Fine-scale Simulations Research Articles

Assisted Upscaling of Miscible CO2-Enhanced Oil Recovery Floods Using an Artificial Neural Network-Based Optimisation Algorithm

Due to the high computing cost of the fine-scale compositional simulations needed to effectively model miscible CO2 flooding, upscaling techniques are needed to approximate the behaviour of these fine-scale grids on more realistic coarse-scale models. The use of transport coefficients to better represent small-scale interactions, such as the time-dependent flux of the components within the hydrocarbon phases (molecular diffusion), and the pseudoisation of relative permeabilities to ensure the matching of large-scale effects, such as the volumetric fluxes of the phases, are two of these procedures. Most times, a mismatch between the phase fluxes of the integrated fine-scale and that of the coarse-scale is observed. By adjusting or calibrating some of the generated coarse-scale pseudo functions, such as the transport coefficients, absolute permeability, or relative permeability endpoints, the accuracy of the upscaling results can be improved. This procedure can be treated a reservoir history matching problem which is typically computationally expensive. In this study, we provide a framework for representing the dynamics of small-scale molecular diffusion and macro-scale heterogeneity-induced channelling related to miscible CO2 displacements on upscaled coarser grid reservoir models. The method used was based on the pseudoisation of relative permeability and transport coefficients and was applied to two benchmark reservoir models from the Society of Petroleum Engineers (SPE). Our results demonstrated that using effectively calibrated transport coefficients improved the upscaling results, so that the calculated pseudo-relative permeability functions can be ignored. We proposed a unique approach to upscaling miscible floods that utilised a genetic algorithm and a neural-network-based proxy model to minimise the associated computing cost. The data-driven approximation model considerably decreased the computing cost associated with the assisted tuning technique, and the optimisation algorithm was used to reduce the error between the predictions of the upscaled models. In conclusion, the methodology described in this study effectively captured the small- and large-scale behaviour related to the miscible displacements on upscaled coarse-scale reservoir models while reduced associated computational costs.

Machine-learning of nonlocal kernels for anomalous subsurface transport from breakthrough curves

<p style='text-indent:20px;'>Anomalous behavior is ubiquitous in subsurface solute transport due to the presence of high degrees of heterogeneity at different scales in the media. Although fractional models have been extensively used to describe the anomalous transport in various subsurface applications, their application is hindered by computational challenges. Simpler nonlocal models characterized by integrable kernels and finite interaction length represent a computationally feasible alternative to fractional models; yet, the informed choice of their kernel functions still remains an open problem. We propose a general data-driven framework for the discovery of optimal kernels on the basis of very small and sparse data sets in the context of anomalous subsurface transport. Using spatially sparse breakthrough curves recovered from fine-scale particle-density simulations, we learn the best coarse-scale nonlocal model using a nonlocal operator regression technique. Predictions of the breakthrough curves obtained using the optimal nonlocal model show good agreement with fine-scale simulation results even at locations and time intervals different from the ones used to train the kernel, confirming the excellent generalization properties of the proposed algorithm. A comparison with trained classical models and with black-box deep neural networks confirms the superiority of the predictive capability of the proposed model.</p>

Fast macroscopic forcing method

The macroscopic forcing method (MFM) of Mani and Park [1] and similar methods for obtaining turbulence closure operators, such as the Green's function-based approach of Hamba [2], recover reduced solution operators from repeated direct numerical simulations (DNS). MFM has already been used to successfully quantify Reynolds-averaged Navier–Stokes (RANS)-like operators for homogeneous isotropic turbulence and turbulent channel flows. Standard algorithms for MFM force each coarse-scale degree of freedom (i.e., degree of freedom in the RANS space) and conduct a corresponding fine-scale simulation (i.e., DNS), which is expensive. We combine this method with an approach recently proposed by Schäfer and Owhadi [3] to recover elliptic integral operators from a polylogarithmic number of matrix–vector products. The resulting Fast MFM introduced in this work applies sparse reconstruction to expose local features in the closure operator and reconstructs this coarse-grained differential operator in only a few matrix–vector products and correspondingly, a few MFM simulations. For flows with significant nonlocality, the algorithm first “peels” long-range effects with dense matrix–vector products to expose a more local operator. We demonstrate the algorithm's performance for the eddy diffusivity and eddy viscosity operators, which correspond to the unclosed parts of the ensemble-averaged transport equations, excluding the analytically known, closed parts of such equations. However, the algorithm can also be applied to the full operators. We focus on scalar transport in a laminar channel flow and momentum transport in a turbulent channel flow. For these problems, we recover eddy diffusivity- and eddy viscosity-like operators, respectively, at 1% of the cost of computing the exact operator via a brute-force approach for the laminar channel flow problem and 13% for the turbulent one. Applying these operators to compute the averaged fields of interest has visually indistinguishable behavior from the exact solution. Our results show that a similar number of simulations are required to reconstruct the operators to the same accuracy under grid refinement. Thus, the accuracy corresponds to the physics of the problem, not the numerics, so long as the grid is sufficiently refined. We glean that for problems in which the RANS space is reducible to one dimension, eddy diffusivity, and eddy viscosity operators can be reconstructed with reasonable accuracy using only a few simulations, regardless of simulation resolution or degrees of freedom.

Multiscale Topology Optimization of modulated fluid microchannels based on asymptotic homogenization

Dehomogenization techniques are becoming increasingly popular for enhancing lattice designs of compliant mechanical systems with ultra-large resolutions. Their effectiveness hinges on computing a deformed periodic grid that enable to reconstruct fine-scale designs with modulated and oriented patterns. In this paper, we propose an approach for extending dehomogenization methods to laminar fluid systems. We initiate our methodology by asymptotically deriving Darcy’s law on a periodically porous medium deformed by a diffeomorphism. Unlike the mechanical context, we reveal that the homogenized permeability matrix depends not solely on local the orientation but also on the local dilation of the deformed periodic medium. This distinction presents one of the several challenges to be tackled when adapting dehomogenization-based topology optimization techniques to porous media. To accommodate existing methodologies, we formulate a simplified “poor man’s” homogenized model, which streamlines various aspects, yet still leans on periodic cell problems to estimate the spatially varying permeability matrix. Specifically, we overlook boundary layer effects, we presume periodic grid deformations, and we neglect local dilation, solely considering the relationship with local cell orientations. Subsequently, we present a numerical approach for designing a system that redistributes an input flow across numerous regularly spaced outlets at an output interface. Leveraging the homogenized model, we deduce optimized geometric arrangements of local channel spacing parameters and orientations. We then use established methods to reconstruct grid deformations and fine-scale designs. The fidelity of these reconstructions is then validated through fine-scale simulations. Our observations indicate that while the proposed designs yield satisfactory performance when subjected to the full-scale model, discernible deviations from the homogenized model persist, appealing to future improvements.

Bioreaction coupled flow simulations: Impacts of methanogenesis on seasonal underground hydrogen storage

Assessing microbial risks is key to feasible hydrogen storage in geological formations. This work quantitatively analyses the impacts of bio-methanation on hydrogen storage performance. Fine-scale flow simulations, coupled with the bio-methanation reaction, are presented to analyse its impact on the storage performance. Based on the reported rates in literature, methanogenesis may slightly degrade the recovery performance of hydrogen but is considered minor compared with the issue of gas mixing. The impacts of methanogenesis on a time scale of months (330 days) becomes observable in the system configured here, when the methanation rate is above 1746 nano molality per hour. The assumed methanation rate is two times greater than the rate reported from the Olla filed. Validated scaling theory generalises findings for gravity-dominated scenarios. But viscous-dominated flows see complications from property variations due to pressure changes at high rates. This study provides definitions of “target properties” (e.g., acceptable methanogenesis rates) for screening hydrogen storage projects.

Fine scale simulation of sea salt aerosol under strong wind: WRF sensitivity test on PBL schemes and LES under Typhoon Hagibis

Identifying which factors among the planetary boundary layer (PBL) scheme, grid resolution, and source or sink terms dominantly influence the sea salt aerosol (SSA) spatial distribution is necessary. The dependence of SSA transport on PBL processes was investigated using a numerical weather prediction (NWP) model with a grid scale of ∼143 m. Size-segregated SSA transport during Typhoon Hagibis, which made landfall in Japan in 2019, was simulated using the Weather Research and Forecasting (WRF) model. The PBL schemes of Yonsei University (YSU), scale-aware YSU (Shin and Hong, 2015's scheme, SH) and Mellor-Yamada Nakanishi-Niino (MYNN), and large-eddy simulation (LES) subjected to turbulence transition enhancement via a cell perturbation method were tested to evaluate the reproducibility of observed values of wind speed, surface pressure, relative humidity, and temperature at meteorology masts. The results showed that wind speed was most influenced by the selection of the PBL scheme and LES. The distance from the coastline, as determined by the wind direction during the most intense typhoon stage, was correlated with a decrease in the time averaged SSA concentration when the storm moved inland. Despite the simulated SSA emission scheme's high sensitivity to wind speed (nearly 3rd power), the near-surface average concentration was less sensitive to the PBL schemes and LES selection than wind speed. However, the maximum arrival of the SSA differed by ∼20 % during the most vigorous stage of the typhoon. Budget analysis of the SSA transport equation highlighted that the advection term and gravity settling were most dependent on the PBL schemes and LES. The impacts of both factors on SSA transport were not limited to near-surface tendencies but were in a broader range over the atmosphere, up to 1–2 km. These results indicated that the vertical structure of the SSA field over the boundary layer depended on the PBL schemes and LES, thereby modulating the inland arrival of the SSA.

Development of the Senseiver for efficient field reconstruction from sparse observations

AbstractThe reconstruction of complex time-evolving fields from sensor observations is a grand challenge. Frequently, sensors have extremely sparse coverage and low-resource computing capacity for measuring highly nonlinear phenomena. While numerical simulations can model some of these phenomena using partial differential equations, the reconstruction problem is ill-posed. Data-driven-strategies provide crucial disambiguation, but these suffer in cases with small amounts of data, and struggle to handle large domains. Here we present the Senseiver, an attention-based framework that excels in reconstructing complex spatial fields from few observations with low overhead. The Senseiver reconstructs n-dimensional fields by encoding arbitrarily sized sparse sets of inputs into a latent space using cross-attention, producing uniform-sized outputs regardless of the number of observations. This allows efficient inference by decoding only a sparse set of output observations, while a dense set of observations is needed to train. This framework enables training of data with complex boundary conditions and extremely large fine-scale simulations. We build on the Perceiver IO by enabling training models with fewer parameters, which facilitates field deployment, and a training framework that allows a flexible number of sensors as input, which is critical for real-world applications. We show that the Senseiver advances the state-of-the-art of field reconstruction in many applications.

Discrete element models for elasto-plastic wave propagation in nonlinear architected materials

Nonlinear architected materials are known to exhibit a plethora of dynamic phenomena that enhance our capacity to manipulate elastic waves. Since these properties stem from complex, subwavelength geometry, dynamic simulations at high resolutions are often intractable at scales of interest. Therefore, prior studies have turned to effective medium models that capture essential properties in the long-wavelength limit. One class of such models is a periodic structure composed of discrete mass, spring, and damper elements, whose constitutive relationships are computed to match behaviors observed in experiments or fine-scale simulations of representative volume elements. While models of this type have been implemented successfully in many cases, the majority of literature has been focused on recoverable deformation, possibly including linear damping. However, history-dependent effects, such as friction and plasticity, have been much less explored. In this work, we develop a discrete element modeling framework for nonlinear architected materials undergoing plastic deformation. Due to the history- and rate-dependent characteristics of plasticity, the framework generally yields a system of differential-algebraic equations whose computational cost is significantly greater than an elastic system of comparable size. We demonstrate the method using several examples from the literature and explore means to obtain phenomenological plasticity models for general geometries.

A cross-scale modeling framework for simulating typhoon-induced compound floods and assessing the emergency response in urban regions

The cities within the Pearl River Delta (PRD), known as the Greater Bay Area (GBA), are flood-prone areas that faced frequent threats from typhoon-induced storm surges and rainfall in recent decades. To assess the potential consequences of a typical typhoon event, fine-scale simulations of inundation induced by both pluvial and fluvial flooding (known as compound flooding) are crucial for policymakers and concerned stakeholders to develop proper measures. In this study, we develop a cross-scale framework for modeling city inundation arising from typhoon-induced surges and torrential rainfall–runoff by coupling a three-dimensional ocean model (FVCOM) and a one-dimensional and two-dimensional coupling river analysis system (HEC-RAS). The ocean model is applied to capture the mesoscale behavior of surge and tide propagation from the deep sea to shallow offshore areas near the PRD. The river analysis system can simulate storm wave propagation along rivers, overtopping banks on floodplains and rainfall–runoff-induced flooding in urban regions at gradually refined scales. Second, the framework integrates GIS components (network analysis) with these hydraulic models. Through this framework, the urban emergency service coverage can be assessed under various potential flood scenarios during typhoon landfalls. Super Typhoon Mangkhut in 2018, the most devastating typhoon in the last decade in the PRD, is used as a case study in simulating and studying the flooding process. The simulated results show that in this typhoon event, pluvial and fluvial flooding play different roles in hindering the accessibility of emergency resources. Moreover, the interaction between pluvial and fluvial flooding significantly affects some urban regions.

Predictive scale-bridging simulations through active learning

Throughout computational science, there is a growing need to utilize the continual improvements in raw computational horsepower to achieve greater physical fidelity through scale-bridging over brute-force increases in the number of mesh elements. For instance, quantitative predictions of transport in nanoporous media, critical to hydrocarbon extraction from tight shale formations, are impossible without accounting for molecular-level interactions. Similarly, inertial confinement fusion simulations rely on numerical diffusion to simulate molecular effects such as non-local transport and mixing without truly accounting for molecular interactions. With these two disparate applications in mind, we develop a novel capability which uses an active learning approach to optimize the use of local fine-scale simulations for informing coarse-scale hydrodynamics. Our approach addresses three challenges: forecasting continuum coarse-scale trajectory to speculatively execute new fine-scale molecular dynamics calculations, dynamically updating coarse-scale from fine-scale calculations, and quantifying uncertainty in neural network models.

Non-intrusive hybrid scheme for multiscale heat transfer: Thermal runaway in a battery pack

Accurate analytical and numerical modeling of multiscale systems is a daunting task. The need to properly resolve spatial and temporal scales spanning multiple orders of magnitude pushes the limits of both our theoretical models as well as our computational capabilities. Rigorous upscaling techniques enable efficient computation while bounding/tracking errors and helping to make informed cost-accuracy tradeoffs. The biggest challenges arise when the applicability conditions of upscaled models break down. Here, we present a non-intrusive two-way (iterative bottom-up top-down) coupled hybrid model, applied to thermal runaway in battery packs, that combines fine-scale and upscaled equations in the same numerical simulation to achieve predictive accuracy while limiting computational costs. First, we develop two methods with different orders of accuracy to enforce continuity at the coupling boundary. Then, we derive weak (i.e., variational) formulations of the fine-scale and upscaled governing equations for finite element (FE) discretization and numerical implementation in FEniCS. We demonstrate that hybrid simulations can accurately predict the average temperature fields within error bounds determined a priori by homogenization theory. Finally, we demonstrate the computational efficiency of the hybrid algorithm against fine-scale simulations.

Scale variance in the carbon dynamics of fragmented, mixed-use landscapes estimated using model–data fusion

Abstract. Many terrestrial landscapes are heterogeneous. Mixed land cover and land use generate a complex mosaic of fragmented ecosystems at fine spatial resolutions with contrasting ecosystem stocks, traits, and processes, each differently sensitive to environmental and human factors. Representing spatial complexity within terrestrial ecosystem models is a key challenge for understanding regional carbon dynamics, their sensitivity to environmental gradients, and their resilience in the face of climate change. Heterogeneity underpins this challenge due to the trade-off between the fidelity of ecosystem representation within modelling frameworks and the computational capacity required for fine-scale model calibration and simulation. We directly address this challenge by quantifying the sensitivity of simulated carbon fluxes in a mixed-use landscape in the UK to the spatial resolution of the model analysis. We test two different approaches for combining Earth observation (EO) data into the CARDAMOM model–data fusion (MDF) framework, assimilating time series of satellite-based EO-derived estimates of ecosystem leaf area and biomass stocks to constrain estimates of model parameters and their uncertainty for an intermediate complexity model of the terrestrial C cycle. In the first approach, ecosystems are calibrated and simulated at pixel level, representing a “community average” of the encompassed land cover and management. This represents our baseline approach. In the second, we stratify each pixel based on land cover (e.g. coniferous forest, arable/pasture) and calibrate the model independently using EO data specific to each stratum. We test the scale dependence of these approaches for grid resolutions spanning 1 to 0.05∘ over a mixed-land-use region of the UK. Our analyses indicate that spatial resolution matters for MDF. Under the community average baseline approach biological C fluxes (gross primary productivity, Reco) simulated by CARDAMOM are relatively insensitive to resolution. However, disturbance fluxes exhibit scale variance that increases with greater landscape fragmentation and for coarser model domains. In contrast, stratification of assimilated data based on fine-resolution land use distributions resolved the resolution dependence, leading to disturbance fluxes that were 40 %–100 % higher than the baseline experiments. The differences in the simulated disturbance fluxes result in estimates of the terrestrial carbon balance in the stratified experiment that suggest a weaker C sink compared to the baseline experiment. We also find that stratifying the model domain based on land use leads to differences in the retrieved parameters that reflect variations in ecosystem function between neighbouring areas of contrasting land use. The emergent differences in model parameters between land use strata give rise to divergent responses to future climate change. Accounting for fine-scale structure in heterogeneous landscapes (e.g. stratification) is therefore vital for ensuring the ecological fidelity of large-scale MDF frameworks. The need for stratification arises because land use places strong controls on the spatial distribution of carbon stocks and plant functional traits and on the ecological processes controlling the fluxes of C through landscapes, particularly those related to management and disturbance. Given the importance of disturbance to global terrestrial C fluxes, together with the widespread increase in fragmentation of forest landscapes, these results carry broader significance for the application of MDF frameworks to constrain the terrestrial C balance at regional and national scales.

Numerical Simulation of the Effect of Freeze-Thaw Cycles on the Axial Compression Strength of Rubber Concrete.

The incorporation of rubber can enhance concrete's durability and effectively reduce the damage caused by freeze-thaw cycling (FTC). Still, there has been only limited research on the damage mechanism of RC at the fine view level. To gain insight into the expansion process of uniaxial compression damage cracks in rubber concrete (RC) and summarize the internal temperature field distribution law during FTC, a fine RC thermodynamic model containing mortar, aggregate, rubber, water, and interfacial transition zone (ITZ) is established in this paper, and the cohesive element is selected for the ITZ part. The model can be used to study the mechanical properties of concrete before and after FTC. The validity of the calculation method was verified by comparing the calculated results of the compressive strength of concrete before and after FTC with the experimental results. On this basis, this study analyzed the compressive crack extension and internal temperature distribution of RC at 0, 5, 10, and 15% replacement rates before and after 0, 50, 100, and 150 cycles of FTC. The results showed that the fine-scale numerical simulation method can effectively reflect the mechanical properties of RC before and after FTC, and the computational results verify the applicability of the method to rubber concrete. The model can effectively reflect the uniaxial compression cracking pattern of RC before and after FTC. Incorporating rubber can impede temperature transfer and reduce the compressive strength loss caused by FTC in concrete. The FTC damage to RC can be reduced to a greater extent when the rubber incorporation is 10%.

A 3D Structure Mapping-Based Efficient Topology Optimization Framework

Abstract It is well known that the computational cost of classic topology optimization (TO) methods increases rapidly with the size of the design problem because of the high-dimensional numerical simulation required at each iteration. Recently, the technical route of replacing the TO process with artificial neural network (ANN) models has gained popularity. These ANN models, once trained, can rapidly produce an optimized design solution for a given design specification. However, the complex mapping relationship between design specifications and corresponding optimized structures presents challenges in the construction of neural networks with good generalizability. In this paper, we propose a new design framework that uses deep learning techniques to accelerate the TO process. Specifically, we present an efficient topology optimization (ETO) framework in which structure update at each iteration is conducted on a coarse scale and a structure mapping neural network (SMapNN) is constructed to map the updated coarse structure to its corresponding fine structure. As such, fine-scale numerical simulations are replaced by coarse-scale simulations, thereby greatly reducing the computational cost. In addition, fragmentation and padding strategies are used to improve the trainability and adaptability of SMapNN, leading to a better generalizability. The efficiency and accuracy of the proposed ETO framework are verified using both benchmark and complex design tasks. It has been shown that with the SMapNN, TO designs of millions of elements can be completed within a few minutes on a personal computer.

Machine learning assisted two-phase upscaling for large-scale oil-water system

The computation of two-phase upscaled functions entails solving time-dependent flow and transport equations over target regions, which is usually the most time-demanding component in the overall two-phase upscaling procedure. For large-scale reservoir models with a great number of coarse grid blocks, it can be very computationally expensive to calculate the two-phase upscaled functions for each individual coarse block. To address this problem, we develop a machine learning assisted upscaling (MLAU) approach, in which the two-phase upscaling is only performed for representative coarse blocks selected by a convolutional neural network (CNN) based clustering model, while the two-phase upscaled functions are quickly predicted for the rest of the coarse blocks using a regression algorithm. The performance of MLAU approach was assessed with three cases involving Gaussian, channelized and SPE 10 sector models, respectively. Numerical results have shown that the MLAU approach consistently provides coarse-scale results with close agreement with the results using full flow-based upscaling. Because two-phase numerical upscaling is only applied for representative coarse blocks (about 5% in each case), the speedups relative to the full flow-based upscaling are significant, ranging from 6.2 to 13.5. Compared to the fine-scale simulations, the speedups range from 27.0 to 47.2.

Evolution TANN and the identification of internal variables and evolution equations in solid mechanics

Data-driven and deep learning approaches have demonstrated to have the potential of replacing classical constitutive models for complex materials, displaying path-dependency and possessing multiple inherent scales. Yet, the necessity of structuring constitutive models with an incremental formulation has given rise to data-driven approaches where physical quantities, e.g. deformation, blend with artificial, non-physical ones, such as the increments in deformation and time. Neural networks and the consequent constitutive models depend, thus, on the particular incremental formulation, fail in identifying material representations locally in time, and suffer from poor generalization.Herein, we propose a new approach which allows, for the first time, to decouple the material representation from the incremental formulation. Inspired by the Thermodynamics-based Artificial Neural Networks (TANN) and the theory of the internal variables, the evolution TANN (eTANN) are continuous-time and, therefore, independent of the aforementioned artificial quantities. Key feature of the proposed approach is the identification of the evolution equations of the internal variables in the form of ordinary differential equations, rather than in an incremental discrete-time form.In this work, we focus attention to juxtapose and show how the various general notions of solid mechanics are implemented in eTANN. The laws of thermodynamics are hardwired in the structure of the network and allow predictions which are always consistent, independently of the range of the training dataset.Inspired by previous works, we propose a methodology that allows to identify, from data and first principles, admissible sets of internal variables from the microscopic fields in complex materials. The capabilities as well as the scalability of the proposed approach are demonstrated through several applications involving a broad spectrum of complex material behaviors, from plasticity to damage and viscosity (and combination of them). Finally, we show that the proposed approach can be used to speed-up state-of-the-art multiscale analyses, by virtue of asymptotic homogenization. eTANN provide excellent results compared to detailed fine-scale simulations and offer the possibility not only to describe the average macroscopic material behavior, but also micromechanical, complex mechanisms.

Advancing understanding of land-atmosphere interactions by breaking discipline and scale barriers.

Vegetation and atmosphere processes are coupled through a myriad of interactions linking plant transpiration, carbon dioxide assimilation, turbulent transport of moisture, heat and atmospheric constituents, aerosol formation, moist convection, and precipitation. Advances in our understanding are hampered by discipline barriers and challenges in understanding the role of small spatiotemporal scales. In this perspective, we propose to study the atmosphere-ecosystem interaction as a continuum by integrating leaf to regional scales (multiscale) and integrating biochemical and physical processes (multiprocesses). The challenges ahead are (1) How do clouds and canopies affect the transferring and in-canopy penetration of radiation, thereby impacting photosynthesis and biogenic chemical transformations? (2) How is the radiative energy spatially distributed and converted into turbulent fluxes of heat, moisture, carbon, and reactive compounds? (3) How do local (leaf-canopy-clouds, 1 m to kilometers) biochemical and physical processes interact with regional meteorology and atmospheric composition (kilometers to 100 km)? (4) How can we integrate the feedbacks between cloud radiative effects and plant physiology to reduce uncertainties in our climate projections driven by regional warming and enhanced carbon dioxide levels? Our methodology integrates fine-scale explicit simulations with new observational techniques to determine the role of unresolved small-scale spatiotemporal processes in weather and climatemodels.

Modeling multi-type urban landscape dynamics along the horizontal and vertical dimensions

Fine-scale urban simulation obtained through vector-based cellular automata (VCA) can provide more authentic parcel-level maps to aid urban planning. However, two aspects of VCA still need improvement for more detailed modeling of urban landscape dynamics. One is the lack of simulation of urban renewal prevalent in megacities. The other is the lack of simulation of vertical urban landscape dynamics at the parcel level. Hence, this study should be the first to propose a comprehensive VCA framework for modeling horizontal and vertical multi-type urban landscape dynamics (HV-MVCA) at the parcel level. The HV-MVCA consists of two modules: a deep learning-based horizontal urban landscape dynamic simulation module and a random forest-based (RF) vertical metrics prediction module. Compared with traditional VCA, the HV-MVCA can obtain the highest simulation accuracy with an enhancement of 1.45%-4.56%. The RF-based fitting model can reasonably interpret the vertical metric as the coefficient of determination and the mean absolute percentage error was 0.88 and 28.93%. By localizing shared socioeconomic pathways (SSPs), the horizontal and vertical urban landscape dynamics in the future can be simulated under different scenarios. As the multi-scenario simulation results can provide abundant horizontal and vertical information at the parcel level, the proposed HV-MVCA can help achieve a more comprehensive assessment of many urban-related issues with these indicators, such as energy consumption, climate change, and pollution emissions.

Comparison of two different types of reduced graph-based reservoir models: Interwell networks (GPSNet) versus aggregated coarse-grid networks (CGNet)

Computerized solutions for field management optimization often require reduced-order models to be computationally tractable. The purpose of this paper is to compare two different graph-based approaches for building such models. The first approach represents the reservoir as a graph of 1D numerical flow models that each connects an injector to a producer. One thus builds a network in which the topology is primarily determined by “well nodes” to which “non-well nodes” can be connected if need be. The second approach aims at building richer models so that the connectivity graph mimics the intercell connections in a conventional, coarse 3D grid model. One thus builds a network with topology defined by a mesh-like placement of “non-well nodes”, to which wells can be subsequently connected. The two approaches thus can be seen as graph-based analogues of traditional streamline and finite-volume simulation models. Both model types can be trained to match well responses obtained from underlying fine-scale simulations using standard misfit minimization methods; herein we rely on adjoint-based gradient optimization. Our comparisons show that graph models having a connectivity graph that mimics the intercell connectivity in coarse 3D models can represent a wider range of fluid connections and are generally more robust and easier to train than graph models built upon 1D subgridded interwell connections between injectors and producers only.

Impact of deformation bands on fault-related fluid flow in field-scale simulations

Subsurface storage of CO2 is predicted to rise exponentially in response to the increasing levels of CO2 in the atmosphere. Large-scale CO2 injections into the subsurface require understanding of the potential for fluid flow through faults to mitigate risk of leakage. Here, we study how to obtain effective permeability of deformation bands in the damage zone of faults. Deformation bands are relatively small, low permeability features that can have a significant effect on flow dynamics, however, the discrepancy of scales is a challenge for field-scale simulation. A new analytical upscaling model is proposed in order to overcome some of the shortcomings of conventional upscaling approaches for heterogeneous porous media. The new model captures the fine-scale impact of deformation bands on fluid flow in the near-fault region, and can be derived from knowledge of large-scale fault properties. To test the accuracy of the model it is compared to fine-scale numerical simulations that explicitly include individual deformation bands. For a wide range of different stochastically generated deformation bands networks, the upscaling model shows improved estimate of effective permeability compared to conventional upscaling approaches. By applying the upscaling model to a full-field simulation of the Smeaheia storage site in the North Sea, we show that deformation bands with a permeability contrast higher than three orders of magnitude may act as an extra layer of protection from fluid flow through faults.