Now showing items 1-20 of 22005

    • Past, Present, and Future of Software for Bayesian Inference

      Štrumbelj, Erik; Bouchard-Côté, Alexandre; Corander, Jukka; Gelman, Andrew; Rue, Haavard; Murray, Lawrence; Pesonen, Henri; Plummer, Martyn; Vehtari, Aki (Accepted by Statistical Science, 2023-09-19) [Article]
      Software tools for Bayesian inference have undergone rapid evolution in the past three decades, following popularisation of the first generation MCMC-sampler implementations. More recently, exponential growth in the number of users has been stimulated both by the active development of new packages by the machine learning community and popularity of specialist software for particular applications. This review aims to summarize the most popular software and provide a useful map for a reader to navigate the world of Bayesian computation. We anticipate a vigorous continued development of algorithms and corresponding software in multiple research fields, such as probabilistic programming, likelihood-free inference, and Bayesian neural networks, which will further broaden the possibilities for employing the Bayesian paradigm in exciting applications.
    • Relative Insignificance of Polyamide Layer Selectivity for Seawater Electrolysis Applications

      Zhou, Xuechen; Shi, Le; Taylor, Rachel; Xie, Chenghan; Bian, Bin; Picioreanu, Cristian; Logan, Bruce (Environmental Science & Technology, American Chemical Society (ACS), 2023-09-18) [Article]
      Low-cost polyamide thin-film composite (TFC) membranes are being explored as alternatives to cation exchange membranes for seawater electrolysis. An optimal membrane should have a low electrical resistance to minimize applied potentials needed for water electrolysis and be able to block chloride ions present in a seawater catholyte from reaching the anode. The largest energy loss associated with a TFC membrane was the Nernstian overpotential of 0.74 V (equivalent to 37 Ω cm<sup>2</sup> at 20 mA cm<sup>-2</sup>), derived from the pH difference between the anolyte and catholyte and not the membrane ohmic overpotential. Based on analysis using electrochemical impedance spectroscopy, the pristine TFC membrane contributed only 5.00 Ω cm<sup>2</sup> to the ohmic resistance. Removing the polyester support layer reduced the resistance by 79% to only 1.04 Ω cm<sup>2</sup>, without altering the salt ion transport between the electrolytes. Enlarging the pore size (∼5 times) in the polyamide active layer minimally impacted counterion transport across the membrane during electrolysis, but it increased the total concentration of chloride transported by 60%. Overall, this study suggests that TFC membranes with thinner but mechanically strong supporting layers and size-selective active layers should reduce energy consumption and the potential for chlorine generation for seawater electrolyzers.
    • Reliable Transmission of Short Packets in Cognitive Radio Inspired NOMA Network

      Xia, Chunli; Xiang, Zhongwu; Meng, Jin; Liu, Hongbo; Pan, Gaofeng (IEEE Systems Journal, Institute of Electrical and Electronics Engineers (IEEE), 2023-09-18) [Article]
      Aiming at the high-performance requirements of Internet of Things (IoT) scenarios, a cognitive radio-inspired nonorthogonal multiple access (CR-NOMA) scheme in short packet communication is investigated. Primary and secondary users share the same nonorthogonal communication resource block to achieve high spectral efficiency. A priority selection combination method is designed and employed by the primary user considering the nonnegligible decoding error rate in short packet communication. We focus on the transmission performance of the secondary users while ensuring the reliable communication of the primary user as a priority. To characterize the performance of the CR-NOMA network, approximate closed-form expressions of the average block error rate (BLER) for primary and secondary users are derived and analyzed. The results show that the communication reliability and throughput of the primary user with the CR-NOMA scheme are enhanced compared with the no-CR scheme. Meanwhile, the secondary users can also achieve reliable communication on the premise of improving the performance of the primary user through spectrum sharing. A performance tradeoff between primary and secondary users can be achieved through power allocation. Furthermore, there is a tradeoff between BLER and throughput, and the maximum throughput can be achieved by choosing an optimal blocklength.
    • Submarine optical fiber communication provides an unrealized deep-sea observation network

      Guo, Yujian; Marin, Juan M.; Ashry, Islam; Trichili, Abderrahmen; Havlik, Michelle-Nicole; Ng, Tien Khee; Duarte, Carlos M.; Ooi, Boon S. (Scientific Reports, Springer Science and Business Media LLC, 2023-09-18) [Article]
      Oceans are crucial to human survival, providing natural resources and most of the global oxygen supply, and are responsible for a large portion of worldwide economic development. Although it is widely considered a silent world, the sea is filled with natural sounds generated by marine life and geological processes. Man-made underwater sounds, such as active sonars, maritime traffic, and offshore oil and mineral exploration, have significantly affected underwater soundscapes and species. In this work, we report on a joint optical fiber-based communication and sensing technology aiming to reduce noise pollution in the sea while providing connectivity simultaneously with a variety of underwater applications. The designed multifunctional fiber-based system enables two-way data transfer, monitoring marine life and ship movement near the deployed fiber at the sea bottom and sensing temperature. The deployed fiber is equally harnessed to transfer energy that the internet of underwater things (IoUTs) devices can harvest. The reported approach significantly reduces the costs and effects of monitoring marine ecosystems while ensuring data transfer and ocean monitoring applications and providing continuous power for submerged IoUT devices.
    • Opportunistic Mobile Networks Content Delivery for Important but Non-Urgent Traffic

      Lau, Chun Pong; Ma, Guoqing; Susanto, Hengky; Dang, Shuping; Ng, Kam Shing; Shihada, Basem (IEEE Access, Institute of Electrical and Electronics Engineers (IEEE), 2023-09-18) [Article]
      As delay-tolerant and large-size content, for example, software updates, TV series, and virtual reality related content, become more prevalent in mobile networks, the need for efficient content delivery mechanisms becomes increasingly important,the traffic that carries these contents is not suitable to be evaluated using traditional network performance metrics, e.g., delay, throughput, and jitter. Based on this insight, we propose the solution of content dissemination from opportunistic mobile social communications (CODOMOC)which utilizes energy cost as an alternative performance metric and exploits daily human activity mobility pattern to determine how, when, and where the contents should be disseminated. Then, we introduce two options in CODOMOC to achieve different network operators’ objectives. The two options are the Only Dense (OD) option which aims at minimizing energy consumption for network operators and the Broadcast Efficiency (BE) option to further reduce the total carbon footprint of network operators. CODOMOC is evaluated by comparing with a mobility-based broadcast method. The results show that CODOMOC reduces the average energy consumption by 51% and 60% in the OD and BE options respectively. The proposed solution equipped with the two modes is expected to provide a higher degree of flexibility and reduce energy consumption for mobile networks, while, admittedly, the application scope of the solution and the associated methodologies proposed in this paper is restricted to important but non-urgent traffic delivery.
    • Comparative DFT Study of Small Anionic Silver and Copper Clusters: Evolution of Structure and Physicochemical Properties

      Matulis, Vitaly E.; Ivashkevich, Oleg A.; Lappo, Daniil D.; Lyakhov, Dmitry; Michels, Dominik L. (The Journal of Physical Chemistry C, American Chemical Society (ACS), 2023-09-18) [Article]
      Based on both total energy calculations and comparison of experimental and calculated characteristics of the photoelectron spectrum (PHES), the structural assignment of clusters Agn– (n = 13–16) and Cum– (m = 14–17) has been made using the density functional theory (DFT) model with our previously developed S2LYP functional. A comparative study of size dependence of geometry, electronic structure, and physicochemical properties has been carried out for a series of anionic silver and copper clusters containing up to 20 atoms. For the cases when two isomers contribute to the experimental PHES, the isomerization barriers and molar ratio of isomers were estimated. It has been shown that the geometry and the properties that are determined mainly by ns-derived electronic states are similar for copper and silver clusters. However, due to the larger contribution of (n–1)d-electrons to the chemical bond, the potential energy surface of copper clusters is less smooth, and these clusters are characterized by higher isomerization energies compared to silver clusters. The isomerization energies of clusters and the number of isomers with similar energies increase with enlarging cluster size. Thus, clusters containing less than 20 atoms easily overcome the barriers of intramolecular isomerization (i.e., behave like liquids). However, it is expected that cooled clusters containing several tens of atoms will have a rigid geometry due to high intramolecular isomerization energies.
    • Coupled fluid flow, solute transport and dissolution processes in discrete fracture networks: an advanced Discontinuous Galerkin model

      Tabrizinejadas, Sara; Younes, Anis; Hoteit, Hussein; Carrayrou, Jerome; Fahs, Marwan (Advances in Water Resources, Elsevier BV, 2023-09-17) [Article]
      Modeling dissolution processes in discrete fracture networks (DFNs) is a challenging task. Challenges are related to the highly nonlinear coupling between flow, mass transport, and reactive processes associated with fracture aperture evolution by dissolution. Further, advection-dominated transport due to fast fluid flow in fractures renders the problem more complex from a computational point of view, as traditional numerical methods may introduce unphysical oscillations or excessive numerical diffusion. The Discontinuous Galerkin (DG) method is known to be suitable for the simulation of advection-dominated transport. In this work, an advanced DG model is developed to model transport with dissolution in DFNs. We propose an upwind formulation to deal with the upstream concentration at the intersection of several fractures. The upstream concentration at an intersection node is calculated based on the average nodal concentrations of all the fractures having an inflow at that node, weighted by the volumetric fluxes of these fractures. The dispersion term is discretized with the Mixed Finite Element (MFE) method, which ensures the continuity of the dispersive flux at the intersection of fractures with different apertures. The obtained nonlinear coupled flow-transport-dissolution equations are discretized in time with a high-order scheme via the method of lines (MOL). Numerical examples and comparisons with standard finite element (FE) and finite volume (FV) solutions are performed to investigate the correctness and efficiency of the developed model. Results show that the new DG-DFN model avoids unphysical oscillations encountered with the standard FE method and strongly reduces the numerical diffusion observed with the upwind FV scheme. The DG-DFN model is then used to investigate the effect of the dissolution rate on the flow, transport, and aperture evolution processes for a single fracture and for a DFN. A quasi-linear evolution of the fracture aperture is observed for low dissolution rates. For high dissolution rates, a funnel-shaped enlargement is observed with a significant widening for the fractures near the inlet and minor effects for those away from the injection location.
    • A marginalized two-part joint model for a longitudinal biomarker and a terminal event with application to advanced head and neck cancers

      Rustand, Denis; Briollais, Laurent; Rondeau, Virginie (Pharmaceutical Statistics, Wiley, 2023-09-17) [Article]
      The sum of the longest diameter (SLD) of the target lesions is a longitudinal biomarker used to assess tumor response in cancer clinical trials, which can inform about early treatment effect. This biomarker is semicontinuous, often characterized by an excess of zeros and right skewness. Conditional two-part joint models were introduced to account for the excess of zeros in the longitudinal biomarker distribution and link it to a time-to-event outcome. A limitation of the conditional two-part model is that it only provides an effect of covariates, such as treatment, on the conditional mean of positive biomarker values, and not an overall effect on the biomarker, which is often of clinical relevance. As an alternative, we propose in this article, a marginalized two-part joint model (M-TPJM) for the repeated measurements of the SLD and a terminal event, where the covariates affect the overall mean of the biomarker. Our simulation studies assessed the good performance of the marginalized model in terms of estimation and coverage rates. Our application of the M-TPJM to a randomized clinical trial of advanced head and neck cancer shows that the combination of panitumumab in addition with chemotherapy increases the odds of observing a disappearance of all target lesions compared to chemotherapy alone, leading to a possible indirect effect of the combined treatment on time to death.
    • Data Center-Enabled High Altitude Platforms: A Green Computing Alternative

      Abderrahim, Wiem; Amin, Osama; Shihada, Basem (Accepted by IEEE Transactions on Mobile Computing (TMC), 2023-09-17) [Article]
      Information technology organizations and companies are seeking greener alternatives to traditional terrestrial data centers to mitigate global warming and reduce carbon emissions. Currently, terrestrial data centers consume a significant amount of energy, estimated at about 1.5% of worldwide electricity use. Furthermore, the increasing demand for data-intensive applications is expected to raise energy consumption, making it crucial to consider sustainable computing paradigms. In this study, we propose a data center-enabled High Altitude Platform (HAP) system, where a flying data center supports the operation of terrestrial data centers. We conduct a detailed analytical study to assess the energy benefits and communication requirements of this approach. Our findings demonstrate that a data center-enabled HAP is more energy-efficient than a traditional terrestrial data center, owing to the naturally low temperature in the stratosphere and the ability to harvest solar energy. Adopting a data center-HAP can save up to 14% of energy requirements while overcoming the offloading outage problem and the associated delay resulting from server distribution. Our study highlights the potential of a data centerenabled HAP system as a sustainable computing solution to meet the growing energy demands and reduce carbon footprint.
    • Effects of multi-observations uncertainty and models similarity on climate change projections

      Pathak, Raju; Dasari, Hari Prasad; Ashok, Karumuri; Hoteit, Ibrahim (npj Climate and Atmospheric Science, Springer Science and Business Media LLC, 2023-09-16) [Article]
      Climate change projections (CCPs) are based on the multimodel means of individual climate model simulations that are assumed to be independent. However, model similarity leads to projections biased toward the largest set of similar models and intermodel uncertainty underestimation. We assessed the influences of similarities in CMIP6 through CMIP3 CCPs. We ascertained model similarity from shared physics/dynamics and initial conditions by comparing simulated spatial temperature and precipitation with the corresponding observed patterns and accounting for intermodel spread relative to the observational uncertainty, which is also critical. After accounting for similarity, the information from 57 CMIP6, 47 CMIP5, and 24 CMIP3 models can be explained by just 11 independent models without significant differences in globally averaged climate change statistics. On average, independent models indicate a lower global-mean temperature rise of 0.25 °C (~0.5 °C–1 °C in some regions) relative to all models by the end of the 21st century under CMIP6’s highest emission scenario.
    • Printed Electrodes Based on Vanadium Dioxide and Gold Nanoparticles for Asymmetric Supercapacitors

      Minyawi, Bashaer A.; Vaseem, Mohammad; Alhebshi, Nuha; AlAmri, Amal; Shamim, Atif (Nanomaterials, MDPI AG, 2023-09-16) [Article]
      Printed energy storage components attracted attention for being incorporated into bendable electronics. In this research, a homogeneous and stable ink based on vanadium dioxide (VO2) is hydrothermally synthesized with a non-toxic solvent. The structural and morphological properties of the synthesized material are determined to be well-crystalline monoclinic-phase nanoparticles. The charge storage mechanisms and evaluations are specified for VO2 electrodes, gold (Au) electrodes, and VO2/Au electrodes using cyclic voltammetry, galvanostatic charge–discharge, and electrochemical impedance spectroscopy. The VO2 electrode shows an electrical double layer and a redox reaction in the positive and negative voltage ranges with a slightly higher areal capacitance of 9 mF/cm2. The VO2/Au electrode exhibits an areal capacitance of 16 mF cm−2, which is double that of the VO2 electrode. Due to the excellent electrical conductivity of gold, the areal capacitance 18 mF cm−2 of the Au electrode is the highest among them. Based on that, Au positive electrodes and VO2 negative electrodes are used to build an asymmetric supercapacitor. The device delivers an areal energy density of 0.45 μWh cm−2 at an areal power density of 70 μW cm−2 at 1.4 V in the aqueous electrolyte of potassium hydroxide. We provide a promising electrode candidate for cost-effective, lightweight, environmentally friendly printed supercapacitors.
    • Methanol oxy-combustion and supercritical water oxidation: A ReaxFF molecular dynamics study

      Monge Palacios, Manuel; Grajales Gonzalez, Edwing; Sarathy, Mani (Energy, Elsevier BV, 2023-09-15) [Article]
      Energy and environmental concerns are motivating the use of renewable fuels such as methanol. Furthermore, the implementation of the oxy-combustion and hydrothermal combustion technologies can help to improve the performance of power generation and reduce NOx emissions. These aspects can contribute to achieve the transition to cleaner sources of energy that is being sought worldwide, and thus we carried out the first molecular dynamics study of the oxidation of methanol at 2700 K and 3000 K in four supercritical environments with compositions CH3OH + O2, CH3OH + O2+CO2, CH3OH + O2+H2O, and CH3OH + O2+CO2+H2O. Reaction mechanisms were obtained and revealed that the initiation reaction is CH3OH unimolecular dissociation in all cases. The CH3OH oxidation chemistry changes when O2 is replaced by supercritical CO2 (sCO2) and/or H2O (sH2O), and a new route for the important oxidation sequence CH3OH→CH2OH→H2CO→CHO→CO→CO2 is reported. The rate constants for the CH3OH unimolecular dissociation were calculated, indicating a positive effect of sH2O. Furthermore, the collisions of CH3OH molecules with those of H2O and CO2 were analyzed with molecular dynamics simulations and quantum chemistry calculations, suggesting that collisions with H2O can activate more efficiently CH3OH for a prospective dissociation event. This study is aimed to help in the development of kinetic models for CH3OH oxidation/pyrolysis in sCO2 and sH2O, and thus in the implementation of the oxy-combustion and hydrothermal combustion techniques for this alternative fuel.
    • Uncertainty quantification in coastal aquifers using the multilevel Monte Carlo method

      Litvinenko, Alexander; Logashenko, Dmitry; Tempone, Raul; Vasilyeva, Ekaterina; Wittum, Gabriel (PAMM, Wiley, 2023-09-15) [Article]
      We are solving a problem of salinisation of coastal aquifers. As a test case example, we consider the Henry saltwater intrusion problem. Since porosity, permeability and recharge are unknown or only known at a few points, we model them using random fields and random variables. The Henry problem describes a two-phase flow and is non-linear and time-dependent. The solution to be found is the expectation of the salt mass fraction, which is uncertain and time-dependent. To estimate this expectation, we use the well-known multilevel Monte Carlo (MLMC) method. The MLMC method takes just a few samples on computationally expensive (fine) meshes and more samples on cheap (coarse) meshes. Then, by building a telescoping sum, the MLMC method estimates the expected value at a much lower computational cost than the classical Monte Carlo method. The deterministic solver used here is the well-known parallel and scalable ug4 solver.
    • FLT3 inhibitors and novel therapeutic strategies to reverse AML resistance: an updated comprehensive review.

      Abdel-Aziz, Amal Kamal; Dokla, Eman M E; Saadeldin, Mona Kamal (Critical reviews in oncology/hematology, Elsevier BV, 2023-09-15) [Article]
      FMS-like tyrosine kinase 3 (FLT3) mutations occur in almost 30% of acute myeloid leukemia (AML) patients. Despite the initial clinical efficacy of FLT3 inhibitors, many treated AML patients with mutated FLT3 eventually relapse. This review critically discusses the potential and challenges of FLT3-targeted therapies and sheds light on their drug interactions as well as potential biomarkers. Furthermore, we focus on the molecular mechanisms underlying the resistance of FLT3 internal tandem duplication (FLT3-ITD) AMLs to FLT3 inhibitors alongside novel therapeutic strategies to reverse resistance. Notably, dynamic heterogeneous patterns of clonal selection and evolution contribute to the resistance of FLT3-ITD AMLs to FLT3 inhibitors. Ongoing preclinical research and clinical trials are actively directed towards devising rational "personalized" or "patient-tailored" combinatorial therapeutic regimens to effectively treat patients with FLT3 mutated AML.
    • Redox-couple-assisted CO2 capture on solid-electrolyte reactor

      Chang, Bin; Feng, Chengyang; Garcia-Melchor, Max; Zhang, Huabin (Chem, Elsevier BV, 2023-09-15) [Article]
      Electrochemical carbon capture offers a promising approach for capturing scarce CO2 from industrial emissions or the atmosphere. Challenges persist in current techniques, such as low capture rates and oxygen sensitivity. This article previews the latest findings by Wang and co-workers published in Nature, reporting the oxygen/water redox couple in a solid-electrolyte reactor for continuous and modular CO2 capture.
    • Microalgae Biomass Harvesting Using Chitosan Flocculant: Optimization of Operating Parameters by Response Surface Methodology

      Elcik, Harun; Karadag, Dogan; Kara, Ayse Irem; Cakmakci, Mehmet (Separations, MDPI AG, 2023-09-15) [Article]
      Bioflocculants can be used for cost-effective harvesting of microalgae biomass on an industrial scale. This study investigates the flocculation-based harvesting approach to recovering Chlorella vulgaris microalgae biomass using chitosan biopolymer. Response surface methodology (RSM) was used to design the experiments and optimize the critical operating parameters. Box-Behnken Design (BBD) was employed at three levels, and 17 experimental runs were conducted to determine the optimal conditions and the relationship between operating parameters. The highest biomass recovery of 99.10% was achieved at the following optimized conditions: pH of 5, flocculation time of 45 min, and chitosan concentration of 10 mg/L. Both experimental results and model outputs indicated that pH significantly impacts microalgae harvesting and that process performance is less dependent on chitosan concentration and flocculation time. The quadratic model has shown the best fit with the experimental results. The results could be applied to large-scale microalgae harvesting applications to promote microalgae biomass recovery and reduce operating costs.
    • Sands Subjected to Repetitive Loading Cycles and Associated Granular Degradation

      Park, Junghee; Santamarina, Carlos (Journal of Geotechnical and Geoenvironmental Engineering, American Society of Civil Engineers (ASCE), 2023-09-14) [Article]
      This study examines the load-deformation response of sands subjected to high- and low-stress cycles, i.e., both ends of the Wöhler’s fatigue curve. At high peak cyclic stress σf, the terminal void ratio decreases with σf due to crushing-dependent densification, and it can be smaller than emin when the peak stress approaches the yield stress σf→σy. When σf≪σy, the soil retains memory of the initial fabric even after a very large number of cycles, and the terminal void ratio correlates with the initial void ratio eo. Data show that the maximum change in relative density leads to simple strategies to estimate the maximum settlement for first-order engineering analyses. In agreement with Wöhler’s fatigue, tipping points in void ratio and stiffness trends occur at a small number of high-stress cycles or after a large number of small-stress cycles. During repetitive loading, sands stiffen with the number of cycles to reflect increased interparticle coordination following crushing, as well as contact flattening due to asperity breakage and fretting. The strong correlation between the resilient modulus Mr and the maximum shear modulus Gmax suggests the possible application of geophysical methods based on shear wave propagation to monitor geosystems subjected to repetitive loading cycles.
    • Multifunctional difluoroboron β-diketonate-based luminescent receiver for a high-speed underwater wireless optical communication system

      Wang, Yue; Wang, Jian-Xin; Alkhazragi, Omar; Gutierrez Arzaluz, Luis; Zhang, Huafan; Kang, Chun Hong; Ng, Tien Khee; Bakr, Osman; Mohammed, Omar F.; Ooi, Boon S. (Optics Express, Optica Publishing Group, 2023-09-14) [Article]
      The last decade has witnessed considerable progress in underwater wireless optical communication in complex environments, particularly in exploring the deep sea. However, it is difficult to maintain a precise point-to-point reception at all times due to severe turbulence in actual situations. To facilitate efficient data transmission, the color-conversion technique offers a paradigm shift in large-area and omnidirectional light detection, which can effectively alleviate the étendue limit by decoupling the field of view and optical gain. In this work, we investigated a series of difluoroboron β-diketonate fluorophores by measuring their photophysical properties and optical wireless communication performances. The emission colors were tuned from blue to green, and >0.5 Gb/s data transmission was achieved with individual color channel in free space by implementing an orthogonal frequency-division multiplexing (OFDM) modulation scheme. In the underwater experiment, the fluorophore with the highest transmission speed was fabricated into a 4×4 cm2 luminescent concentrator, with the concentrated emission from the edges coupled with an optical fiber array, for large-area photodetection and optical beam tracking. The net data rates of 130 Mb/s and 217 Mb/s were achieved based on nonreturn- to-zero on-off keying and OFDM modulation schemes, respectively. Further, the same device was used to demonstrate the linear light beam tracking function with high accuracy, which is beneficial for sustaining a reliable and stable connection in a dynamic, turbulent underwater environment.
    • TinyML Models for a Low-cost Air Quality Monitoring Device

      Wardana, I Nyoman Kusuma; Fahmy, Suhaib A.; Gardner, Julian W. (IEEE Sensors Letters, Institute of Electrical and Electronics Engineers (IEEE), 2023-09-14) [Article]
      Low-cost air quality monitoring devices can provide high-density spatiotemporal pollution data, thus offering a better opportunity to apply machine learning. Low-cost sensor nodes usually utilize microcontrollers as the main processors, and tinyML brings machine learning (ML) models to these resource-constrained devices. In this letter, we reported the development of a low-cost air quality monitoring device with embedded tinyML models. We deployed two tinyML models on a single microcontroller and performed two tasks: predicting air quality and power parameters (using model predictor) and imputing missing features (using model imputer). The proposed model predictor can estimate parameters with a coefficient of determination above 0.70, and the model imputer effectively estimates the testing data when missing rates are below 80%. By performing the post-training quantization technique, we can further reduce the model size but slightly degrade the accuracies.
    • Rapid Inference of Reservoir Permeability from Inversion of Traveltime Data Under a Fast Marching Method-Based Deep Learning Framework

      Li, Chen; Yan, Bicheng; Kou, Rui; Gao, Sunhua (SPE Journal, Society of Petroleum Engineers (SPE), 2023-09-14) [Article]
      The fast marching method (FMM) is a highly efficient numerical algorithm used to solve the Eikonal equation. It calculates traveltime from the source point to different spatial locations and provides a geometric description of the advancing front in anisotropic and heterogeneous media. As the Eikonal solution, the diffusive time of flight (DTOF) can be used to formulate an asymptotic approximation to the pressure diffusivity equation to describe transient flow behavior in subsurface porous media. For the infinite-acting flow that occurs in porous media with smoothly varying heterogeneity, traveltime of the pressure front from the active production or injection well to the observation well can be directly estimated from the DTOF using the concept of radius (or depth) of investigation (ROI or DOI), which is defined as the moment when a maximum magnitude of the partial derivative of pressure to time occurs. Based on the ROI or DOI definition, we propose a deep neural network called the inversion neural network (INN) to inversely estimate heterogeneous reservoir permeability by inverting the traveltime data. The INN is trained by traveltime data created for a large data set of distinct permeability fields from FMM simulations, which can be two orders of magnitude faster than conventional reservoir simulators. A convolutional neural network (CNN), the U-Net architecture, is incorporated into the INN, which establishes a nonlinear mapping between the heterogeneous permeability fields and the traveltime data collected at sparse observation wells. The loss function used for the INN is defined as the root mean square error (RMSE) between the logarithm of the predicted permeability and the logarithm of the true permeability. The performance of the INN is tested on reservoir models with both smoothly varying heterogeneity and high-contrast media properties. For the 2D smoothly varying heterogeneous models with a grid size of 49×49, the permeability predicted by the INN has an average estimation error of 8.73% when a set of 7×7 uniformly distributed observation wells is used to collect “observational” traveltime data from the FMM simulation. For models with the same grid size and observation well density but with high-contrast media properties, the INN can still capture the general heterogeneity distribution, although with reduced prediction accuracy. Using a graphics processing unit (GPU) for training and prediction allows the entire inverse modeling process for a 2D 49×49 reservoir model to be completed within 7 minutes.