Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “system partition”

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 379 records · Page 21
All sourcesNASA NTRSDOE OSTI
DOE OSTI2024

IRIS: A Performance-Portable Framework for Cross-Platform Heterogeneous Computing

From edge to exascale, computer architectures are becoming more heterogeneous and complex. The systems typically have fat nodes, with multicore CPUs and multiple hardware accelerators such as GPUs, FPGAs, and DSPs. This complexity is causing a crisis in programming systems and performance portability. Several programming systems are working to address these challenges, but the increasing architectural diversity is forcing software stacks and applications to be specialized for each architecture. As we show, all of these approaches critically depend on their software framework for discovery, execution, scheduling, and data orchestration. To address this challenge, we believe that a more agile and proactive software framework is essential to increase performance portability and improve user productivity. To this end, we have designed and implemented IRIS: a performance-portable framework for cross-platform heterogeneous computing. IRIS can discover available resources, manage multiple diverse programming platforms (e.g., CUDA, Hexagon, HIP, Level Zero, OpenCL, OpenMP) simultaneously in the same execution, respect data dependencies, orchestrate data movement proactively, and provide for user-configurable scheduling. To simplify data movement, IRIS introduces a shared virtual device memory with relaxed consistency among different heterogeneous devices. IRIS also adds an automatic kernel workload partitioning technique using the polyhedral model so that it can resize kernels for a wide range of devices. Our evaluation on three architectures, ranging from Qualcomm Snapdragon to a Summit supercomputer node, shows that IRIS improves portability across a wide range of diverse heterogeneous architectures with negligible overhead.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2024

Recent Advances of PyROS: A Pyomo Solver for Nonconvex Two-Stage Robust Optimization in Process Systems Engineering

The document presents recent algorithmic and implementation advances of the two-stage robust optimization (RO) solver PyROS, and a benchmarking study which demonstrates the utility of PyROS for two-stage RO problems. The advances include extensions of the scope of PyROS to models with uncertain variable bounds, improvements to the initializations of the subproblems used by the underlying cutting set algorithm, and extensions of the uncertainty set interfaces. The benchmarking study is performed on a library of over 8,500 instances, with variations in the nonlinearities, degree-of-freedom partitioning, uncertainty sets, and polynomial decision rule approximations. An amine-based CO2 capture case study is presented to demonstrate the utility of PyROS for large-scale process models. Overall, the results highlight the effectiveness of PyROS for obtaining robust solutions to optimization problems with uncertain equality constraints.

Sherman, Jason↗
DOE OSTI2024

Localization of proteins involved in the biogenesis and repair of the photosynthetic apparatus to thylakoid subdomains in Arabidopsis

Abstract Thylakoid membranes in chloroplasts and cyanobacteria harbor the multisubunit protein complexes that catalyze the light reactions of photosynthesis. In plant chloroplasts, the thylakoid membrane system comprises a highly organized network with several subcompartments that differ in composition and morphology: grana stacks, unstacked stromal lamellae, and grana margins at the interface between stacked and unstacked regions. The localization of components of the photosynthetic apparatus among these subcompartments has been well characterized. However, less is known about the localization of proteins involved in the biogenesis and repair of the photosynthetic apparatus, the partitioning of proteins between two recently resolved components of the traditional margin fraction (refined margins and curvature), and the effects of light on these features. In this study, we analyzed the partitioning of numerous thylakoid biogenesis and repair factors among grana, curvature, refined margin, and stromal lamellae fractions of Arabidopsis thylakoid membranes, comparing the results from illuminated and dark‐adapted plants. Several proteins previously shown to localize to a margin fraction partitioned in varying ways among the resolved curvature and refined margin fractions. For example, the ALB3 insertase and FtsH protease involved in photosystem II (PSII) repair were concentrated in the refined margin fraction, whereas TAT translocon subunits and proteins involved in early steps in photosystem assembly were concentrated in the curvature fraction. By contrast, two photosystem assembly factors that facilitate late assembly steps were depleted from the curvature fraction. The enrichment of the PSII subunit OE23/PsbP in the curvature fraction set it apart from other PSII subunits, supporting the previous conjecture that OE23/PsbP assists in PSII biogenesis and/or repair. The PSII assembly factor PAM68 partitioned differently among thylakoid fractions from dark‐adapted plants and illuminated plants and was the only analyzed protein to convincingly do so. These results demonstrate an unanticipated spatial heterogeneity of photosystem biogenesis and repair functions in thylakoid membranes and reveal the curvature fraction to be a focal point of early photosystem biogenesis.

Chotewutmontri, Prakitchai↗
DOE OSTI2024

Performance Results on CPU/GPU Exascale Architectures for OMEGA: The Ocean Model for E3SM Global Applications

The US Department of Energy (DOE) conducts climate simulations on some of the world’s largest supercomputers. These exascale machines use heterogeneous architectures with both CPUs and GPUs, and scientific codes must adapt to make full use of this computing power. Los Alamos National Lab is developing Omega: The Ocean Model for E3SM Global Applications, which is specifically designed for modern exascale computers. It uses external libraries that have been optimized for a variety of architectures to run on different supercomputers. Omega is an unstructured-mesh ocean model based on TRiSK numerical methods. It will be the new ocean component of the DOE’s Energy Exascale Earth System Model (E3SM). The algorithms in Omega follow those of the current ocean component, MPAS-Ocean, but it will be written in C++ rather than Fortran to take advantage of the Kokkos performance portability library. Omega spatial operators are written as Kokkos kernels to run efficiently on both CPUs and GPUs. Work on Omega began in 2023 with a new C++ framework for unstructured mesh partitioning, halo exchanges, parallel IO, and Kokkos interfaces. The current version, Omega-0, is being developed to solve the shallow water equations and at present includes all of the tendency terms but not time stepping. Here we share the results of Omega-0 verification and performance testing. Verification includes unit tests implemented with CTest as well as convergence tests in Polaris, an in-house python package with a large suite of test problems. Performance tests compare simulations conducted on CPUs versus GPUs and across different architectures: tests are run on Frontier, which has AMD “Optimized 3rd Gen EPYC” CPUs and AMD MI250X GPUs, as well as Perlmutter, which is composed of AMD EPYC 7763 CPUs and NVIDIA A100 GPUs.

58 GEOSCIENCES↗
DOE OSTI2022

QUEST: systematically approximating Quantum circuits for higher output fidelity

We present QUEST, a procedure to systematically generate approximations for quantum circuits to reduce their CNOT gate count. Our approach employs circuit partitioning for scalability with procedures to 1) reduce circuit length using approximate synthesis, 2) improve fidelity by running circuits that represent key samples in the approximation space, and 3) reason about approximation upper bound. Our evaluation results indicate that our approach of "dissimilar"approximations provides close fidelity to the original circuit. Overall, the results indicate that QUEST can reduce CNOT gate count by 30-80% on ideal systems and decrease the impact of noise on existing and near-future quantum systems.

Patel, Tirthak↗
DOE OSTI2024

Privacy-Preserving Control of Partitioned Energy Resources

Distributed energy resources are an increasingly important part of the electric grid. We examine the problem of partitioning a distributed energy resource among many users while providing privacy to them. In this model, clients can send requests to a server, the server can verify that the requests are valid and aggregate them, but it cannot see the actual values in the requests. Without privacy, each user is forced to reveal their daily schedule or energy use. Energy resources add a novel challenge that prior systems do not address: they require verifying limits on private power (a rate over time) and energy (a sum) values. Furthermore, the cryptographic mechanisms must run on embedded energy control systems. We describe Weft, a novel cryptographic system that verifies both power (rate) and energy (integral) constraints on private client values and aggregates them. The key insight behind the approach is to rely on additively homomorphic secret shares, which allows servers to compute sums from rates. We present 3 cryptographic proof systems with different system trade-off for embedded systems: bit-splitting proofs minimize memory use, sorting proofs minimize computation, and commitment proofs minimize network communication. Using bit-splitting proofs, it takes an IoT client using a CortexM microcontroller 4 minutes of compute time to privately control its share of an energy resource for a day at 20s granularity.

Laufer, Evan↗
DOE OSTI2020

Ecogenomics reveals community interactions in a long-term methanogenic bioreactor and a rapid switch to sulfate-reducing conditions

ABSTRACT The anaerobic digestion of wastes is globally important in the production of methane (CH4) as a biofuel. When sulfate is present, sulfate-reducing bacteria (SRB) are stimulated, competing with methanogens for common substrates, which decreases CH4 production and results in the formation of corrosive, odorous hydrogen sulfide gas (H2S). Here, we show that a population of SRB within a methanogenic bioreactor fed only butyrate for years immediately (within hours) responded to sulfate availability and shifted the microbial community dynamics within the bioreactor. By mapping shotgun metatranscriptomes to metagenome-assembled genomes, we shed light on the transcriptomic responses of key community members in response to increased sulfate provision. We link these short-term transcriptional responses to long-term niche partitioning using comparative metagenomic analyses. Our results suggest that sulfate provision supports a syntrophic butyrate oxidation community that disfavors poly-β-hydroxyalkanoate storage and that hydrogenotrophic SRB populations effectively exclude obligately hydrogenotrophic, but not aceticlastic, methanogens when sulfate is readily available. These findings elucidate key ecological dynamics between SRB, methanogens and syntrophic butyrate-oxidizing bacteria, which can be applied to a variety of engineered and natural systems.

St. James, Andrew R.↗
DOE OSTI2023

Consideration of Nonideal Detonation Regimes Influenced by Wave Modes in a Water-Cooled Rotating Detonation Engine Using OH* Chemiluminescence

Although inherently unstable, existing research in rotating detonation combustion supports its application in notionally steady processes resulting in greater availability compared to conventional, constant pressure combustion. Further improvements rely on a more in-depth understanding of system losses and identifying conditions which optimize device performance. Within this study, the presence and proportion of ideal and nonideal combustion regimes are compared across a variety of process conditions and wave modes. Here, large-scale data analysis seeks to summarize proportional heat release associated with commensal, parasitic, and detonative combustion averaged across individual traces of OH* chemiluminescent data acquired at the detonation plane. Means of regime partitioning based on the anatomy of the time-resolved OH* signal are proposed to ensure consistent analysis throughout the current and future studies concerning combustion regimes. Of particular interest is the possible influence of wave on the nonideal combustion in relative proportion to the desired detonation. Results showed improved percent detonation with increasing significance for the following trends: decreasing equivalence ratio, increasing wave count, decreasing wave velocity, and increasing detonation time. Increased wave number, brought on by decreased equivalence ratios and wave velocities, is thought to decrease fill region surface area, and therefore, decrease nonideal contact burning. Additional analysis is performed to consider possible trend variation due to the presence of stable galloping waves, which were found to have minimal influence on relative percent detonation values. The outcome of this study suggests operational states, which correspond to increased wave quantities for increased proportions of reactants consumed by the targeted detonative combustion regime.

42 ENGINEERING↗
DOE OSTI2021

Distributed ADMM Using Private Blockchain for Power Flow Optimization in Distribution Network With Coupled and Mixed-Integer Constraints

The optimization problem for scheduling distributed energy resources (DERs) and battery energy storage systems (BESS) integrated with the power grid is important to minimize energy consumption from conventional sources in response to demand. Conventionally this optimization problem is solved in a centralized manner, limiting the size of the problem that can be solved and creating a high communication overhead because all the data is transferred to the central controller. These limitations are addressed by the proposed distributed consensus-based alternating direction method of multiplier (DC-ADMM) optimization algorithm, which decomposes the optimization problem into subproblems with private cost function and constraints. The distribution feeder is partitioned into low coupling subnetworks/regions, which solves the private subproblem locally and exchanges information with the neighboring regions to reach consensus. The relaxation strategy is employed for mixed-integer and coupled constraints introduced in the optimal power flow (OPF) problem by stationary and transportable BESS because DC-ADMM convergence is only guaranteed for strict convex problems. The information exchange and synchronization between subnetworks/regions are vital for distributed optimization. In this work, both of these aspects are addressed by the blockchain. The smart contract deployed on the blockchain network acts as a mediator for secure data exchange and synchronization in distributed computation. The blockchain-based distributed optimization problem’s effectiveness is tested for a 0.5-MW laboratory microgrid for one hour ahead and day-ahead for the IEEE 123-bus and EPRI J1 test feeders, and results are compared with a centralized solution.

25 ENERGY STORAGE↗
DOE OSTI2023

Automated identification and calculation of prompt effects in kinetic mechanisms using statistical models

The kinetics of prompt dissociation involves rovibrationally excited species (generally formed by exothermic reactions) which may dissociate or isomerize prior to thermalization via collisions with the bath gas. Treating such rovibrationally excited species (so-called "hot" species) with standard kinetic phenomenology may result in incorrect macroscopic representation of their reactivity. Here this work presents the first fully automated methodology for the calculation of prompt effects of a chosen species in a kinetic mechanism, including (i) reaction selection; (ii) theoretical calculation of rate constants and prompt branching fractions; and (iii) final rate constant fitting. The energy partition between hot fragments is estimated using a variety of statistical models, including a new physically sound microcanonical statistical model based on the rovibrational density of states of the fragments. The methodology is validated against literature data for the prompt dissociations of HCO and C 3 H 7 radicals. The microcanonical statistical model is in better agreement with trajectory simulations for larger species and is thus applicable for practical systems that typically involve large molecules, for which direct dynamics calculations are impractical. The automated workflow is applied to the evaluation of the effects of prompt dissociation for two isomeric radicals C 4 H 7 1-3 (1-methylallyl) and C 4 H 7 1-4 (3-buten-1-yl). Twelve H-atom abstraction reactions are selected and the corresponding rate constants are computed with first principles theory. The microcanonical statistical model predicts that prompt dissociations of C 4 H 7 1-3 and C 4 H 7 1-4 are already significant at 1000K, resulting in differences of up to an order of magnitude at 2000K with respect to the phenomenological thermal rate constants. To illustrate the effects of prompt dissociation on simulations of experimental data, the calculated prompt rate constants are implemented in both CRECK and C3MechV3.3 kinetic mechanisms. Simulations of experimental flame data illustrate the noticeable impact of prompt dissociation kinetics on the high-temperature combustion reactivity of C 4 H 8 -1 and C 4 H 8 -2.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2023

Software Defined Architectures for Portability and Performance

The Software Defined Architectures for Portability and Performance (SODAPOP) project developed a co-design framework to partition and map converged applications on specialized heterogeneous architectures. We started from key domain applications that combine scientific simulation with data analytics and machine learning as drivers to integrate our framework. The framework includes high-level compilers that interfaces with high-level programming frameworks, domain-specific optimization passes, and hardware-oriented optimizations. The framework leverages hardware generators to enable specialization and facilitate exploration of custom system designs.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2020

Predicting secondary organic aerosol phase state and viscosity and its effect on multiphase chemistry in a regional-scale air quality model

Atmospheric aerosols are a significant public health hazard and have substantial impacts on the climate. Secondary organic aerosols (SOAs) have been shown to phase separate into a highly viscous organic outer layer surrounding an aqueous core. This phase separation can decrease the partitioning of semi-volatile and low-volatile species to the organic phase and alter the extent of acid-catalyzed reactions in the aqueous core. A new algorithm that can determine SOA phase separation based on their glass transition temperature (T g ), oxygen to carbon (O:C) ratio and organic mass to sulfate ratio, and meteorological conditions was implemented into the Community Multiscale Air Quality Modeling (CMAQ) system version 5.2.1 and was used to simulate the conditions in the continental United States for the summer of 2013. SOA formed at the ground/surface level was predicted to be phase separated with core–shell morphology, i.e., aqueous inorganic core surrounded by organic coating 65.4 % of the time during the 2013 Southern Oxidant and Aerosol Study (SOAS) on average in the isoprene-rich southeastern United States. Our estimate is in proximity to the previously reported ~70 % in literature. The phase states of organic coatings switched between semi-solid and liquid states, depending on the environmental conditions. The semi-solid shell occurring with lower aerosol liquid water content (western United States and at higher altitudes) has a viscosity that was predicted to be 10 2 –10 12 Pa s, which resulted in organic mass being decreased due to diffusion limitation. Organic aerosol was primarily liquid where aerosol liquid water was dominant (eastern United States and at the surface), with a viscosity <10 2 Pa s. Phase separation while in a liquid phase state, i.e., liquid–liquid phase separation (LLPS), also reduces reactive uptake rates relative to homogeneous internally mixed liquid morphology but was lower than aerosols with a thick viscous organic shell. The sensitivity cases performed with different phase-separation parameterization and dissolution rate of isoprene epoxydiol (IEPOX) into the particle phase in CMAQ can have varying impact on fine particulate matter (PM 2.5 ) organic mass, in terms of bias and error compared to field data collected during the 2013 SOAS. This highlights the need to better constrain the parameters that govern phase state and morphology of SOA, as well as expand mechanistic representation of multiphase chemistry for non-IEPOX SOA formation in models aided by novel experimental insights.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2025

Accelerating eigenvalue computation for nuclear structure calculations via perturbative corrections

Subspace projection methods utilizing perturbative corrections have been proposed for computing the lowest few eigenvalues and corresponding eigenvectors of large Hamiltonian matrices. In this paper, we build upon these methods and introduce the term Subspace Projection with Perturbative Corrections (SPPC) method to refer to this approach. We tailor the SPPC for nuclear many-body Hamiltonians represented in a truncated configuration interaction subspace, i.e., the no-core shell model (NCSM). We use the hierarchical structure of the NCSM Hamiltonian to partition the Hamiltonian as the sum of two matrices. The first matrix corresponds to the Hamiltonian represented in a small configuration space, whereas the second is viewed as the perturbation to the first matrix. Eigenvalues and eigenvectors of the first matrix can be computed efficiently. Because of the split, perturbative corrections to the eigenvectors of the first matrix can be obtained efficiently from the solutions of a sequence of linear systems of equations defined in the small configuration space. These correction vectors can be combined with the approximate eigenvectors of the first matrix to construct a subspace from which more accurate approximations of the desired eigenpairs can be obtained. We show by numerical examples that the SPPC method can be more efficient than conventional iterative methods for solving large-scale eigenvalue problems such as the Lanczos, block Lanczos and the locally optimal block preconditioned conjugate gradient (LOBPCG) method. The method can also be combined with other methods to avoid convergence stagnation.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗
DOE OSTI2026

Enhancing segmentation fairness through curriculum learning and progressive loss: a centralized and federated perspective on radiograph analysis

Bias in medical image segmentation can lead to unequal performance across demographic subgroups, raising concerns about fairness and reliability in clinical AI systems. While deep learning models have achieved high segmentation accuracy, ensuring equitable performance across race and gender remains a significant challenge, particularly in privacy-sensitive healthcare environments. This study investigates fairness-aware medical image segmentation for hip and knee radiographs using deep learning models evaluated in both centralized and Federated Learning (FL) settings. We introduce Curriculum Learning (CL) strategies and Progressive Loss (PL) functions to regulate sample difficulty during training. In addition, we propose two novel fairness-oriented federated learning algorithms, Federated Intersection over Union (FedIoU) and Federated Intersection over Union with Outlier Analysis (FedIoUoutlier). Experiments are conducted using multiple segmentation backbones and simulated multi-site data partitions derived from the Osteoarthritis Initiative dataset. Model performance is evaluated using Intersection over Union (IoU), IoU standard deviation, Skewed Error Ratio (SER), and Min-Max Disparity across race and gender subgroups. Statistical significance was verified using paired t-tests to compare per-sample IoU performance against baseline configurations. Across both hip and knee segmentation tasks, curriculum learning and progressive loss strategies consistently improved segmentation accuracy and reduced demographic performance disparities in centralized training. In federated settings, fairness-aware aggregation further enhanced performance. Notably, FedIoUoutlier combined with balanced curriculum learning and tiered progressive loss achieved the highest mean IoU while yielding the lowest SER and Min-Max Disparity, indicating improved fairness without sacrificing accuracy. In several configurations, federated models matched or exceeded the performance of optimized centralized models, with statistically significant improvements in per-sample IoU over baseline configurations. The results demonstrate that structured training strategies and fairness-aware federated aggregation can jointly improve accuracy, stability, and demographic fairness in medical image segmentation. By integrating curriculum learning, progressive loss, and novel FL algorithms, this work provides a practical pathway toward equitable and privacy-preserving AI systems for medical imaging.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2022

Global Dust Cycle and Direct Radiative Effect in E3SM Version 1: Impact of Increasing Model Resolution

Quantification of dust aerosols in Earth System Models (ESMs) has important implications for water cycle and biogeochemistry studies. This study examines the global life cycle and direct radiative effects (DREs) of dust in the U.S. Department of Energy's Energy Exascale Earth System Model version 1 (E3SMv1), and the impact of increasing model resolution both horizontally and vertically. The default 1° E3SMv1 captures the spatial and temporal variability in the observed dust aerosol optical depth (DAOD) reasonably well, but overpredicts dust absorption in the shortwave (SW). Simulations underestimate the dust vertical and long-range transport, compared with the satellite dust extinction profiles. After updating dust refractive indices and correcting for a bias in partitioning size-segregated emissions, both SW cooling and longwave (LW) warming of dust simulated by E3SMv1 are increased and agree better with other recent studies. The estimated net dust DRE of –0.42 Wm –2 represents a stronger cooling effect than the observationally based estimate –0.2 Wm –2 (–0.48 to +0.2), due to a smaller LW warming. Constrained by a global mean DAOD, model sensitivity studies of increasing horizontal and vertical resolution show strong influences on the simulated global dust burden and lifetime primarily through the change of dust dry deposition rate; there are also remarkable differences in simulated spatial distributions of DAOD, DRE, and deposition fluxes. Thus, constraining the global DAOD is insufficient for accurate representation of dust climate effects, especially in transitioning to higher- or variable-resolution ESMs. Better observational constraints of dust vertical profiles, dry deposition, size, and LW properties are needed.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2022

m-CUBES An efficient and portable implementation of multi-dimensional integration for gpus

The task of multi-dimensional numerical integration is frequently encountered in physics and other scientific fields, e.g., in modeling the effects of systematic uncertainties in physical systems and in Bayesian parameter estimation. Multi-dimensional integration is often time-prohibitive on CPUs. Efficient implementation on many-core architectures is challenging as the workload across the integration space cannot be predicted a priori. We propose m-Cubes, a novel implementation of the well-known Vegas algorithm for execution on GPUs. Vegas transforms integration variables followed by calculation of a Monte Carlo integral estimate using adaptive partitioning of the resulting space. m-Cubes improves performance on GPUs by maintaining relatively uniform workload across the processors. As a result, our optimized Cuda implementation for Nvidia GPUs outperforms parallelization approaches proposed in past literature. We further demonstrate the efficiency of m-Cubes by evaluating a six-dimensional integral from a cosmology application, achieving significant speedup and greater precision than the CUBA library's CPU implementation of VEGAS. We also evaluate m-Cubes on a standard integrand test suite. m-Cubes outperforms the serial implementations of the Cuba and GSL libraries by orders of magnitude speedup while maintaining comparable accuracy. Our approach yields a speedup of at least 10 when compared against publicly available Monte Carlo based GPU implementations. In summary, m-Cubes can solve integrals that are prohibitively expensive using standard libraries and custom implementations. A modern C++ interface header-only implementation makes m-Cubes portable, allowing its utilization in complicated pipelines with easy to define stateful integrals. Compatibility with non-Nvidia GPUs is achieved with our initial implementation of m-Cubes using the Kokkos framework.

Sakiotis, Ioannis↗
DOE OSTI2020

Ion separations with membranes

Ion separations are important for resource recovery, water treatment, and energy production and storage. Techniques such as chemical precipitation, selective adsorption, and solvent extraction are effective, but membranes may separate ions continuously with less waste and lower energy costs. Separation of monovalent and multivalent ions with nanofiltration or electrodialysis membranes already enables water softening and edible salt purification. Similar membranes are attractive as separators in vanadium redox flow batteries. Selective partitioning of divalent counter-ions into ion-exchange membranes even allows transport of these ions against their concentration gradients in salt mixtures. However, separations of ions with the same charge is more challenging. Recent research demonstrated highly selective ion “sieving” at small scales. Here, separations using electrical potentials and differences in ion electrophoretic mobilities are promising, but relatively unexplored. Carrier-mediated transport affords high selectivity in liquid membranes, but these systems are not very stable, and selective transport via hopping between anchored carriers has proven elusive. Finally, this paper discusses how concentration polarization decreases selectivities in many membrane processes. Although development of selective, inexpensive ion-separation membranes is a work in progress, successes in water softening and edible salt purification suggests that future selective membranes will serve as complementary methods to traditional purification techniques.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2026

A hybrid Penman-Monteith and machine learning model for simulating evapotranspiration and its components

Integrating physical processes with machine learning has advanced evapotranspiration (ET) simulation, yet most hybrid models fail to partition total ET into its components: soil evaporation (E) and vegetation transpiration (T). This study introduces Residual Neural Network–Penman–Monteith (RNN-PM), a novel hybrid dual-source ET model designed to overcome this limitation. The model synergizes the physically-based Penman–Monteith framework with three specialized residual neural networks trained to estimate key conductance parameters (canopy conductance, soil surface conductance, and aerodynamic conductance). Furthermore this explicit parameterization allows for the direct partitioning of total ET. Validation at National Ecological Observatory Network (NEON) flux sites using high-frequency partitioned E and T shows that RNN-PM reliably reproduces ET and the transpiration fraction (T/ET). For ET, the model achieves an average Kling–Gupta efficiency (KGE) of 0.89 and a root-mean-square error (RMSE) of 0.55 mm/day; for T/ET, the KGE is 0.87 with an RMSE of 0.06. Furthermore, RNN-PM demonstrates robust generalization, accurately simulating ET and its components well beyond the initial training dataset, even under extreme climatic conditions. This study extended the analysis by comparing the RNN-PM model with seven established dual-source ET models. The results indicate that RNN-PM outperforms both conventional machine learning models and purely physical process-based models in simulating ET components in most cases. Among the purely physical process-based dual-source models, those based on surface temperature decomposition showed improved performance as the leaf area index (LAI) decreased when evaluated against high-frequency ET component datasets. In contrast, the performance of conductance-based dual-source models declined with decreasing LAI. Although purely machine learning-based models can produce relatively accurate simulations of ET components, they often exhibit limited generalization capability, an issue that the RNN-PM model effectively overcomes. Ultimately, the RNN-PM model represents a significant advance in simulating ET components, offering a novel and scalable approach for improving the representation of land–atmosphere interactions in Earth system models.

54 ENVIRONMENTAL SCIENCES↗