Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Dynamic Timing Analysis”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 1,765 records · Page 98

Quantum statistical effects in nuclear reactions, fission, and open quantum systems

Quantum diffusion equations with time-dependent transport coefficients are derived from generalized non-Markovian Langevin equations. Generalized fluctuation-dissipation relations and analytical formulas for calculating friction and diffusion coefficients in nuclear processes are obtained. The asymptotics of the transport coefficients and of the correlation functions are investigated. The problem of correlation decay in quantum dissipative systems is studied. A comparative analysis of diffusion coefficients for the harmonic and inverted oscillators is performed. The role of quantum statistical effects during passage through a parabolic potential barrier is investigated. Sets of diffusion coefficient assuring the purity of states at any time instant are found in cases of non-Markovian dynamics. The influence of different sets of transport coefficients on the rate of decay from a metastable state is studied in the framework of the master equation for reduced density matrices describing open quantum systems. The approach developed is applied to investigation of fission processes and the processes of projectile-nuclei capture by target nuclei for bombarding energies in the vicinity of the Coulomb barrier. The influence of dissipation and fluctuation on these processes is taken into account in a self-consistent way. The evaporation residue cross sections for asymmetric fusion reactions are calculated from the derived capture probabilities averaged over all orientations of the deformed projectile and target nuclei.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

Two temperature regimes of triplet transfer in the dissociation of the correlated triplet pair after singlet fission

The ability to undergo spin-allowed exciton multiplication makes singlet fission materials promising for photovoltaic applications. Here, we examine the separation of correlated triplet pairs, 1 (T…T), in polycrystalline pentacene films via temperature-dependent transient absorption spectroscopy. Single wavelength analysis reveals a profound delay in 1 (T…T) dynamics. Moreover, the dynamics of 1 (T…T) exhibit temperature dependence, whereas other features show no discernable temperature dependence. Previous literatures have suggested that correlated triplet separation is mediated by a thermally activated hopping process. Surprisingly, we found that the time constants governing triplet pair separation display two distinct temperature-dependent regimes of triplet transport. The high temperature regime follows a thermally activated hopping mechanism. The experimentally derived reorganization energy and electronic coupling is verified by density matrix renormalization group quantum chemical calculations. In addition, we evaluated the low temperature regime and show that the trend can be modelled by a Miller–Abrahams-type model that incorporates the effects of energetic disorder. We conclude that the correlated triplet pair separation is mediated by thermally activated hopping or a disorder driven Miller–Abrahams-type mechanism at high and low temperature, respectively. We observe that crossover between two regimes occurs ~226 K. We find the time constant for triplet–triplet energy transfer to be 1.8 ps at ambient temperature and 21 ps at 77 K.

Chemistry↗

Modes and Mechanisms of Pacific Decadal-Scale Variability

The modes of Pacific decadal-scale variability (PDV), traditionally defined as statistical patterns of variance, reflect to first order the ocean's integration (i.e., reddening) of atmospheric forcing that arises from both a shift and a change in strength of the climatological (time-mean) atmospheric circulation. While these patterns concisely describe PDV, they do not distinguish among the key dynamical processes driving the evolution of PDV anomalies, including atmospheric and ocean teleconnections and coupled feedbacks with similar spatial structures that operate on different timescales. In this review, we synthesize past analysis using an empirical dynamical model constructed from monthly ocean surface anomalies drawn from several reanalysis products, showing that the PDV modes of variance result from two fundamental low-frequency dynamical eigenmodes: the North Pacific–central Pacific (NP-CP) and Kuroshio–Oyashio Extension (KOE) modes. Both eigenmodes highlight how two-way tropical–extratropical teleconnection dynamics are the primary mechanisms energizing and synchronizing the basin-scale footprint of PDV. While the NP-CP mode captures interannual- to decadal-scale variability, the KOE mode is linked to the basin-scale expression of PDV on decadal to multidecadal timescales, including contributions from the South Pacific.

54 ENVIRONMENTAL SCIENCES↗

Cloud Radiative Feedbacks and El Niño–Southern Oscillation

Cloud radiative feedbacks are disabled via “cloud-locking” in the Community Earth System Model, version 1.2 (CESM1.2), to result in a shift in El Niño–Southern Oscillation (ENSO) periodicity from 2–7 years to decadal time scales. We hypothesize that cloud radiative feedbacks may impact the periodicity in three ways: by 1) modulating heat flux locally into the equatorial Pacific subsurface through negative shortwave cloud feedback on sea surface temperature anomalies (SSTA), 2) damping the persistence of subtropical southeast Pacific SSTA such that the South Pacific meridional mode impacts the duration of ENSO events, or 3) controlling the meridional width of off-equatorial westerly winds, which impacts the periodicity of ENSO by initiating longer Rossby waves. The result of cloud-locking in CESM1.2 contrasts that of another study, which found that cloud-locking in a different global climate model led to decreased ENSO magnitude across all time scales due to a lack of positive longwave feedback on the anomalous Walker circulation. CESM1.2 contains this positive longwave feedback on the anomalous Walker circulation, but either its influence on the surface is decoupled from ocean dynamics or the feedback is only active on interannual time scales. The roles of cloud radiative feedbacks in ENSO in other global climate models are additionally considered. In particular, it is shown that one cannot predict the role of cloud radiative feedbacks in ENSO through a multimodel diagnostic analysis. Instead, they must be directly altered.

54 ENVIRONMENTAL SCIENCES↗

Photo-Switching of Protein Dynamical Collectivity

We examine changes in the picosecond structural dynamics with irreversible photobleaching of red fluorescent proteins (RFP) mCherry, mOrange2 and TagRFP-T. Measurements of the protein dynamical transition using terahertz time-domain spectroscopy show in all cases an increase in the turn-on temperature in the bleached state. The result is surprising given that there is little change in the protein surface, and thus, the solvent dynamics held responsible for the transition should not change. A spectral analysis of the measurements guided by quasiharmonic calculations of the protein absorbance reveals that indeed the solvent dynamical turn-on temperature is independent of the thermal stability/photostate however the protein dynamical turn-on temperature shifts to higher temperatures. This is the first demonstration of switching the protein dynamical turn-on temperature with protein functional state. The observed shift in protein dynamical turn-on temperature relative to the solvent indicates an increase in the required mobile waters necessary for the protein picosecond motions, that is, these motions are more collective. Melting-point measurements reveal that the photobleached state is more thermally stable, and structural analysis of related RFP’s shows that there is an increase in internal water channels as well as a more uniform atomic root mean squared displacement. These observations are consistent with previous suggestions that water channels form with extended light excitation providing O2 access to the chromophore and subsequent fluorescence loss. We report that these same channels increase internal coupling enhancing thermal stability and collectivity of the picosecond protein motions. The terahertz spectroscopic characterization of the protein and solvent dynamical onsets can be applied generally to measure changes in collectivity of protein motions.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Revisiting the Architecture of the KOI-89 System

While high stellar obliquities observed in exoplanetary systems may be attributed to processes that tilt the planetary orbits, it is also possible that they reflect misalignments between protoplanetary disks and stellar spins. This latter hypothesis predicts the presence of coplanar multiplanetary systems misaligned with their central stars. Here we reevaluate the evidence of such an architecture that has been claimed for the KOI-89 system. An early-type star, KOI-89 has one validated transiting planet, KOI-89.01/Kepler-462b (period 84.7 days, radius 3.0 R {sub ⊕}), and one transiting planet candidate, KOI-89.02 (period 207.6 days, radius 4.0 R {sub ⊕}), where the latter exhibits transit timing variations (TTVs). A previous modeling of the stellar gravity-darkening effect in the transit light curves inferred a high stellar obliquity of ≈70°. We perform photodynamical modeling of the Kepler transit light curves and use the resulting constraints on the orbital configuration and transit times to update the gravity-darkened transit model. As a result, we find no firm evidence for the gravity-darkening effect in the transit shapes and conclude that stellar obliquity is not constrained by the data. Given the evidence for low orbital eccentricities from the dynamical analysis, the system architecture can thus be consistent with many other multitransiting systems with flat, near-circular orbits aligned with the stellar spin. We find that the TTVs imparted on its neighbor imply that KOI-89.01 has a mass ≳20 M {sub ⊕}. This would render it one of the densest known sub-Neptunes, mostly composed of a solid core. Lower masses are possible if the TTVs are instead due to an unseen third planet.

79 ASTRONOMY AND ASTROPHYSICS↗

Model-Agnostic Algorithm for Real-Time Attack Identification in Power Grid using Koopman Modes

Malicious activities on measurements from sensors like Phasor Measurement Units (PMUs) can mislead the control center operator into taking wrong control actions resulting in disruption of operation, financial losses, and equipment damage. In particular, false data attacks initiated during power systems transients caused due to abrupt changes in load and generation can fool the conventional model-based detection methods relying on thresholds comparison to trigger an anomaly. In this paper, we propose a Koopman mode decomposition (KMD) based algorithm to detect and identify false data attacks in real-time. The Koopman modes (KMs) are capable of capturing the nonlinear modes of oscillation in the transient dynamics of the power networks and reveal the spatial embedding of both natural and anomalous modes of oscillations in the sensor measurements. The Koopman-based spatio-temporal nonlinear modal analysis is used to filter out the false data injected by an attacker. The performance of the algorithm is illustrated on the IEEE 68-bus test system using synthetic attack scenarios generated on GridSTAGE, a recently developed multivariate spatio-temporal data generation framework for simulation of adversarial scenarios in cyber-physical power systems.

Nandanoori, Sai Pushpak↗

Dynamic domain kinematic modelling for predicting interflow over leaky impeding layers

Traditional Boussinesq or kinematic simulations of interflow (i.e., lateral subsurface flow) assume no leakage through the impeding layer and require a no-flow boundary condition at the ridge top. However, recent analyses of many interflow-producing landscapes indicate that leaky impeding layers are common, that most interflow percolates well before reaching the toe slope, and therefore, the downslope contributing length is shorter than the hillslope length. In watersheds characterised by perched interflow over a low conductivity layer through permeable topsoil, interflow with percolation may be modelled with a kinematic wave model using a mobile upslope boundary condition defining the hillslope portion contributing interflow to valleys. Here, we developed and applied a dynamic interflow model to simulate interflow using a downslope travel distance concept such that only the active contributing length is modelled at any time. The model defines a variable active area based on the depth of the perched layer, the topographic slope and the ratio of the hydraulic conductivity of topsoil to that of the impeding layer. It incorporates a two-layer soil moisture accounting water balance analysis, a pedo-transfer function, and percolation and evaporation routines to predict interflow rates in continuous and event-based scenarios. We tested the modelling concept on two sets of data (2-year dataset of rainfall observations for the continuous simulation and a multi-day irrigation experiment for the event simulation) from a 121-m-long open interflow collection trench on an experimental hillslope at the Savannah River Site, South Carolina. The continuous model simulation partially represented the observed interflow hydrograph and perched water depth in the experimental hillslope with correlation coefficients of 0.85 and 0.35, respectively. Model performance improved significantly at event-scale analysis. The modelling approach realistically represents interflow dynamics in hillslopes with leaky impeding layers and can be integrated into catchment-scale hydrology models for more detailed hillslope process modelling.

54 ENVIRONMENTAL SCIENCES↗

Investigation of the Use of Dynamic Probabilistic Risk Assessment Methodologies for Identifying Digital I&C System Common Cause Failures

Digital Instrumentation and Control (I&C) systems have a key role in nuclear power plants in the upgrade of aging analog systems. Digital systems improve plant safety and reliability through features such as increased hardware reliability and stability and improved failure detection capability. There is no consensus on which of the current probabilistic risk assessment methods are most suitable for use in the reliability analysis of digital I&C systems. While the traditional event-tree/fault-tree (ET/FT) approach is still used for their reliability modeling, there are concerns regarding this approach in properly accounting for dynamic interactions among system components since potentially significant dependencies among failure events may not be identified and/or their likelihood may not be properly quantified. Dynamic methodologies are expected to provide a much more accurate representation of probabilistic evolution of the I&C systems in time due to their capability to more properly account for complex interactions than the static approach. The applicability of dynamic PRA methodologies for digital I&C system is investigated using the criteria presented in the NUREG/CR-6901, and the comparisons made in NUREG/CR-6901 are updated in light of the latest studies. The Dynamic Event Tree (DET) approach has been identified as one of the top dynamic methods when evaluated against the requirements for the reliability modeling of digital I&C systems. The DET method is a strong candidate for integration into existing PRA studies, as it bears many similarities to the traditional ET approach. In this study, the DET approach has been applied to the Plant Protection System of the APR1400 design, and the results are compared to results from its available traditional ET/FT analysis. Possible approaches to evaluate and quantify the effects of common cause failures on system safety using dynamic methods are also examined.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

Destruction dynamics of a heterogeneous body (diorite) under friction

Triboluminescence bursts are observed in two heterogeneous (diorite) specimens under friction against each other. Triboluminescence appears upon the relaxation of excited free ≡Si–O{sup –} radicals and Fe{sup 3+} ions and the capture of electrons with acceptor traps formed upon the destruction of the plagioclase crystal lattice. The analysis of the time dependence of these bursts shows that the friction surfaces accumulate clusters, in which the concentration of free ≡Si–O{sup –} radicals and electron traps is at least an order of magnitude higher than in their surrounding. The time interval between the appearance of two sequential clusters variates from 0.1 to 1 μs. The linear sizes of clusters are ~0.5 μm.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗

Multi-scale, Multi-disciplinary, and Multi-agent Explainable AI with Koopman-Undergirded Learning, Prediction, and Analysis (M3EA KULPA) (Project Closeout Report)

The goal of this project was to develop and use domain-aware machine learning formulations, based on the Koopman Operator (KO), for modelling multi-scale, multi-disciplinary (e.g., multi-physics), and/or multi-agent systems. The project developed these formulations for the following cases: • Systems with dynamics at two separate time scales, • Systems with a bi-level hierarchical control structure, • Systems with bi-level hierarchical control and dynamics at two separate time scales (the lower level controls operating at the faster time scale), and • Systems with n separate but interacting agents/disciplines (with/without control, respectively); the controls for each agent could include bi-level hierarchical control and dynamics at two separate time scales as described above. The project then defined a set of dynamical systems consisting of different nonlinear oscillators that could be used to test these different formulations and then subsequently learned the KO models for those systems. With the KO models, we were able to do the following: • Quantify system stability, including both long-term and transient behavior, • Quantify the effects of feedbacks between the different time scales and agents/disciplines in terms of those feedbacks’ effects on system stability, • Replace a standard Proportional-Integral (PI) control in the hierarchical control structure with a KO-based Linear-Quadratic Regular (LQR), a form of optimal control, • Calculate optimal supervisory control policies a) with and without time scale separated dynamics at the lower level control levels and b) with both PI and KO-based LQR lower level control policies, and • Calculate dynamic Nash equilibria for multi-agent systems where each agent makes its own control decisions.

97 MATHEMATICS AND COMPUTING↗

Rewarding Grid-Friendly Behavior: Estimating the Potential Bill Reduction and Load Shifting Benefits of Dynamic Prices

Shifting electric load from times of peak demand can be a key strategy to slow price growth as reducing peak demand avoids the cost of upgrading generation, transmission and distribution infrastructure. Utilities are releasing time-varying prices, such as time of use rates or dynamic prices, to incentivize grid-friendly load shifting. New dynamic price programs provide insight into the true cost of operating electricity grids and the potential economic benefits of load shifting. Program developers and device manufacturers need to understand the economic opportunities in terms of 1) the variation in prices across hours, days, and seasons; 2) the change in utility bills for customers who don’t shift load; and 3) the potential load shifted and economic value of different technologies if manufacturers or aggregators deploy price-responsive controls. This paper estimates possible impacts of dynamic price adoption and load shifting controls if customers paid the dynamic rate from one pilot program. Statistical analysis of historical prices identified annual and seasonal metrics as well as representative price curves for each circuit in the pilot. Simulations for residential technologies with price-responsive controls including unitary heat pump water heaters, central multifamily heat pump water heaters, heating & cooling + storage systems, and pool pumps estimated the potential impacts of highly dynamic prices both with and without load shifting controls. Results showed the potential to reduce electricity costs on representative days by 42-94% and reduce consumption during times of high electricity prices by 63-100% compared to baseline operation for those flex-friendly devices.

Grant, Peter↗

Decadal Spiciness Variability in the Subtropical-Tropical Pacific in the CESM2 Large Ensemble

Tropical Pacific decadal variations impact weather and climate around the world and are also connected to variations in the global warming trend. The mechanisms driving these long-term modulations, particularly the role of subsurface ocean dynamics, are still debated. Here, we investigate the dynamics of spiciness (density-compensated temperature and salinity) anomalies in the tropical and subtropical Pacific, which are hypothesized as a possible driving mechanism of decadal climate variability. Based on the analysis of 100 realizations from the Community Earth System Model Version 2 Large Ensemble (CESM2-LE), we demonstrate a coupling between the subtropics and the equatorial Pacific by propagating spiciness anomalies at decadal time scales. The CESM2-LE simulates spiciness variability along a subduction path from the subtropics to the equator with frequency spectra that show the highest power at low frequencies and a power decay proportional to a −4 slope for frequencies greater than 0.01 cycles per months, corresponding to periods smaller than ∼8.5 years. Signals that originate in the Southern Hemisphere (SH) dominate and arrive with a larger magnitude at the equator compared to spiciness anomalies from the Northern Hemisphere (NH). Spiciness anomalies from the SH have shorter propagation times and are strengthened along their pathway as stochastic wind stress curl forcing generates anomalous baroclinic ocean pressure gradients. These pressure gradients generate spiciness anomalies via anomalous advection across climatological spiciness gradients in the SH. We conclude that the observed spiciness variance at decadal time scales is consistent with a forcing by stochastic wind variations that are low-pass filtered by ocean dynamics.

54 ENVIRONMENTAL SCIENCES↗

PROPERTIES OF A CORONAL SHOCK WAVE AS A DRIVER OF EARLY SEP ACCELERATION

Coronal mass ejectmons (CMEs) are thought to drive collisionless shocks in the solar corona, which in turn have been shown to be capable of accelerating solar energetic particles (SEPs) in minutes. It has been notoriously difficult to extract information about energetic particle spectra in the corona, owing to a lack of in situ measurements. It is possible, however, to combine remote observations with data-driven models in order to deduce coronal shock properties relevant to the local acceleration of SEPs and their heliospheric connectivity to near-Earth space. We present such novel analysis applied to the 2011 May 11 CME event on the western solar limb, focusing on the evolution of the eruption-driven, dome-like shock wave observed by the Atmospheric Imaging Assembly (AIA) EUV telescopes on board the Solar Dynamics Observatory spacecraft. We analyze the shock evolution and estimate its strength using emission measure modeling. We apply a new method combining a geometric model of the shock front with a potential field source surface model to estimate time-dependent field-to-shock angles and heliospheric connectivity during shock passage in the low corona. We find that the shock was weak, with an initial speed of ∼450 km s{sup –1}. It was initially mostly quasi-parallel, but a significant portion of it turned quasi-perpendicular later in the event. There was good magnetic connectivity to near-Earth space toward the end of the event as observed by the AIA instrument. The methods used in this analysis hold a significant potential for early characterization of coronal shock waves and forecasting of SEP spectra based on remote observations.

79 ASTRONOMY AND ASTROPHYSICS↗

FERMI view of the TeV blazar Markarian 421

The high energy component of the TeV blazar Markarian 421 has been extensively studied since the beginning of the 90s, when the source was first detected at gamma-rays with EGRET and the Whipple Telescope, yet the source is still far from being understood. The high sensitivity, large dynamic range, and excellent time coverage of the Fermi Large Area Telescope (LAT), all representing significant advances over previous gamma-ray observations, will play a key role in the elucidation of the physical processes underlying the high energy emission of this blazar. In this presentation we show the results from almost 6 months (4 August 2008 to 20 January 2009) of observation with LAT. We report significant flux/spectral variability on a range of time scales from weeks to days, and an energy spectrum from 0.1 GeV to 300 GeV, overlapping with the energy ranges covered by the current generation of Cherenkov Telescopes. Results on the observations of the BLLac object Markarian 421 collected in the first months of operation of the Fermi satellite have been presented. Light curves on weekly and daily timescales have been shown, as well as the results of the spectral analysis in the energy range between 100 MeV and 300 GeV, covered for the first time by a satellite experiment overlapping the lower energy observations from Cherenkov telescopes on earth. These results are still preliminary and will be enriched and completed soon by a forthcoming publication. The results shown here demonstrate the great performance of Fermi-LAT to study the gamma-emission from Mrk421 (and blazars in general) over a large dynamic range and also on short timescales, which is expected to be of key importance for the study of the emission of the source in a coordinated way with other instruments covering other energy ranges.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Review: Recent advances of ToF-SIMS for environmental analysis and imaging

Background: Time-of-flight secondary ion mass spectrometry (ToF-SIMS) is a powerful surface analysis technique, initially developed and applied in inorganic materials and semiconductors. In past decades, ToF-SIMS has attracted more attention in its analysis capabilities of organic materials, with increased applications in biology, medical, and health development. It has also become a versatile and effective tool in environmental analysis due to its high mass resolution, mass accuracy, and depth profiling. Results: In this review, we first give an overview of the principle of ToF-SIMS and follow with recent ToF-SIMS applications in exemplary environmental study cases, including atmospheric aerosol, soil, water, plant, and organic solvent analysis. Moreover, sample preparation techniques are summarized in relation to corresponding environmental applications. Specifically, we call attention to ToF-SIMS investigations showcasing studies in surface chemical compositions, images, and depth profile analysis. These findings emphasize the important role of interfacial chemistry in environmental processes and provide valuable insights into dynamic processes, such as chemical transformation, particle formation, plant biology, and microbial inspired biotechnology development. The mass spectral imaging results acquired by ToF-SIMS offer a deeper understanding of intermediate stages and transient phases for environmental specimens. Significance: In situ and operando imaging offer new possibilities in studying phenomena in real time with high spatial resolution. Furthermore, it is anticipated that more research groups will use ToF-SIMS in environmental research given recent advances in measurement capabilities and surging needs in chemical mapping of complex analytes and systems.

Aerosol↗

An integrated computational materials engineering framework to analyze the failure behaviors of carbon fiber reinforced polymer composites for lightweight vehicle applications

A bottom-up multi-scale modeling approach is used to develop an Integrated Computational Materials Engineering (ICME) framework for carbon fiber reinforced polymer (CFRP) composites, which has the potential to reduce development to deployment lead time for structural applications in lightweight vehicles. In this work, we develop and integrate computational models comprising of four size scales to fully describe and characterize three types of CFRP composites. In detail, the properties of the interphase region are determined by an analytical gradient model and molecular dynamics analysis at the nano-scale, which is then incorporated into micro-scale unidirectional (UD) representative volume element (RVE) models to characterize the failure strengths and envelopes of UD CFRP composites. Then, the results are leveraged to propose an elasto-plastic-damage constitutive law for UD composites to study the fiber tows of woven composites as well as the chips of sheet molding compound (SMC) composites. Subsequently, the failure mechanisms and failure strengths of woven and SMC composites are predicted by the meso-scale RVE models. Finally, building upon the models and results from lower scales, we show that a homogenized macro-scale model can capture the mechanical performance of a hat-section-shaped part under four-point bending. Along with the model integration, we will also demonstrate that the computational results are in good agreement with experiments conducted at different scales. The present work illustrates the potential and significance of integrated multi-scale computational modeling tools that can virtually evaluate the performance of CFRP composites and provide design guidance for CFRP composites used in structural applications.

36 MATERIALS SCIENCE↗

Advancing spatiotemporal forecasts of CO 2 plume migration using deep learning networks with transfer learning and interpretation analysis

Accurate and timely forecasts of CO 2 plume distribution throughout the injection and post-injection phases are crucial for detecting plume migration, assessing leakage risks, and supporting operational decisions in geologic carbon storage (GCS). Current convolutional neural network-based approaches primarily focus on spatial information and overlook temporal dependencies in plume distributions, thus limiting their ability to capture dynamic movement effects and provide accurate predictions of plume migration. In this work, we propose two deep learning models, Auto-Encoder (AE)-LSTM and Encoder-Decoder (ED)-ConvLSTM, each uniquely designed to capture both spatial and temporal features. We apply the proposed methods to forecast the dynamic distribution of CO 2 plumes based on 108 reservoir simulations over a 30-year injection and a 30-year post-injection period. The results indicate that the ED-ConvLSTM model outperforms the AE-LSTM model in accurately predicting the spatiotemporal dynamics of CO 2 plume migration, achieving R 2 values above 0.99. To provide a deeper understanding of these model predictions, we employ a gradient-based explanation method on the trained models. This approach provides insights into the influence of input variables on plume migration forecasts and uncovers the underlying prediction mechanisms of the proposed models. Furthermore, we introduce a transfer learning technique, enabling fast and accurate plume migration forecasting in the post-injection phase by leveraging the trained model during the injection phase. This reduces the necessity for extensive data collection or re-training. In conclusion, the methods proposed in our work enhances the performance and interpretability of CO 2 plume migration forecasts, thereby facilitating informed decision-making throughout the entire lifecycle of GCS applications.

58 GEOSCIENCES↗