences of anisotropy and attenuation, resulting in the high-quality imaging results.","PeriodicalId":54820,"journal":{"name":"Journal of Geophysics and Engineering","volume":null,"pages":null},"PeriodicalIF":1.4,"publicationDate":"2024-06-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141356291","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
In this paper, we introduce the deep numerical technique DeepNM, which is designed for solving one-dimensional (1D) hyperbolic conservation laws, particularly wave equations. By creatively integrating traditional numerical schemes with deep learning techniques, the method yields improvements over conventional approaches. Specifically, we compare this approach against two established classical numerical methods: the Discontinuous Galerkin method (DG) and the Lax-Wendroff correction method (LWC). While maintaining a comparable level of accuracy, DeepNM significantly improves computational speed, surpassing conventional numerical methods in this aspect by more than tenfold, and reducing storage requirements by over 1000 times. Furthermore, DeepNM facilitates the utilization of higher-order numerical schemes and allows for an increased number of grid points, thereby enhancing precision. In contrast to the more prevalent PINN method, DeepNM optimally combines the strengths of conventional mathematical techniques with deep learning, resulting in heightened accuracy and expedited computations for solving partial differential equations (PDEs). Notably, DeepNM introduces a novel research paradigm for numerical equation-solving that can be seamlessly integrated with various traditional numerical methods.
{"title":"A deep learning operator-based numerical scheme method for solving 1-D wave equations","authors":"Yunfan Chang, Dinghui Yang, Xijun He","doi":"10.1093/jge/gxae062","DOIUrl":"https://doi.org/10.1093/jge/gxae062","url":null,"abstract":"\u0000 In this paper, we introduce the deep numerical technique DeepNM, which is designed for solving one-dimensional (1D) hyperbolic conservation laws, particularly wave equations. By creatively integrating traditional numerical schemes with deep learning techniques, the method yields improvements over conventional approaches. Specifically, we compare this approach against two established classical numerical methods: the Discontinuous Galerkin method (DG) and the Lax-Wendroff correction method (LWC). While maintaining a comparable level of accuracy, DeepNM significantly improves computational speed, surpassing conventional numerical methods in this aspect by more than tenfold, and reducing storage requirements by over 1000 times. Furthermore, DeepNM facilitates the utilization of higher-order numerical schemes and allows for an increased number of grid points, thereby enhancing precision. In contrast to the more prevalent PINN method, DeepNM optimally combines the strengths of conventional mathematical techniques with deep learning, resulting in heightened accuracy and expedited computations for solving partial differential equations (PDEs). Notably, DeepNM introduces a novel research paradigm for numerical equation-solving that can be seamlessly integrated with various traditional numerical methods.","PeriodicalId":54820,"journal":{"name":"Journal of Geophysics and Engineering","volume":null,"pages":null},"PeriodicalIF":1.4,"publicationDate":"2024-06-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141355375","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The trial-and-error method for calibrating rock mechanics parameters has the disadvantages in complexity, time-consuming and difficulty in ensuring accuracy. Harnessing the repeatability and scalability intrinsic to numerical simulation calculations and amalgamating them with the data-driven attributes of machine learning methods. The study utilised the finite element analysis software RS2 to establish 252 sets of sandstone sample data. The Recursive Feature Elimination and Cross-Validation (RFECV) method was employed for feature selection. The shear strength parameters of sandstone were predicted using machine learning models optimised by Particle Swarm Optimization (PSO) algorithm, including BP neural network (BP), Bayesian Ridge Regression (BRR), Support Vector Regression (SVR), and Light Gradient Boosting Machine (LightGBM). The predicted value of cohesion is proposed as the input feature to predict the friction angle. The results indicate that the optimal input characteristics for predicting cohesion are elastic modulus, Poisson's ratio, peak stress, and peak strain, while the optimal input characteristics for predicting friction angle are peak stress and cohesion. The PSO-SVR model demonstrates the best performance. The maximum error between the predicted values of cohesion and friction angle and the calculated results of RSData program is 3.5% and 4.31%, respectively. The finite element calculation is in good agreement with the stress-strain curve obtained in the laboratory. The sensitivity analysis indicates that SVR's prediction performance for cohesion and friction angle tends to be stable when the sample size is greater than 25. These results offer a valuable reference for accurately predicting rock mechanics parameters.
{"title":"Fusion of finite element and machine learning methods to predict rock shear strength parameters","authors":"Defu Zhu, Biaobiao Yu, Deyu Wang, Yujiang Zhang","doi":"10.1093/jge/gxae064","DOIUrl":"https://doi.org/10.1093/jge/gxae064","url":null,"abstract":"\u0000 The trial-and-error method for calibrating rock mechanics parameters has the disadvantages in complexity, time-consuming and difficulty in ensuring accuracy. Harnessing the repeatability and scalability intrinsic to numerical simulation calculations and amalgamating them with the data-driven attributes of machine learning methods. The study utilised the finite element analysis software RS2 to establish 252 sets of sandstone sample data. The Recursive Feature Elimination and Cross-Validation (RFECV) method was employed for feature selection. The shear strength parameters of sandstone were predicted using machine learning models optimised by Particle Swarm Optimization (PSO) algorithm, including BP neural network (BP), Bayesian Ridge Regression (BRR), Support Vector Regression (SVR), and Light Gradient Boosting Machine (LightGBM). The predicted value of cohesion is proposed as the input feature to predict the friction angle. The results indicate that the optimal input characteristics for predicting cohesion are elastic modulus, Poisson's ratio, peak stress, and peak strain, while the optimal input characteristics for predicting friction angle are peak stress and cohesion. The PSO-SVR model demonstrates the best performance. The maximum error between the predicted values of cohesion and friction angle and the calculated results of RSData program is 3.5% and 4.31%, respectively. The finite element calculation is in good agreement with the stress-strain curve obtained in the laboratory. The sensitivity analysis indicates that SVR's prediction performance for cohesion and friction angle tends to be stable when the sample size is greater than 25. These results offer a valuable reference for accurately predicting rock mechanics parameters.","PeriodicalId":54820,"journal":{"name":"Journal of Geophysics and Engineering","volume":null,"pages":null},"PeriodicalIF":1.4,"publicationDate":"2024-06-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141357473","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Weihua Liu, Yang Wang, Hui Shen, Min Li, Wenhao Fan
Given the growing importance of organic-rich shale as unconventional reservoirs, a thorough understanding of the elastic and anisotropic behavior of shales is of great concern. However, for lacustrine shales, the complex lithofacies assemblage with geological deposition makes it challenging. Four lithofacies (argillaceous, mixed, siliceous, and calcareous) are recognized for 40 lacustrine shale samples from Jurassic formation in Sichuan basin based on their mineral compositions. We perform ultrasonic velocity measurements on 40 pairs of shale plugs at varied confining pressures, attempting to uncover the controls on the anisotropic properties of different lithofacies. The experimental results reveal that the total porosity, clay, and organic matter would positively contribute to velocity anisotropy of Jurassic shales. Combined with micro-structure and pressure-dependent velocity analysis, the preferred orientations of platy clay particles and lenticular kerogen, the development of clay pores along clay fabric, and the subparallel micro-cracks induced by hydrocarbon expulsion are treated to be the controlling mechanisms. We sum the total porosity, clay content, and kerogen volume together, intending to distinguish the elastic and anisotropic properties of four lithofacies. Generally, argillaceous shales, the dominant lithofacies in Jurassic formation, could be characterized by the highest clay and TOC content, the lowest bedding-normal velocities, and the strongest velocity anisotropy. Finally, with the laboratory data, two rock-physics-driven exponential relationships are proposed to predict the P- and S-wave velocity anisotropy with the bedding-normal velocities.
鉴于富含有机质的页岩作为非常规储层的重要性与日俱增，全面了解页岩的弹性和各向异性行为是非常重要的。然而，对于湖相页岩来说，地质沉积的复杂岩性组合使其具有挑战性。我们根据四川盆地侏罗系地层中 40 个湖相页岩样本的矿物成分，确定了四种岩性（霰粒质、混合、硅质和钙质）。我们在不同约束压力下对 40 对页岩塞进行了超声波速度测量，试图揭示不同岩性各向异性的控制因素。实验结果表明，总孔隙度、粘土和有机质会对侏罗纪页岩的速度各向异性产生积极影响。结合微观结构和随压力变化的速度分析，我们认为板状粘土颗粒和透镜状角质的优先取向、粘土孔隙沿粘土结构的发展以及烃类排出引起的近平行微裂缝是其控制机制。我们将总孔隙度、粘土含量和角质体积相加，以区分四种岩性的弹性和各向异性。一般来说，侏罗纪地层中的主要岩性--霰粒页岩具有粘土和总有机碳含量最高、层位速度最低、速度各向异性最强的特点。最后，根据实验室数据，提出了两种岩石物理学驱动的指数关系，以预测 P 波和 S 波速度各向异性与层位正常速度的关系。
{"title":"Ultrasonic velocity anisotropy of jurassic shales with different lithofacies","authors":"Weihua Liu, Yang Wang, Hui Shen, Min Li, Wenhao Fan","doi":"10.1093/jge/gxae061","DOIUrl":"https://doi.org/10.1093/jge/gxae061","url":null,"abstract":"\u0000 Given the growing importance of organic-rich shale as unconventional reservoirs, a thorough understanding of the elastic and anisotropic behavior of shales is of great concern. However, for lacustrine shales, the complex lithofacies assemblage with geological deposition makes it challenging. Four lithofacies (argillaceous, mixed, siliceous, and calcareous) are recognized for 40 lacustrine shale samples from Jurassic formation in Sichuan basin based on their mineral compositions. We perform ultrasonic velocity measurements on 40 pairs of shale plugs at varied confining pressures, attempting to uncover the controls on the anisotropic properties of different lithofacies. The experimental results reveal that the total porosity, clay, and organic matter would positively contribute to velocity anisotropy of Jurassic shales. Combined with micro-structure and pressure-dependent velocity analysis, the preferred orientations of platy clay particles and lenticular kerogen, the development of clay pores along clay fabric, and the subparallel micro-cracks induced by hydrocarbon expulsion are treated to be the controlling mechanisms. We sum the total porosity, clay content, and kerogen volume together, intending to distinguish the elastic and anisotropic properties of four lithofacies. Generally, argillaceous shales, the dominant lithofacies in Jurassic formation, could be characterized by the highest clay and TOC content, the lowest bedding-normal velocities, and the strongest velocity anisotropy. Finally, with the laboratory data, two rock-physics-driven exponential relationships are proposed to predict the P- and S-wave velocity anisotropy with the bedding-normal velocities.","PeriodicalId":54820,"journal":{"name":"Journal of Geophysics and Engineering","volume":null,"pages":null},"PeriodicalIF":1.4,"publicationDate":"2024-06-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141265285","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Monitoring underground fluid migration caused by injection/production processes is crucial for guiding petroleum exploitation in mature oilfields and ultimately enhancing petroleum production. In this paper, we propose a time-lapse reverse time imaging (RTI) to dynamically monitor the injection/production processes within oilfield. By utilizing RTI to track microseimicities at different time periods, we can analyze the relationship between injection/production activities and the spatiotemporal changes in microseismic distribution. The inferred relationship enables the time-lapse RTI to infer fluid migration patterns within oil reservoirs. To assess the accuracy and spatiotemporal resolution of the time-lapse RTI, we conducted numerical experiments to evaluate the imaging quality under different microseismic distribution scenarios. In addition, we assessed the method's stability under low signal-to-noise ratio conditions. Numerical results indicate that the time-lapse RTI can effectively distinguish the spatiotemporal variations of seismic swarms at depths of 0.5 kilometers within the target layer, even in the presence of strong noise. Practical applications show a significant correlation between changes in swarm distribution surrounding reservoirs and fluctuations in oil production. Utilizing time-lapse RTI enables real-time monitoring of oilfield injection/production processes, thereby offering valuable insights for optimizing oilfield development and fostering future increases in petroleum production.
{"title":"Passive seismic monitoring of injection-production process in oilfield using reverse time imaging","authors":"Runbi Yuan, Zhihui Zou, Song Xu, Wenhuan Kuang","doi":"10.1093/jge/gxae060","DOIUrl":"https://doi.org/10.1093/jge/gxae060","url":null,"abstract":"\u0000 Monitoring underground fluid migration caused by injection/production processes is crucial for guiding petroleum exploitation in mature oilfields and ultimately enhancing petroleum production. In this paper, we propose a time-lapse reverse time imaging (RTI) to dynamically monitor the injection/production processes within oilfield. By utilizing RTI to track microseimicities at different time periods, we can analyze the relationship between injection/production activities and the spatiotemporal changes in microseismic distribution. The inferred relationship enables the time-lapse RTI to infer fluid migration patterns within oil reservoirs. To assess the accuracy and spatiotemporal resolution of the time-lapse RTI, we conducted numerical experiments to evaluate the imaging quality under different microseismic distribution scenarios. In addition, we assessed the method's stability under low signal-to-noise ratio conditions. Numerical results indicate that the time-lapse RTI can effectively distinguish the spatiotemporal variations of seismic swarms at depths of 0.5 kilometers within the target layer, even in the presence of strong noise. Practical applications show a significant correlation between changes in swarm distribution surrounding reservoirs and fluctuations in oil production. Utilizing time-lapse RTI enables real-time monitoring of oilfield injection/production processes, thereby offering valuable insights for optimizing oilfield development and fostering future increases in petroleum production.","PeriodicalId":54820,"journal":{"name":"Journal of Geophysics and Engineering","volume":null,"pages":null},"PeriodicalIF":1.4,"publicationDate":"2024-06-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141267293","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Lei Li, Xiaotao Wen, Chao Tang, Dongyong Zhou, Songgen Zhang
Based on the wave equation, scholars worldwide have proposed various methods for numerical simulation of seismic wave propagation in underground and surface media. The finite element method offers a unique advantage in accurately depicting the undulating surfaces and steep palaeoburial hills with its triangular mesh. However, its computational efficiency cannot meet our needs while lots of memories are occupied. To address this, we optimized and improved the critical Mass matrix and Stiffness matrix of spatial discretization of the acoustic wave equation. We first fully utilized the flexibility of triangles to fit different undulating terrains, then reorganized the numbering of triangle mesh nodes and elements to reduce the bandwidth of the matrices, and then used optimized matrices for solving. The Crank-Nicolson scheme was adopted for time discretization, and the Perfectly Matched Layer condition was utilized to eliminate false waves reflected from the boundary. The numerical experiments with simple and significant fluctuation models proved that this method can accelerate computational efficiency while ensuring computational accuracy.
{"title":"Numerical simulation of acoustic waves propagation by finite element method based on optimized matrices","authors":"Lei Li, Xiaotao Wen, Chao Tang, Dongyong Zhou, Songgen Zhang","doi":"10.1093/jge/gxae055","DOIUrl":"https://doi.org/10.1093/jge/gxae055","url":null,"abstract":"\u0000 Based on the wave equation, scholars worldwide have proposed various methods for numerical simulation of seismic wave propagation in underground and surface media. The finite element method offers a unique advantage in accurately depicting the undulating surfaces and steep palaeoburial hills with its triangular mesh. However, its computational efficiency cannot meet our needs while lots of memories are occupied. To address this, we optimized and improved the critical Mass matrix and Stiffness matrix of spatial discretization of the acoustic wave equation. We first fully utilized the flexibility of triangles to fit different undulating terrains, then reorganized the numbering of triangle mesh nodes and elements to reduce the bandwidth of the matrices, and then used optimized matrices for solving. The Crank-Nicolson scheme was adopted for time discretization, and the Perfectly Matched Layer condition was utilized to eliminate false waves reflected from the boundary. The numerical experiments with simple and significant fluctuation models proved that this method can accelerate computational efficiency while ensuring computational accuracy.","PeriodicalId":54820,"journal":{"name":"Journal of Geophysics and Engineering","volume":null,"pages":null},"PeriodicalIF":1.4,"publicationDate":"2024-05-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141100597","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Full waveform inversion (FWI) can simultaneously update low-to-medium wavenumber velocity components and high-wavenumber velocity components. However, if seismic data lack large-offset data and effective low-frequency components, FWI updates will be dominated by high-wavenumber velocity perturbation. Meanwhile, providing that the initial model is inaccurate, inversion will have the problem of local minima. In this study, FWI is developed with structural regularizing constraint based on gradient decomposition (RGDFWI). By correlating the separated forward wavefield and backward wavefield with specific propagating direction, FWI gradient is decomposed into tomography-mode gradient and migration-mode gradient. We propose an optimized strategy taking full advantage of the two modes of FWI gradient. On the one hand, we use tomography-mode gradient to enhance low-to-medium wavenumber updates. On the other hand, we use migration-mode gradient to apply structural regularizing constraint by estimating structure dip and adding sparsity constraint in Seislet domain. During the inversion process, high-wavenumber structural information constrains and guides low-wavenumber model updates. The results of two numerical tests, Marmousi model test and Overthrust model test, validate the optimized strategy, which can produce a better initial velocity model for FWI. The inversion finally generates a high-precision and high-resolution velocity model.
{"title":"Waveform inversion with structural regularizing constraint based on gradient decomposition","authors":"Ziying Wang, Jianhua Wang, Wenbo Sun, Jianping Huang, Zhenchun Li, Yandong Wang","doi":"10.1093/jge/gxae057","DOIUrl":"https://doi.org/10.1093/jge/gxae057","url":null,"abstract":"\u0000 Full waveform inversion (FWI) can simultaneously update low-to-medium wavenumber velocity components and high-wavenumber velocity components. However, if seismic data lack large-offset data and effective low-frequency components, FWI updates will be dominated by high-wavenumber velocity perturbation. Meanwhile, providing that the initial model is inaccurate, inversion will have the problem of local minima. In this study, FWI is developed with structural regularizing constraint based on gradient decomposition (RGDFWI). By correlating the separated forward wavefield and backward wavefield with specific propagating direction, FWI gradient is decomposed into tomography-mode gradient and migration-mode gradient. We propose an optimized strategy taking full advantage of the two modes of FWI gradient. On the one hand, we use tomography-mode gradient to enhance low-to-medium wavenumber updates. On the other hand, we use migration-mode gradient to apply structural regularizing constraint by estimating structure dip and adding sparsity constraint in Seislet domain. During the inversion process, high-wavenumber structural information constrains and guides low-wavenumber model updates. The results of two numerical tests, Marmousi model test and Overthrust model test, validate the optimized strategy, which can produce a better initial velocity model for FWI. The inversion finally generates a high-precision and high-resolution velocity model.","PeriodicalId":54820,"journal":{"name":"Journal of Geophysics and Engineering","volume":null,"pages":null},"PeriodicalIF":1.4,"publicationDate":"2024-05-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141104939","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Yajie Liu, Yan Zhang, Cheng Guo, Song Zhang, Houqin Kang, Qing Zhao
As an emerging geophysical exploration technology in recent years, airborne electromagnetic exploration has the advantages of adapting to diverse terrains, wide coverage, and providing a large amount of electromagnetic data, and can be applied to the rapid collection of large amounts of data. Scenarios are often used in fields such as deep geological structures, mineral resource exploration, and environmental engineering research. However, traditional airborne electromagnetic data inversion technology usually takes a long time to process a large amount of airborne electromagnetic data, and it is difficult to remove the noise in the later signals. Therefore, this paper proposes a multi-task learning network structure based on Transformer. By constraining the two network branches of imaging and denoising, a sub-network with simultaneous denoising and imaging is established to process aeronautical electromagnetic data. The noise test set is introduced for testing. This model achieved a 582.61% signal-to-noise ratio improvement in smooth Gaussian noise denoising, and a 129.69% and 112.74% signal-to-noise ratio improvement in non-smooth Gaussian noise and random impulse noise denoising, respectively. The method proposed in this article overcomes the shortcomings of traditional inversion imaging such as slow speed and low resolution, and at the same time eliminates the influence of noise in airborne electromagnetic data. This is of great significance for the application of deep learning in the field of geophysical exploration.
{"title":"A multi-task learning network based on transformer network for airborne electromagnetic detection imaging and denoising","authors":"Yajie Liu, Yan Zhang, Cheng Guo, Song Zhang, Houqin Kang, Qing Zhao","doi":"10.1093/jge/gxae054","DOIUrl":"https://doi.org/10.1093/jge/gxae054","url":null,"abstract":"\u0000 As an emerging geophysical exploration technology in recent years, airborne electromagnetic exploration has the advantages of adapting to diverse terrains, wide coverage, and providing a large amount of electromagnetic data, and can be applied to the rapid collection of large amounts of data. Scenarios are often used in fields such as deep geological structures, mineral resource exploration, and environmental engineering research. However, traditional airborne electromagnetic data inversion technology usually takes a long time to process a large amount of airborne electromagnetic data, and it is difficult to remove the noise in the later signals. Therefore, this paper proposes a multi-task learning network structure based on Transformer. By constraining the two network branches of imaging and denoising, a sub-network with simultaneous denoising and imaging is established to process aeronautical electromagnetic data. The noise test set is introduced for testing. This model achieved a 582.61% signal-to-noise ratio improvement in smooth Gaussian noise denoising, and a 129.69% and 112.74% signal-to-noise ratio improvement in non-smooth Gaussian noise and random impulse noise denoising, respectively. The method proposed in this article overcomes the shortcomings of traditional inversion imaging such as slow speed and low resolution, and at the same time eliminates the influence of noise in airborne electromagnetic data. This is of great significance for the application of deep learning in the field of geophysical exploration.","PeriodicalId":54820,"journal":{"name":"Journal of Geophysics and Engineering","volume":null,"pages":null},"PeriodicalIF":1.4,"publicationDate":"2024-05-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140969003","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Wei Shi, Weihong Wang, Ying Shi, S. Chen, Zhiwei Li, Ning Wang
Multiple reflections are among the most challenging noises to suppress in seismic data, as they differ from effective waves only in terms of apparent velocity. Besides, the Radon transform, an essential technique for attenuating multiple reflections, has been widely incorporated into various commercial software packages. Thus, this study introduces a 3D Radon transform method based on the LP‒1 norm to enhance sparsity-constraining capability in the transform domain, leveraging high-resolution Radon transform techniques. Specifically, an iteratively reweighted least squares (IRLS) algorithm is employed to obtain the transformed data in the Radon domain. Given that the LP‒1 norm is applied to seismic data processing for the first time, this paper theoretically demonstrates its powerful sparsity-constraining capability. Indeed, the proposed strategy enhances energy concentration in the Radon transform domain, better-separating primaries from multiples and ultimately suppressing the multiples. Both model tests and real data indicate that the 3D Radon transform constrained by the LP‒1 norm outperforms existing high-resolution Radon transform methods with sparsity constraints regarding energy concentration and effectiveness in multiple reflection attenuation.
{"title":"3D high-resolution Radon transform based on strong sparse LP‒1 norm and its applications","authors":"Wei Shi, Weihong Wang, Ying Shi, S. Chen, Zhiwei Li, Ning Wang","doi":"10.1093/jge/gxae052","DOIUrl":"https://doi.org/10.1093/jge/gxae052","url":null,"abstract":"\u0000 Multiple reflections are among the most challenging noises to suppress in seismic data, as they differ from effective waves only in terms of apparent velocity. Besides, the Radon transform, an essential technique for attenuating multiple reflections, has been widely incorporated into various commercial software packages. Thus, this study introduces a 3D Radon transform method based on the LP‒1 norm to enhance sparsity-constraining capability in the transform domain, leveraging high-resolution Radon transform techniques. Specifically, an iteratively reweighted least squares (IRLS) algorithm is employed to obtain the transformed data in the Radon domain. Given that the LP‒1 norm is applied to seismic data processing for the first time, this paper theoretically demonstrates its powerful sparsity-constraining capability. Indeed, the proposed strategy enhances energy concentration in the Radon transform domain, better-separating primaries from multiples and ultimately suppressing the multiples. Both model tests and real data indicate that the 3D Radon transform constrained by the LP‒1 norm outperforms existing high-resolution Radon transform methods with sparsity constraints regarding energy concentration and effectiveness in multiple reflection attenuation.","PeriodicalId":54820,"journal":{"name":"Journal of Geophysics and Engineering","volume":null,"pages":null},"PeriodicalIF":1.4,"publicationDate":"2024-05-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140994099","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
This study developed a noise-reduction method for acoustic logging array signals using a deep neural network algorithm in the time-frequency domain. Initially, we derived analytical solutions for the received waveforms when the acoustic logging tool was positioned either at the centre or eccentrically within the borehole. To simulate the received waveforms across various formations, we developed a real-axis integration algorithm. Subsequently, we devised a noise-reduction algorithm workflow based on a convolutional neural network (CNN) and configured the structure and parameters of the U-net using TensorFlow. To address the scarcity of open datasets, we established both signal and noise datasets. The signal dataset was generated using theoretical simulation encompassing various model parameters, while the noise dataset was collected during tool testing and downhole operations. The trained model demonstrated substantial noise-reduction capabilities during validation. To validate the effectiveness of the algorithm, we applied noise reduction to actual data collected during downhole operations in the TangGu oilfield, yielding impressive results across different types of noisy data. Therefore, the U-net-based time-domain noise-reduction algorithm proposed in this paper holds the potential to significantly improve the quality of acoustic logging array signals.
{"title":"Acoustic logging array signal denoising using U-net and a case study in TangGu oil field","authors":"Xin Fu, Yang Gou, Fuqiang Wei","doi":"10.1093/jge/gxae051","DOIUrl":"https://doi.org/10.1093/jge/gxae051","url":null,"abstract":"\u0000 This study developed a noise-reduction method for acoustic logging array signals using a deep neural network algorithm in the time-frequency domain. Initially, we derived analytical solutions for the received waveforms when the acoustic logging tool was positioned either at the centre or eccentrically within the borehole. To simulate the received waveforms across various formations, we developed a real-axis integration algorithm. Subsequently, we devised a noise-reduction algorithm workflow based on a convolutional neural network (CNN) and configured the structure and parameters of the U-net using TensorFlow. To address the scarcity of open datasets, we established both signal and noise datasets. The signal dataset was generated using theoretical simulation encompassing various model parameters, while the noise dataset was collected during tool testing and downhole operations. The trained model demonstrated substantial noise-reduction capabilities during validation. To validate the effectiveness of the algorithm, we applied noise reduction to actual data collected during downhole operations in the TangGu oilfield, yielding impressive results across different types of noisy data. Therefore, the U-net-based time-domain noise-reduction algorithm proposed in this paper holds the potential to significantly improve the quality of acoustic logging array signals.","PeriodicalId":54820,"journal":{"name":"Journal of Geophysics and Engineering","volume":null,"pages":null},"PeriodicalIF":1.4,"publicationDate":"2024-05-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140997030","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：