SEARCH · Engineering Papers
Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.
Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.
Explore the source record for details and available documents.
Explore the source record for details and available documents.
Explore the source record for details and available documents.
SUMMARY Full-waveform inversion has become an essential technique for mapping geophysical subsurface structures. However, proper uncertainty quantification is often lacking in current applications. In theory, uncertainty quantification is related to the inverse Hessian (or the posterior covariance matrix). Even for common geophysical inverse problems its calculation is beyond the computational and storage capacities of the largest high-performance computing systems. In this study, we amend the Broyden–Fletcher–Goldfarb–Shanno (BFGS) algorithm to perform uncertainty quantification for large-scale applications. For seismic inverse problems, the limited-memory BFGS (L-BFGS) method prevails as the most efficient quasi-Newton method. We aim to augment it further to obtain an approximate inverse Hessian for uncertainty quantification in FWI. To facilitate retrieval of the inverse Hessian, we combine BFGS (essentially a full-history L-BFGS) with randomized singular value decomposition to determine a low-rank approximation of the inverse Hessian. Setting the rank number equal to the number of iterations makes this solution efficient and memory-affordable even for large-scale problems. Furthermore, based on the Gauss–Newton method, we formulate different initial, diagonal Hessian matrices as pre-conditioners for the inverse scheme and compare their performances in elastic FWI applications. We highlight our approach with the elastic Marmousi benchmark model, demonstrating the applicability of pre-conditioned BFGS for large-scale FWI and uncertainty quantification.
Abstract Inverse analysis has been utilized to understand unknown underground geological properties by matching the observational data with simulators. To overcome the underconstrained nature of inverse problems and achieve good performance, an approach is presented with embedded physics and a technique known as algorithmic differentiation. We use a physics-embedded generative model, which takes statistically simple parameters as input and outputs subsurface properties (e.g., permeability or P-wave velocity), that embeds physical knowledge of the subsurface properties into inverse analysis and improves its performance. We tested the application of this approach on four geologic problems: two heterogeneous hydraulic conductivity fields, a hydraulic fracture network, and a seismic inversion for P-wave velocity. This physics-embedded inverse analysis approach consistently characterizes these geological problems accurately. Furthermore, the excellent performance in matching the observational data demonstrates the reliability of the proposed method. Moreover, the application of algorithmic differentiation makes this an easy and fast approach to inverse analysis when dealing with complicated geological structures.
This study builds and expands on previous CO 2 storage resource assessment studies of the southeastern offshore Atlantic margin by providing a detailed evaluation on how rock porosities and permeabilities are distributed across the Upper Cretaceous strata restricted to the South Georgia Embayment (SGE). Using legacy industry two-dimensional seismic reflection and well data, this assessment is the first application of multiple seismic inversion techniques in this area. This workflow provides a reliable and repeatable model-based inversion which gives an improved image to discriminate lithology and predict porosity. The workflow is applicable to future CO 2 storage resource assessment studies elsewhere. The inversion results indicate that distinct porosity and permeability regimes are present and distributed in the Upper Cretaceous strata within the SGE. The impedance and porosity relationships show well-founded and reliable correlation. These relationships reveal low impedance coincident to the high porosity intervals which are proposed as potential reservoir intervals for CO 2 storage. In addition, the result shows that the Upper Cretaceous strata have two main potential reservoirs in the lower part. These are overlain by a thick impermeable interval, mostly shale, which has high impedance, low porosity, and low permeability and extends within the SGE. This result is in agreement with a previous study that also proposed two significant storage reservoirs for CO 2 in the Upper Cretaceous strata. Since porosity distribution is estimated using multiple methods, it follows the trends of seismic signature and structures of the Upper Cretaceous strata. The extracted values of porosity, ranging from 15 to 36%, and permeability, ranging from 1 to 100 mD, are close to the measured values from the well core data at the Upper Cretaceous strata interval.
Carbon capture and storage is being pursued globally as a geoengineering measure for reducing the emission of anthropogenic CO 2 the atmosphere. Comprehensive monitoring, verification, and accounting programs must be established for demonstrating the safe storage of injected CO 2 . One of the most commonly deployed monitoring techniques is time-lapse seismic reservoir monitoring (also known as 4-D seismic), which involves comparing 3-D seismic survey data taken at the same study site but over different times. Analyses of 4-D seismic data volumes can help improve the quality of storage reservoir characterization, track the movement of injected CO 2 plume, and identify potential CO 2 spillover/leakage from the storage reservoirblue. However, the derivation of high-resolution CO 2 saturation maps from 4-D seismic data is a highly nonlinear and ill-posed inverse problem, often requiring significant computational effort. In this research, we apply a physics-based deep learning method to facilitate the solution of both the forward and inverse problems in seismic inversion while honoring physical constraints. A cycle generative adversarial neural network (CycleGAN) model is trained to learn the bidirectional functional mappings between the reservoir dynamic property changes and seismic attribute changes, such that both forward and inverse solutions can be obtained efficiently from the trained model. We show that our CycleGAN-based approach not only improves the reliability of 4-D seismic inversion but also expedites the quantitative interpretation. Our deep learning-based workflow is generic and can be readily used for reservoir characterization and reservoir model updates involving the use of 4-D seismic data.
Here, we present deep learning (DL) networks for three-dimensional (3D) joint inversion of active seismic full waveform and passive seismic traveltime data to image reservoirs and their properties and quantify imaging uncertainties. Active seismic full-waveform data can provide high-resolution monitoring images but are collected only intermittently because of their high acquisition cost. In contrast, passive seismic data can be gathered at relatively low cost between regular active surveys, although their imaging quality can be compromised by factors such as low signal-to-noise ratios and limited ray coverage of the target. Although these datasets are routinely acquired together at CO 2 storage sites, their combined inversion within a 3D DL framework has not been previously demonstrated. To our knowledge, this is the first study to address this gap, combining the strength of both data types. For efficient data storage and DL training with large 3D seismic datasets, we use a 3D data matrix in which a random number of passive seismic traveltime data are stored as parabolic envelopes using one-hot encoding and a 3D full-waveform data matrix in which multiple shot gathers are summed. Two network architectures are evaluated: a single-encoder U-Net for single-data type inversion and a dual-encoder U-Net for joint inversion of active and passive seismic data. We also evaluate the single-encoder U-Net for joint inversion by concatenating full-waveform data and traveltime data. We propose a systematic approach for selecting an optimal dropout rate that balances regularization during training and Monte Carlo dropout-based uncertainty quantification during prediction by examining the correlation coefficient between standard deviation and prediction error, along with the training misfit, across a range of dropout rates. 3D DL inversion experiments include five different network configurations, with evaluations under ideal, noisy and dropout-enabled conditions. Both model and data uncertainties are assessed, as well as their combined effects. Across all conditions, the networks consistently predict accurate CO 2 saturation models with low prediction errors, such as a structural similarity index of 0.993 and CO 2 difference of 1.1%. Uncertainty estimates show strong spatial correlation with prediction errors, confirming the effectiveness of the proposed dropout selection approach. The results demonstrate that our DL approach, utilizing compact data representations and appropriate uncertainty quantification, yields accurate subsurface images under various inversion conditions and provides valuable insights into the reliability of predictions.
Abstract Partial differential equation (PDE)-constrained inverse problems are some of the most challenging and computationally demanding problems in computational science today. Fine meshes required to accurately compute the PDE solution introduce an enormous number of parameters and require large-scale computing resources such as more processors and more memory to solve such systems in a reasonable time. For inverse problems constrained by time-dependent PDEs, the adjoint method often employed to compute gradients and higher order derivatives efficiently requires solving a time-reversed, so-called adjoint PDE that depends on the forward PDE solution at each timestep. This necessitates the storage of a high-dimensional forward solution vector at every timestep. Such a procedure quickly exhausts the available memory resources. Several approaches that trade additional computation for reduced memory footprint have been proposed to mitigate the memory bottleneck, including checkpointing and compression strategies. In this work, we propose a close-to-ideal scalable compression approach using autoencoders to eliminate the need for checkpointing and substantial memory storage, thereby reducing the time-to-solution and memory requirements. We compare our approach with checkpointing and an off-the-shelf compression approach on an earth-scale ill-posed seismic inverse problem. The results verify the expected close-to-ideal speedup for the gradient and Hessian-vector product using the proposed autoencoder compression approach. To highlight the usefulness of the proposed approach, we combine the autoencoder compression with the data-informed active subspace (DIAS) prior showing how the DIAS method can be affordably extended to large-scale problems without the need for checkpointing and large memory.
Fast assimilation of monitoring data to update forecasts of pressure buildup and carbon dioxide (CO 2 ) plume migration under geologic uncertainties is a challenging problem in geologic carbon storage. The high computational cost of data assimilation with a high-dimensional parameter space impedes fast decision-making for commercial-scale reservoir management. We propose to leverage physical understandings of porous medium flow behavior with deep learning techniques to develop a fast data assimilation-reservoir response forecasting workflow. Applying an Ensemble Smoother Multiple Data Assimilation (ES-MDA) framework, the workflow updates geologic properties and predicts reservoir performance with quantified uncertainty from pressure history and CO 2 plumes interpreted through seismic inversion. As the most computationally expensive component in such a workflow is reservoir simulation, we developed surrogate models to predict dynamic pressure and CO 2 plume extents under multi-well injection. The surrogate models employ deep convolutional neural networks, specifically, a wide residual network and a residual U-Net. The workflow is validated against a flat threedimensional reservoir model representative of a clastic shelf depositional environment. Intelligent treatments are applied to bridge between quantities in a true-3D reservoir model and those in a single-layer reservoir model. The workflow can complete history matching and reservoir forecasting with uncertainty quantification in less than one hour on a mainstream personal workstation.
Abstract A well-recognized characteristic of the World Stress Map (WSM) database is the continued presence of large spatial gaps in the distribution of the data records despite the more than 40-year development history of the database. The current release has more than 30 000 high-quality (A–C) data records (often referred to as ‘stress indicators’), but while some continental areas (such as Australia) have seen a significant increase in spatial converge with the latest release, other continental regions (Africa, central Asia, most of South America) remain markedly sparse. In this contribution we (1) review the current state of the spatial distribution of stress indicators in the continental regions (above sea-level); (2) quantify the clustering of the stress indicators in the latest WSM release using the Hopkins statistic as a way to explore the current spatial distribution of the indicators and assess future WSM releases; and (3) present three approaches (joint inversion, seismic anisotropy and InSAR) that provide a way to fill in the gaps (both in the S Hmax orientation and principal stress magnitudes) in regions that lack active seismicity and where borehole drilling is cost prohibitive. These three approaches have the potential to guide procedures for improving a priori estimates of the ambient stress field in the Earth's crust and reduce the uncertainty in predicting both the magnitude and orientation of the principal tectonic stresses.
We have analyzed a 3D seismic survey acquired for a carbon sequestration project on top of the Moxa Arch in southwestern Wyoming. We observed a zone of discontinuous reflectors on vertical slices of seismic amplitude volume, whereas, the northwest–southeast lineations were observed on the time slices. We performed a seismic to well tie that suggested that the lineations occur within the Nugget Sandstone. The Nugget Sandstone is an eolian sandstone deposit of Early Jurassic age, deposited as a subtropical dune field, and equivalent to the Navajo Sandstone of southwestern Utah. Petrophysical analysis indicates that the Nugget Sandstone is dominated by clean sandstone (70%–80%), whereas evaporites, including halite and anhydrite, are present in certain zones. Previous outcrop studies on the Navajo Sandstone indicate the wind direction to be predominantly northeast–southwest. Seismic attributes, including coherence and curvature, displayed on stratal slices within the Nugget Sandstone interval indicate the presence of lineations in the northwest–southeast direction with irregular spacing. These lineations are approximately perpendicular to the inferred dominant wind direction. We computed the dominant wind direction from the average azimuth of the lineations as seen on the curvature attribute in the Nugget Sandstone interval. Geological feature: Eolian sand dunes with interdunal evaporites Seismic appearance: Parallel lineations with irregular spacing on seismic attribute horizon slices Alternative interpretations: Canyons at continental slopes; slope failures Features with a similar appearance: Marine bars; contourites Formation: The Nugget Sandstone — equivalent to the Navajo Sandstone Age: Early Jurassic Location: Moxa Arch, Wyoming Seismic data: Obtained by the University of Wyoming with U.S. DOE funding Contributors: Dhruv Agrawal, Brady Lujan, Sumit Verma, Shuvajit Bhattacharya, and Subhashis Mallick Analysis tools: Coherence and curvature attributes; seismic inversion; petrophysical inversion
Conference presentation at American Institute of Chemical Engineers (AIChE) Annual Meeting, Phoenix, AZ, November 13–18, 2022. An analysis of rock physics crossplots and diagnostic models suggests that the baseline seismic data may be sensitive to changes in lithology and porosity of the Broom Creek Formation and that these changes may be detectable by prestack seismic inversion.
Seismic full-waveform inversion (FWI), which uses iterative methods to estimate high-resolution subsurface models from seismograms, is a powerful imaging technique in exploration geophysics. In recent years, the computational cost of FWI has grown exponentially due to the increasing size and resolution of seismic data. Moreover, it is a nonconvex problem and can encounter local minima due to the limited accuracy of the initial velocity models or the absence of low frequencies in the measurements. To overcome these computational issues, we develop a multiscale data-driven FWI method based on fully convolutional networks (FCNs). In preparing the training data, we first develop a real-time style transform method to create a large set of synthetic subsurface velocity models from natural images. We then develop two convolutional neural networks with encoder-decoder structures to reconstruct the low- and high-frequency components of the subsurface velocity models, separately. To validate the performance of our data-driven inversion method and the effectiveness of the synthesized training set, we compare it with conventional physics-based waveform inversion approaches using both synthetic and field data. Finally, these numerical results demonstrate that, once our model is fully trained, it can significantly reduce the computation time and yield more accurate subsurface velocity models in comparison with conventional FWI.
Seismic source characterization is an essential function of global nuclear explosion monitoring (NEM). While large events (roughly with moment magnitude, M w , greater than 5.0) can often be easily detected, located and identified with high signal-to-noise ratios at teleseismic distances (> 20°), trends in NEM research require confident source characterization at much lower magnitudes (say down to 3.0) and exploitation of sparse observations (from only a few stations) at regional distance (< 20°). Regional distance waveform inversion to characterize sources is now widely used and effective (e.g. Ford et al., 2009; Alvizuri and Tape, 2018; Alvizuri et al., 2018; Chiang et al., 2018; Ford et al., 2022). These methods obtain the magnitude, depth and seismic moment tensor, which represents the forces that excited the observed seismic waves (slip on an earthquake fault, explosion, collapse or a combination of various forces). Common to many problems in seismology, the isolation of the source 2 properties requires removal of path propagation effects that waves experience while traveling through the three-dimensional (3D) Earth (the structure exists due to different rock types, material properties, temperature and tectonic processes).
Explore the source record for details and available documents.
Explore the source record for details and available documents.
In full waveform inversion (FWI) high-resolution subsurface model parameters are sought. FWI is normally treated as a nonlinear least-squares inverse problem, in which the minimum of the corresponding misfit function is found by updating the model parameters. When multiple elastic or acoustic properties are solved for, simple gradient methods tend to confuse parameter classes. This is referred to as parameter cross-talk; it leads to incorrect model solutions, poor convergence and strong dependence on the scaling of the different parameter types. Determining step lengths in a subspace domain, rather than directly in terms of gradients of different parameters, is a potentially valuable approach to address this problem. The particular subspace used can be defined over a span of different sets of data or different parameter classes, provided it involves a small number of vectors compared to those contained in the whole model space. Additionally, in a subspace method, the basis vectors are defined first, and a local minimum is found in the space spanned by these. We examine the application of the subspace method within acoustic FWI in determining simultaneously updates for velocity and density. We first discuss the choice of basis vectors to construct the spanned space, from linear updates by distinguishing only the contributions of different parameter classes towards nonlinear updates by adding the contributions of higher-order perturbations of each parameter class. The numerical character of FWI solutions generated via subspace methods involving different basis vectors is then analyzed and compared with traditional FWI methods. The subspace methods can provide better reconstructions of the model, especially for the velocity, as well as improved convergence rates, while the computational costs are still comparable with the traditional FWI methods.