Proposed Workflow Research Articles

Considerations for the Design and Rapid Manufacturing of Pop‐Up MEMS Devices

AbstractThe rising number of devices created using pop‐up microelectromechanical systems (MEMS) and related folding‐based assembly techniques highlights the need for robust design and manufacturing workflows to support the wide range of device variations. To push the bounds of miniaturization, design for manufacturing is key in dealing with fabrication challenges. Iterative building of intermediate device prototypes is a promising way to explore an often very large design space and can highlight mechanical limitations that may not be obvious to the designer. Manufacturing of pop‐up MEMS devices is, however, typically a lengthy process with lamination taking a significant percentage of the build time, compared to layer machining and component assembly. To expedite the design process, researchers often prototype multiple iterations at larger scales prior to committing to at‐scale designs. In this study, an at‐scale rapid prototyping workflow for pop‐up MEMS devices is introduced. This study also includes flexure design considerations for castellated hinges, to approach the behavior of an ideal pin joint. The new proposed workflow uses more accessible lower‐cost equipment and materials and reduces lamination time by over 95% (10 min lamination vs a 3.5 h lamination) compared to the previous heated press process, which is validated through a series of prototypes.

Deep learning-based workflow for automatic extraction of atria and epicardial adipose tissue on cardiac computed tomography in atrial fibrillation.

Preoperative estimation of the volume of the left atrium (LA) and epicardial adipose tissue (EAT) on computed tomography (CT) images is associated with an increased risk of atrial fibrillation (AF) recurrence. We aimed to design a deep learning-based workflow to provide reliable automatic segmentation of the atria, pericardium, and EAT for future applications in the management of AF. This study enrolled 157 patients with AF who underwent first-time catheter ablation between January 2015 and December 2017 at Taipei Veterans General Hospital. Three-dimensional (3D) U-Net models of the LA, right atrium (RA), and pericardium were used to develop a pipeline for total, LA-EAT, and RA-EAT automatic segmentation. We defined fat within the pericardium as tissue with attenuation between -190 and -30 HU and quantified the total EAT. Regions between the dilated endocardial boundaries and endocardial walls of the LA or RA within the pericardium were used to detect voxels attributed to fat, thus estimating LA-EAT and RA-EAT. The LA, RA, and pericardium segmentation models achieved Dice coefficients of 0.960 ± 0.010, 0.945 ± 0.013, and 0.967 ± 0.006, respectively. The 3D segmentation models correlated well with the ground truth for the LA, RA, and pericardium ( r = 0.99 and p < 0.001 for all). The Dice coefficients of our proposed method for EAT, LA-EAT, and RA-EAT were 0.870 ± 0.027, 0.846 ± 0.057, and 0.841 ± 0.071, respectively. Our proposed workflow for automatic LA, RA, and EAT segmentation using 3D U-Nets on CT images is reliable in patients with AF.

Model Shows Computational Gains, Preserves Accuracy in Tight Rock EOR

_ This article, written by JPT Technology Editor Chris Carpenter, contains highlights of paper URTeC 3834855, “Advanced Dual-Porosity and Dual-Permeability Model for Tight Rock Integrated Primary Depletion and Enhanced Oil Recovery Simulation,” by Daegil Yang, SPE, Tyler Do, and Changdong Yang, SPE, Chevron, et al. The paper has not been peer reviewed. _ Discrete fracture or dual-porosity modeling has been widely adopted for tight rock simulation. To represent the complex fracture heterogeneity and simplify its characterization, the authors present in the complete paper a new simulation model to holistically optimize the tight rock reservoir completion, primary depletion, and enhanced oil recovery modeling and retain the injection mechanism and simplify the complex fracture characterization to enhance computational efficiency. The model enables users to characterize different properties in multiple regions. This tool will have broad applications in supporting asset-development decisions. Introduction Traditional dual-porosity, dual-permeability (DPDK) simulation incorporates microscale fracture characterization through the shape factor (matrix/fracture coupling), which offers a significant contact area for the gas/oil mixture to represent the field response reliably. However, its homogeneous fracture network cannot be used to optimize the completion design. The authors have developed what they refer to as the advanced DPDK (A-DPDK) simulation work flow to maintain heterogeneous fracture characterization and to simplify microfracture characterization. The work flow contains the following features: - A newly derived shape-factor algorithm that can represent the scale and heterogeneity of the fracture network - Local grid refinement (LGR) that accelerates the speed of computation (five-times-faster acceleration) and improves accuracy - A fracture-characterization feature that serves as a helpful diagnostic tool to assign a flexible cutoff value for both propped fractures (PF) and unpropped fractures to enhance simulation speed and robustness - A connected PF feature that can label the effective fracture region to simplify completion design work flow and prioritize reservoir simulation The authors validated the new features implemented in the work flow systematically and demonstrated the benefits of the proposed work flow. The authors also applied the work flow in one gas-injection pilot and successfully history-matched the primary depletion and the gas-injection response. Modeling Approach The A-DPDK model design is aimed at modeling a complex fracture network using a structured grid system while maintaining the background natural fracture. In the modeling process, the main fracture network is screened and identified and its transmissibility calculated. Then, the microfracture is calculated to matrix mass transfer. This can be numerically modeled using a DPDK framework. The geometry of the hydraulic fracture network is preserved explicitly as grid cells in the fracture domain. Also, in the matrix domain, the enhanced permeability region (EPR), which represents the microfracture, is included. The mass transfer between matrix cells and the EPR is modeled as a grid-cell connection (transmissibility), and the mass transfer between the hydraulic fracture cells and the microfractures is modeled as a grid-cell connection as well. A new way of calculating the coupling term that models the mass transfer between the hydraulic fracture and the EPR and the matrix is introduced.

Stochastic Economic Ranking Addresses Risk and Uncertainty for Decision-Makers

_ This article, written by JPT Technology Editor Chris Carpenter, contains highlights of paper SPE 213385, “Stochastic Economic Ranking: A Prudent Way To Address Risk and Uncertainty for Decision-Makers,” by Mahesh Narayanan, SPE, Mehaboob Abdulazeez, SPE, and Khadija Bukhamsin, SPE, Saudi Aramco, et al. The paper has not been peer reviewed. _ Project screening and ranking is inevitable even when an organization faces no resource bottlenecks. In project economic predictions, the stochastic approach is prevalent and proves to be the most natural approach to address technical and commercial uncertainties. In contrast with a deterministic approach, a stochastic Monte Carlo simulation helps in understanding the range of possible outcomes to enable effective decision-making. The objective of the complete paper is to present an integrated stochastic work flow that addresses the subsurface, surface, and cost uncertainties to screen, rank, and grade opportunities early in the derisking process. Introduction The proposed work flow simultaneously can address all technical and commercial uncertainties with a Monte Carlo model and predict a range of possible economic outcomes with customized economic analysis tools. In other words, it can achieve complete unbiased sample randomness in expected project outcomes. The work flow also creates a stage-gate decision-tree approach and reviews economics as a series of staged investments. This helps to expose incremental capital in the life cycle of a project responsibly and help with course correction where necessary. Value Addition: Stochastic Approach A common practice in the oil and gas industry is to simply examine low, middle, and high cases. This is a limited view of possible scenarios and provides a false sense of precision regarding expected project outcomes. The term “stochastic” refers to the property of being well-described by a random probability distribution. A stochastic approach features the following advantages: - A stochastic model addresses uncertainties in inputs with complete unbiased randomness in the outcomes. - By running thousands of calculations and using many different estimates of future economic conditions, stochastic models predict a range of possible future investment results, showing the potential upside and downside of each. - Once the work flow is followed and the model is set up, multiple realizations can be created with a single click. Proposed Integrated Stochastic Methodology In the world of project economics, classical stochastic economics is a tool to address uncertainty and risk. In simple terms, a mean or deterministic view of a production profile, along with one view of a schedule, is taken and blended with probabilistic distribution of cost inputs. This data is then taken through a Monte Carlo model to create multiple scenarios to predict a range of possible economic outcomes. This narrow range of economic outcomes only covers the uncertainty in cost but does not cover variability in well performance and time.

Reservoir heterogeneity analysis using multi-directional textural attributes from deep learning-based enhanced acoustic impedance inversion: A study from Poseidon, NW shelf Australia

Reservoir heterogeneities play a crucial role in governing reservoir performance and management. Traditionally, detailed and inter-well heterogeneity analyses are commonly performed by mapping seismic facies change in the seismic data, which is a time-intensive task. Many researchers have utilized a robust Grey-level co-occurrence matrix (GLCM)-based texture attributes to map reservoir heterogeneity. However, these attributes take seismic data as input and might not be sensitive to lateral lithology variation. To incorporate the lithology information, we have developed an innovative impedance-based texture approach using GLCM workflow by integrating 3D acoustic impedance volume (a rock property-based attribute) obtained from a deep convolution network-based impedance inversion. Our proposed workflow is anticipated to be more sensitive toward mapping lateral changes than the conventional amplitude-based texture approach, wherein seismic data is used as input. To evaluate the improvement, we applied the proposed workflow to the full-stack 3D seismic data from the Poseidon field, NW-shelf, Australia. This study demonstrates that a better demarcation of reservoir gas sands with improved lateral continuity is achievable with the presented approach compared to the conventional approach. In addition, we assess the implication of multi-stage faulting on facies distribution for effective reservoir characterization. This study also suggests a well-bounded potential reservoir facies distribution along the parallel fault lines. Thus, the proposed approach provides an efficient strategy by integrating the impedance information with texture attributes to improve the inference on reservoir heterogeneity, which can serve as a promising tool for identifying potential reservoir zones for both production benefits and fluid storage.

Hierarchical automated machine learning (AutoML) for advanced unconventional reservoir characterization

Recent advances in machine learning (ML) have transformed the landscape of energy exploration, including hydrocarbon, CO2 storage, and hydrogen. However, building competent ML models for reservoir characterization necessitates specific in-depth knowledge in order to fine-tune the models and achieve the best predictions, limiting the accessibility of machine learning in geosciences. To mitigate this issue, we implemented the recently emerged automated machine learning (AutoML) approach to perform an algorithm search for conducting an unconventional reservoir characterization with a more optimized and accessible workflow than traditional ML approaches. In this study, over 1000 wells from Alberta’s Athabasca Oil Sands were analyzed to predict various key reservoir properties such as lithofacies, porosity, volume of shale, and bitumen mass percentage. Our proposed workflow consists of two stages of AutoML predictions, including (1) the first stage focuses on predicting the volume of shale and porosity by using conventional well log data, and (2) the second stage combines the predicted outputs with well log data to predict the lithofacies and bitumen percentage. The findings show that out of the ten different models tested for predicting the porosity (78% in accuracy), the volume of shale (80.5%), bitumen percentage (67.3%), and lithofacies classification (98%), distributed random forest, and gradient boosting machine emerged as the best models. When compared to the manually fine-tuned conventional machine learning algorithms, the AutoML-based algorithms provide a notable improvement on reservoir property predictions, with higher weighted average f1-scores of up to 15–20% in the classification problem and 5–10% in the adjusted-R2 score for the regression problems in the blind test dataset, and it is achieved only after ~ 400 s of training and testing processes. In addition, from the feature ranking extraction technique, there is a good agreement with domain experts regarding the most significant input parameters in each prediction. Therefore, it is evidence that the AutoML workflow has proven powerful in performing advanced petrophysical analysis and reservoir characterization with minimal time and human intervention, allowing more accessibility to domain experts while maintaining the model’s explainability. Integration of AutoML and subject matter experts could advance artificial intelligence technology implementation in optimizing data-driven energy geosciences.

Calibration and data-analysis routines for nanoindentation with spherical tips

Instrumented spherical nanoindentation with a continuous stiffness measurement has gained increased popularity in microphysical investigations of grain boundaries, twins, dislocation densities, ion-induced damage, and more. These studies rely on different methodologies for instrument and tip calibration. Here, we test, integrate, and re-adapt published strategies for tip and machine-stiffness calibration for spherical tips. We propose a routine for independently calibrating the effective tip radius and the machine stiffness using standard reference materials, which requires the parametrization of the effective radius as a function of load. We validate our proposed workflow against key benchmarks and apply the resulting calibrations to data collected in materials with varying ductility to extract indentation stress–strain curves. We also test the impact of the machine stiffness on recently proposed methods for identification of yield stress. Finally, we synthesize these analyses in a single workflow for use in future studies aiming to extract and process data from spherical nanoindentation.Graphic abstract

Three-dimensional physical modeling of the wet manufacturing process of solid-state battery electrodes

Poor design of composite electrodes in Solid-State Batteries (SSBs) is one of the main reasons for their low performance. Although modeling techniques offer great potential in investigating multiple conditions for microstructure optimization, SSB manufacturing remains almost unaddressed in terms of computational studies. We present here a three-dimensional physics-based modeling workflow to investigate the impact of wet manufacturing process parameters on the properties of SSB tape casted composite electrodes based on LiNi0.6Mn0.2Co0.2O2 as the active material and Li6PSCl5 sulfide solid electrolyte. Our proposed workflow models the entire process of wet manufacturing of SSB electrodes, starting with the slurry consisting of active material, carbon additive, binder, solid electrolyte and solvent, and continues through its drying process, and the calendering of the resulting electrodes. Our focus is in particular on the impact of the calendering degree on the microstructure of the electrode. We characterize the resulting microstructures in terms of electronic and ionic conduction properties. We believe that this first-of-its-kind wet manufacturing process model for SSB cathodes is an important step towards the development of systematic modeling approaches that can provide practical optimization of the interfaces between materials in electrodes to improve SSB performance and durability.

A Fast and Cost-Effective Genotyping Method for CRISPR-Cas9-Generated Mutant Rice Lines.

With increasing throughput in both the generation and phenotyping of mutant lines in plants, it is important to have an efficient and reliable genotyping method. Traditional workflows, still commonly used in many labs, have time-consuming and expensive steps, such as DNA purification, cloning and growing E. coli cultures. We propose an alternative workflow where these steps are bypassed, using Phire polymerase on fresh plant tissue, and ExoProStar treatment as preparation for sequencing. We generated CRISPR-Cas9 mutants for ZAS (ZAXINONE SYNTHASE) in rice with two guide RNAs. Using both a traditional workflow and our proposed workflow, we genotyped nine T1 plants. To interpret the sequencing output, which is often complex in CRISPR-generated mutants, we used free online automatic analysis systems and compared the results. Our proposed workflow produces results of the same quality as the old workflow, but in 1 day instead of 3 days and about 35 times cheaper. This workflow also consists of fewer steps and reduces the risk of cross contamination and mistakes. Furthermore, the automated sequence analysis packages are mostly accurate and could easily be used for bulk analysis. Based on these advantages, we encourage academic and commercial labs conducting genotyping to consider switching over to our proposed workflow.

The role of hearing professionals for over-the-counter hearing aids

Adults with all levels of hearing loss have faced financial and accessibility constraints in obtaining prescription hearing aids. To increase access to hearing aids, the Food and Drug Administration (FDA) has created a new designation of over-the-counter (OTC) hearing aids for adults with perceived mild to moderate hearing loss. The goal of this article is to describe the incorporation of OTC hearing aid users into established audiology services. A proposed workflow was created by otolaryngologists and audiologists at Johns Hopkins University School of Medicine and is being implemented at outpatient audiology offices. We used published criteria for appropriate OTC hearing aid usage and disqualifying criteria hearing specialists should be aware of. The proposed workflow describes the multiple entry points for OTC hearing aid users seeking access to care and appropriate objectives for these visits. The new FDA ruling has increased the accessibility of hearing assistance devices; however, the varying degrees of retailer support systems could lead to improper fitting and other technical problems for some patients trialing new hearing aids. Our proposed workflow aims to describe the incorporation of established hearing assistance services with the new OTC hearing aid market.

Enabling cryo-EM density interpretation from yeast native cell extracts by proteomics data and AlphaFold structures.

In the cellular context, proteins participate in communities to perform their function. The detection and identification of these communities as well as in-community interactions has long been the subject of investigation, mainly through proteomics analysis with mass spectrometry. With the advent of cryogenic electron microscopy and the "resolution revolution," their visualization has recently been made possible, even in complex, native samples. The advances in both fields have resulted in the generation of large amounts of data, whose analysis requires advanced computation, often employing machine learning approaches to reach the desired outcome. In this work, we first performed a robust proteomics analysis of mass spectrometry (MS) data derived from a yeast native cell extract and used this information to identify protein communities and inter-protein interactions. Cryo-EM analysis of the cell extract provided a reconstruction of a biomolecule at medium resolution (∼8 Å (FSC=0.143)). Utilizing MS-derived proteomics data and systematic fitting of AlphaFold-predicted atomic models, this density was assigned to the 2.6 MDa complex of yeast fatty acid synthase. Our proposed workflow identifies protein complexes in native cell extracts from Saccharomyces cerevisiae by combining proteomics, cryo-EM, and AI-guided protein structure prediction.

Monitoring tree mortality in Ukrainian Pinus sylvestris L. forests using remote sensing data from earth observing satellites

This article considers the application of remote sensing data to solve the problems of forestry in the Polissia zone (Ukraine). The satellite remote sensing was shown to be applicable to monitoring the damage caused by diseases and pests to forest resources and to assessing the effects of fires. During the research, a detailed analysis and optimization of the information content of Sentinel-2 long-term data sets was performed to detect changes in the forest cover of Polissia, affected by pests and damaged by fires. The following classification algorithms were used for automated decryption: the maximum likelihood method; cluster classification without training; Principal Component Analysis (PCA); Random Forest classification. The results of this study indicate the high potential of Sentinel-2 data for application in applied problems of forestry and vegetation analysis, despite the decametric spatial resolution. Our proposed workflow has achieved an overall classification accuracy of 90 % for the Polissia region, indicating its reliability and potential for scaling to a higher level, and the proposed forecast model is stationary and does not depend on time parameters. To improve the classification results, testing of different combinations of bands emphasized the importance of Band 8 in combination with red edge bands, as well as other bands with a resolution of 10 m for summer scenes. The red margin shows clearly visible differences in the spectral profiles, but bands with a higher resolution of 10 m were crucial for good results.

Approach Evaluates Geothermal Potential in Existing Oil and Gas Wells

_ This article, written by JPT Technology Editor Chris Carpenter, contains highlights of paper SPE 207801, “Geothermal Production From Existing Oil and Gas Wells: A Sustainable Repurposing Model,” by Oscar M. Molina, SPE, and Camilo Mejia, Enovate Upstream, and Mayank Tyagi, SPE, Louisiana State University, et al. The paper has not been peer reviewed. _ In the complete paper, the authors discuss an integrated cloud-based work flow aimed at evaluating the cost-effectiveness of adopting geothermal production in low- to medium-enthalpy systems either by repurposing existing oil and gas wells or by coproducing thermal and fossil energy. The work flow introduces an automated and intrinsically secure decision-making process to convert mature oil and gas wells into geothermal wells, enabling both operational and financial assessment of the conversion process, whether partial or complete. Background The strong projection of geothermal-market growth allows oil and gas companies to consider geothermal energy coproduction from existing wells. This process considers, first, full conversion of wells in depleted hydrocarbon-bearing formations and, second, partial conversion to enable coproduction of hydrocarbons and economical volumes of water (wells with high water cut are suitable candidates for the partial-conversion approach). Aside from the financial gain resulting from repurposing candidate wells for geothermal production, the conversion process yields operational cost optimization by reducing the amount of wells in need of plugging and abandonment, extending the productive life of an otherwise depleted asset, and significantly reducing carbon dioxide emissions on location. Methodology The premise of the work flow is that candidate wells for conversion are producing hydrocarbons from low- to medium-enthalpy systems. The conversion from hydrocarbon to geothermal energy producer can be full or partial, meaning that a candidate well can produce water either by itself (full conversion) or alongside oil and gas (partial conversion). The authors discuss the fundamentals of the geothermal-conversion work flow, which is based on a well-structured and physics-informed decision-making process. Operational and financial metrics pertaining to the conversion process, whether partial or complete, are assessed by the work flow. The proposed work flow focuses on the reliability of the surface-to-subsurface model and the production system to ensure the financial success of the conversion project in terms of the realizable heat production and associated cost of development. The decision-making part of the work flow considers technical-, social-, and environmental-effect metrics that ultimately translate into return on investment (ROI) for either conversion case for geothermal production. All components of the work flow are evaluated using artificial intelligence algorithms that will help reduce biases in the decision-making process. The automated section of the work flow considers the assessment of the following components: - Heat potential - Integration of geological, petrophysical, and geophysical interpretation - Predictive methods for reservoir characterization - Dynamic reservoir modeling - Environmental, social, and governance (ESG) Each component is considered cloud-ready so that the entire work flow can be assembled in a cloud environment.

Detecting clonemate pairs in multicellular diploid clonal species based on a shared heterozygosity index.

Clonal reproduction, the formation of nearly identical individuals via mitosis in the absence of genetic recombination, is a very common reproductive mode across plants, fungi and animals. To detect clonal genetic structure, genetic similarity indices based on shared alleles are widely used, such as the Jaccard index, or identity by state. Here we propose a new pairwise genetic similarity index, the SH index, based on segregating genetic marker loci (typically single nucleotide polymorphisms) that are identically heterozygous for pairs of samples (NSH ). To test our method, we analyse two old seagrass clones (Posidonia australis, estimated to be around 8500 years old; Zostera marina, >750 years old) along with two young Z. marina clones of known age (17 years old). We show that focusing on shared heterozygosity amplifies the power to distinguish sample pairs belonging to different clones compared to methods focusing on all shared alleles. Our proposed workflow can successfully detect clonemates at a location dominated by a single clone. When the collected samples involve two or more clones, the SH index shows a clear gap between clonemate pairs and interclone sample pairs. Ideally NSH should be on the order of approximately ≥3000, a number easily achievable via restriction-site associated DNA (RAD) sequencing or whole-genome resequencing. Another potential application of the SH index is to detect possible parent-descendant pairs under selfing. Our proposed workflow takes advantage of the availability of the larger number of genetic markers in the genomic era, and improves the ability to distinguish clonemates from nonclonemates in multicellular diploid clonal species.

Deeper insights into long-term survival heterogeneity of pancreatic ductal adenocarcinoma (PDAC) patients using integrative individual- and group-level transcriptome network analyses

Pancreatic ductal adenocarcinoma (PDAC) is categorized as the leading cause of cancer mortality worldwide. However, its predictive markers for long-term survival are not well known. It is interesting to delineate individual-specific perturbed genes when comparing long-term (LT) and short-term (ST) PDAC survivors and integrate individual- and group-based transcriptome profiling. Using a discovery cohort of 19 PDAC patients from CHU-Liège (Belgium), we first performed differential gene expression analysis comparing LT to ST survivor. Second, we adopted systems biology approaches to obtain clinically relevant gene modules. Third, we created individual-specific perturbation profiles. Furthermore, we used Degree-Aware disease gene prioritizing (DADA) method to develop PDAC disease modules; Network-based Integration of Multi-omics Data (NetICS) to integrate group-based and individual-specific perturbed genes in relation to PDAC LT survival. We identified 173 differentially expressed genes (DEGs) in ST and LT survivors and five modules (including 38 DEGs) showing associations to clinical traits. Validation of DEGs in the molecular lab suggested a role of REG4 and TSPAN8 in PDAC survival. Via NetICS and DADA, we identified various known oncogenes such as CUL1 and TGFB1. Our proposed analytic workflow shows the advantages of combining clinical and omics data as well as individual- and group-level transcriptome profiling.

Integrated detection and investigation of bad borehole section in the Wolfcamp Formation in the Midland Basin using machine learning, petrophysics, and core characterization

The quality of formation evaluation and reservoir estimates depends a lot on the quality of open hole logs, which is intricately tied to the borehole geometry and stand-offs. Bad borehole sections cause interesting patterns on wireline logs that can result in multiple interpretations. We have conducted an automated identification of bad borehole sections (&gt;50 ft [15 m] thick) in multiple wells in the Wolfcamp Formation in the Midland Basin of the United States and investigated the causalities behind the unstable borehole section. The wells impacted by bad hole conditions appear to be within a particular region of the northern Midland Basin, and the bad hole condition impacts a specific zone in the Wolfcamp D. We consider well logs as multivariate time series (or depth series), where they share interdependencies to a large extent. We use this concept and unsupervised multivariate time series clustering to automatically identify bad borehole sections. Our proposed workflow simultaneously clusters the wireline logs, resulting in a large labeled regional-scale data set that can be further used for log quality assurance, processing, and petrophysical modeling. More importantly, we investigate the reasons behind the borehole washout, integrating sedimentology and geochemical data. Thick borehole washout sections are related to the presence of a significant amount of clay (&gt;73 wt%, a mixture of smectite and illite) in the Wolfcamp D. The results also indicate that the manual editing of wireline logs and even supervised machine learning-based log reconstruction (commonly known as a missing log prediction problem) can lead to erroneous and inconsistent interpretation of subsurface characteristics. Such exercise can be meaningless if derived models are not validated with the ground truth (preserved subsurface rocks). Geologic feature: Clay-rich unstable borehole section Log appearance: Large-to-small caliper response associated with erratic density and neutron porosity logs Features with similar appearance: Gypsum, siderite nodules, and tool failure Formation: Wolfcamp Formation Age: Wolfcampian Location: Northern Midland Basin, Texas, USA Well data: Obtained from the IHS Analysis tools: Unsupervised machine learning, core description, geochemistry, and petrophysical inversion modeling

Application of Machine Learning Solutions to Optimize Parameter Prediction to Enhance Automatic NMR Metabolite Profiling.

The quality of automatic metabolite profiling in NMR datasets from complex matrices can be affected by the numerous sources of variability. These sources, as well as the presence of multiple low-intensity signals, cause uncertainty in the metabolite signal parameters. Lineshape fitting approaches often produce suboptimal resolutions to adapt them in a complex spectrum lineshape. As a result, the use of software tools for automatic profiling tends to be restricted to specific biological matrices and/or sample preparation protocols to obtain reliable results. However, the analysis and modelling of the signal parameters collected during initial iteration can be further optimized to reduce uncertainty by generating narrow and accurate predictions of the expected signal parameters. In this study, we show that, thanks to the predictions generated, better profiling quality indicators can be outputted, and the performance of automatic profiling can be maximized. Our proposed workflow can learn and model the sample properties; therefore, restrictions in the biological matrix, or sample preparation protocol, and limitations of lineshape fitting approaches can be overcome.

RAW images and videos for 30K panoramic projection using the ACES workflow

According to our recent paper [1], the concept of creating a still image panorama with the additional inclusion of video footage up to 30K resolution has proven to be successful in various application examples. However, certain aspects of the production pipeline need some optimization, especially the color workflow and the spatial placement of the video content. This paper aims to compare two workflows to overcome these problems. In particular, the following two methods are described in detail: 1) Improving the current workflow with the Canon EOS D5 Mark IV camera as the central device, 2) Establishing a new workflow using the new possibilities of the Apple iPhone 12 Pro MAX. The following aspects are the subject of our investigation: a) The fundamental idea is to use the ACES as the central color management system. It is investigated if the direct import from RAW to ACEScg via dcraw and rawtoaces shows advantages. In addition, the conversion from Dolby Vision to ACES for the video processing is investigated, and the result is evaluated. Furthermore, the influence of stitching programs (e.g., PTGUI) on the color workflow is observed and optimized. b) The second part of the paper deals with the spatial integration of the videos into the still panoramas. Due to the different crop factors, specific focal lengths must be applied when using the Canon EOS D5 Mark IV; this distorts the image and video materials differently and makes it difficult to place the video footage in the panorama. We investigate if the usage of the lens distortion removal algorithm improves results. Furthermore, the comparison of the performance and capabilities of the Apple iPhone 12 Pro MAX is also evaluated regarding this aspect. Finally, the recorded resolution of detailed vegetation and foliage in video footage is compared. The paper summarizes the results of the new proposed workflow and indicates necessary further investigation. [1] Hasche, Eberhard; Benning, Dominik; Karaschewski, Oliver; Carstens, Florian; Creutzburg, Reiner: Creating high-resolution 360-degree single-line 25K video content for modern conference rooms using film compositing techniques. In: Electronic Imaging, Mobile Devices and Multimedia: Technologies, Algorithms &amp; Applications 2020, pp. 206-1-206-14(14), https://doi.org/10.2352/ISSN.2470-1173.2020.3.MOBMU-206

A method of predicting oil and gas resource spatial distribution based on Bayesian network and its application

Abstract The spatial distribution prediction of hydrocarbon resource is essential to reduce exploration risks and improve investment returns. Determining the location of the spatial distribution of oil and gas resources is a complex and uncertain problem. This paper systematically analyzes oil and gas exploration risk assessment and resource spatial distribution prediction technology, and proposes a method of predicting hydrocarbon spatial distribution based on averaged one-dependence estimators(AODE). Firstly, the transformation process of petroleum geological problem to mathematical model is described. Then, the classification principle and decision rule of AODE model are expounded. Finally, a case study of Sangonghe Formation in the hinterland of Junggar Basin in China is given to further illustrate the proposed method and work flow. The case prediction results not only reveal that the accuracy of the AODE model can reach 85.2% on the data set of 203 exploratory wells, which is higher than state-of-the-art methods (such as tree augmented Bayesian network method and the Mahalanobis distance method), but also point out the areas with high probability of remaining hydrocarbon resources in Sangonghe Formation, which provides decision-making basis for the next exploration. The application results show that the AODE model can effectively predict the spatial distribution of oil and gas and visualize the risk of geological exploration, thereby optimizing drilling strategy and increasing economic benefits.

Machine learning assisted history matching for a deepwater lobe system

High exploration costs resulting in sparse datasets and complicated geological structures in deepwater depositional systems make the reservoir characterization extremely difficult. To meet this challenge, rule-based geostatistical subsurface modeling has been developed to fill this data gap with enhanced integration of geological concepts for deepwater reservoirs based on geological rules constrained by this sparse well data. The rule-based model simulates sediment dynamics through depositional rules to calculate reservoir architecture and the associated rock properties distributions. As a result, rule-based models incorporate conceptual and qualitative information, such as temporal deposition sequence and consequent compensational stacking patterns. Integrating quantitative data, such as fluid production history, is a remaining obstacle to the broad application of rule-based subsurface models. We propose a new machine learning assisted history matching workflow for rule-based models. First, multiple rule-based models are calculated as training data for a Generative Adversarial Network (GAN). The successfully trained GAN enables exploration of the latent reservoir manifold with only a small number of numeric values (i.e., latent random vector); this is a massive reduction in the dimensionality of potential subsurface models. The initial ensemble models, generated by the trained GAN with a range of latent random vectors, are coupled with physics-based reservoir flow simulation to obtain production forecasts. Ensemble Kalman filter (EnKF) updates the latent random vectors of the ensemble by minimizing the misfit (i.e., error norm) between the production forecasts and the production observation over time. Our proposed workflow finds an ensemble of optimized reservoir models that honor realistic geological heterogeneity and production history. Besides, as the GAN's latent random vectors are Gaussian distributed, one of the fundamental mathematical assumptions of EnKF, this workflow effectively alleviates possible artifacts (e.g., filter divergence or overshooting) in EnKF. Moreover, this proposed workflow is more computationally efficient than updating the entire high-dimensional reservoir properties and removes the limitation to only Gaussian simulation models with their limited geological realism. The proposed workflow may be expanded to various reservoir depositional settings. • We propose a machine learning assisted history matching workflow for complex deepwater subsurface models. • We apply generative adversarial networks (GAN) to represent the complex subsurface models by latent random vectors. • We update latent random vectors of the subsurface model to match the production history by the ensemble Kalman filter (EnKF). • By combining GAN and EnKF, we can 1) enhance computational efficiency, 2) remove artifacts, and 3) preserve geologic realism.