Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “computer system benchmarking”

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 649 records · Page 36
All sourcesNASA NTRSDOE OSTI
DOE OSTI2021

A Feynman-Kac based numerical method for the exit time probability of a class of transport problems

The exit time probability, which gives the likelihood that an initial condition leaves a prescribed region of the phase space of a dynamical system at, or before, a given time, is arguably one of the most natural and important transport problems. In this work, we present an accurate and efficient numerical method for computing this probability for systems described by non-autonomous (time-dependent) stochastic differential equations (SDEs) or their equivalent Fokker-Planck partial differential equations. The method is based on the direct approximation of the Feynman-Kac formula that establishes a link between the adjoint Fokker-Planck equation and the forward SDE. The Feynman-Kac formula is approximated using the Gauss-Hermite quadrature rules and piecewise cubic Hermite interpolating polynomials, and a GPU accelerated matrix representation is used to compute the entire time evolution of the exit time probability using a single pass of the algorithm. The method is unconditionally stable, exhibits second order convergence in space, first order convergence in time, and it is straightforward to parallelize. Applications are presented to the advection diffusion of a passive tracer in a fluid flow exhibiting chaotic advection, and to the runaway acceleration of electrons in a plasma in the presence of an electric field, collisions, and radiation damping. Benchmarks against analytical solutions as well as comparisons with explicit and implicit finite difference standard methods for the adjoint Fokker-Planck equation are presented.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2025

Spin-informed universal graph neural networks for simulating magnetic ordering

The screening and discovery of magnetic materials are hindered by the computational cost of first-principles density-functional theory (DFT) calculations required to find the ground state magnetic ordering. Although universal machine-learning interatomic potentials (uMLIPs), also known as atomistic foundation models, offer high-fidelity models of many atomistic systems with significant speedup, they currently lack the inputs required for predicting magnetic ordering. In this work, we present a data-efficient, spin-informed graph neural network framework that incorporates spin degrees of freedom as inputs and preserves physical symmetries, extending the functionality of uMLIPs to simulate magnetic orderings. This framework speeds up DFT calculations through better initial guesses for magnetic moments, determines the ground-state ordering of bulk materials and even generalizes to magnetic ordering in surfaces. Furthermore, we implement a closed-loop anomaly detection approach that effectively addresses the classic "chicken-and-egg" problem of creating a high-quality dataset while developing a uMLIP, unearthing anomalies in large benchmark datasets and boosting model accuracy.

Xu, Wenbin↗
DOE OSTI2020

Progress Toward Simulating Departure from Nucleate Boiling at High-Pressure Applications with Selected Wall Boiling Closures

Recently, a Eulerian-based two-fluid computational fluid dynamics (CFD) framework with a wall heat flux partitioning approach has been intensively investigated for departure from nucleate boiling (DNB) simulation under the U.S. Department of Energy–funded Consortium for Advanced Simulation of Light Water Reactors (CASL) program. Understanding of the DNB characteristics over a range of pressurized water reactor–like operating conditions and accurate prediction of boiling crisis in the nuclear power system have been grand challenges because of the large impact of DNB on reactor safety and operational economics. The ultimate goal of this task in the CASL program is to introduce a robust multiphase CFD–based DNB modeling framework that is capable of characterizing an entire boiling history in which the wall boiling mode experiences the following through multiple stages of heat transfer mode: (1) single-phase convective heat transfer, (2) nucleate boiling heat transfer, and (3) identification of the departure of nucleate boiling. To validate the CASL boiling model, we have benchmarked simulated DNB over three different flow channel configurations (pipe flow, 5 × 5 fuel bundle with mixing vane tests, and 5 × 5 fuel bundle without mixing vane tests) against experimental measurements, and the validation result with open literature is reported. The DNB detection criteria in the simulation are checked by monitoring the peak wall temperature, wall dryout factor, and net energy balance. In addition to the DNB performance test, some preliminary sensitivity results on closure model selection are reported to address the prediction capability of local void profile against measurements. The boiling simulation tested in this study exhibits a maximum deviation of 24% from the measured DNB value in a high-pressure (i.e., 138 bars) subcooled pipe flow test. The ranges of operating conditions are as follows: 1650 to 2650 kg/m 2 ·s for mass flux and 8.5 to 96 K for subcooled inlet temperature. The deviation is even reduced to 7% when the subcooled temperature is less than 40 K. Besides accuracy, base practice guidelines for DNB detection criteria are tested by monitoring three simulation variables: (1) maximum wall temperature, (2) wall dryout factor (i.e., K-value), and (3) energy balance. Numerical robustness of DNB simulation is largely achieved in most of the validation test except for a few high subcooled test cases.

42 ENGINEERING↗
DOE OSTI2020

Evolution of the ROOT Tree I/O

The ROOT TTree data format encodes hundreds of petabytes of High Energy and Nuclear Physics events. Its columnar layout drives rapid analyses, as only those parts (“branches”) that are really used in a given analysis need to be read from storage. Its unique feature is the seamless C++ integration, which allows users to directly store their event classes without explicitly defining data schemas. In this contribution, we present the status and plans of the future ROOT 7 event I/O. Along with the ROOT 7 interface modernization, we aim for robust, where possible compile-time safe C++ interfaces to read and write event data. On the performance side, we show first benchmarks using ROOT’s new experimental I/O subsystem that combines the best of TTrees with recent advances in columnar data formats. A core ingredient is a strong separation of the high-level logical data layout (C++ classes) from the low-level physical data layout (storage backed nested vectors of simple types). We show how the new, optimized physical data layout speeds up serialization and deserialization and facilitates parallel, vectorized and bulk operations. This lets ROOT I/O run optimally on the upcoming ultra-fast NVRAM storage devices, as well as file-less storage systems such as object stores.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2024

Stochastic finite volume method for uncertainty quantification of transient flow in gas pipeline networks

We develop a weakly intrusive framework to simulate the propagation of uncertainty in solutions of generic hyperbolic partial differential equation systems on graph-connected domains with nodal coupling and boundary conditions. The method is based on the Stochastic Finite Volume (SFV) approach and can be applied for uncertainty quantification (UQ) of the dynamical state of fluid flow over actuated transport networks. The numerical scheme has specific advantages for modeling intertemporal uncertainty in time-varying boundary parameters, which cannot be characterized by strict upper and lower (interval) bounds. We describe the scheme for a single pipe, and then formulate the controlled junction Riemann problem (JRP) that enables the extension to general network structures. In conclusion, we demonstrate the method's capabilities and performance characteristics using a standard benchmark test network.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2026

In situ quantum verification of polarization-stabilized optical channels

The active stabilization of polarization channels is a task of growing importance as quantum networks move to deployed demonstrations over existing fiber infrastructure. However, the uniquely strict requirements for high-fidelity qubit transmission complicate the extent to which classical solutions may apply to future quantum networks, particularly in terms of recognizing noise sources present in low-flux, nonunitary channels. Here we introduce an in situ benchmarking approach that augments a classical polarization tracking system, limited to unitary correction, with simultaneously transmitted quantum light for ancilla-assisted process tomography of the full quantum map. Implemented in a quantum local-area network, our method uses the reconstructed map both to validate the classical compensation and to expose noise sources it fails to capture. A sliding measurement window that continuously updates the estimated quantum process further increases sensitivity to rapid channel fluctuations. Our results should unlock new opportunities for in situ channel characterization in quantum-classical coexistence networks.

Stevens, Matthew L [Arizona State University]↗
DOE OSTI2020

Local Phasor-Based Control of DER Inverters for Voltage Regulation on Distribution Feeders

We introduce a new control paradigm termed Phasor-Based Control (PBC) to coordinate Distributed Energy Resources (DER) for improved voltage regulation and other objectives. PBC is a multi-layer control framework where the state of the distribution grid is represented by voltage phasors, and Phasor Measurement Units (PMUs) measure the state at critical points in the distribution grid. In the PBC paradigm, a supervisory controller (S-PBC) sets phasor targets which local controllers track. In this paper, we propose a Proportional Integral (PI) implementation for the local control layer (LPBC), to regulate real and reactive power output of distributed inverters to track a voltage phasor target computed by the supervisory layer. We tune the PI controller gains offline with a genetic algorithm that yields better performance than the Ziegler-Nichols method. We benchmark the proposed L-PBC controller against droop volt-var control (DVVC) and observe improved voltage regulation in simulation on the IEEE 13 node unbalanced test feeder (IEEE 13NF). Furthermore, by means of numerical simulation we observe the response of the L-PBC controllers to small and large grid disturbances. Simulations on larger feeders, IEEE 123-node and 647-node, demonstrate challenges with tuning controllers in setups with different amounts of power-voltage coupling.

42 ENGINEERING↗
DOE OSTI2021

HIPPO – A Software Platform for Electricity Market Research and Development

The goal of this project is to provide Regional transmission organizations (RTOs) and independent system operators (ISOs) a market design and prototyping software, High-Performance Power-Grid Optimization (HIPPO), that they can evaluate electricity market design options, calculate market planning strategies and operational performance. With the high standards and strict reliability requirements for operating power systems, impacts of new technologies need to be fully investigated prior to any consideration for adoption. A market design and prototyping software tool which can be used to prototype electricity market design options, to calculate market planning strategies and operational performance with high precision, and to investigate the impacts for integrating future power grid technologies will be valuable to RTOs/ISOs who operate power systems, to vendors like GE and ABB who provide the market solvers, and to market participants and researchers who are actively doing market research. HIPPO is a such tool that can be used to improve the current market operations and provide capabilities for rigorous forward-looking design and prototyping of next-generation energy markets. HIPPO has a high-resolution model for the day-ahead SCUC, which was validated with MISO and GE-Grid Solutions. HIPPO is built with parallel and distributed computing capabilities and can be executed in both multi-thread and high-performance computing (HPC) settings. This capability provides fast solution speed necessary to handle the larger and more complex SCUC problems of real-world cases and the potentially growing size and complexity of future scenarios. In addition, HIPPO has a concurrent optimizer (CO) which manages multiple algorithm executions simultaneously and leverages the advantages from different algorithms. This structure provides flexibility to better benchmark competing approaches. Highly accurate market model, fast solution technologies and flexible model and algorithm control are the features which will make HIPPO an extensible platform for developing and testing multiple approaches to meet a wide range of future market needs.

24 POWER TRANSMISSION AND DISTRIBUTION↗
DOE OSTI2023

TransformerG2G: Adaptive time-stepping for learning temporal graph embeddings using transformers

Dynamic graph embedding has emerged as a very effective technique for addressing diverse temporal graph analytic tasks (i.e., link prediction, node classification, recommender systems, anomaly detection, and graph generation) in various applications. Such temporal graphs exhibit heterogeneous transient dynamics, varying time intervals, and highly evolving node features throughout their evolution. Hence, incorporating long-range dependencies from the historical graph context plays a crucial role in accurately learning their temporal dynamics. In this paper, we develop a graph embedding model with uncertainty quantification, TransformerG2G, by exploiting the advanced transformer encoder to first learn intermediate node representations from its current state (t) and previous context (over timestamps [t–1,t–l], l is the length of context). Moreover, we employ two projection layers to generate lower-dimensional multivariate Gaussian distributions as each node's latent embedding at timestamp t. We consider diverse benchmarks with varying levels of "novelty" as measured by the TEA (Temporal Edge Appearance) plots. Here, our experiments demonstrate that the proposed TransformerG2G model outperforms conventional multi-step methods and our prior work (DynG2G) in terms of both link prediction accuracy and computational efficiency, especially for high degree of novelty. Furthermore, the learned time-dependent attention weights across multiple graph snapshots reveal the development of an automatic adaptive time stepping enabled by the transformer. Importantly, by examining the attention weights, we can uncover temporal dependencies, identify influential elements, and gain insights into the complex interactions within the graph structure. For example, we identified a strong correlation between attention weights and node degree at the various stages of the graph topology evolution.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2023

Computing the Properties of Matter with Leadership Computing Resources (Closeout Report for DE-SC0018121)

In order to add more capabilities to Halide, we have designed a new framework called Tiramisu and integrated this framework into Halide. Since Tiramisu enables Halide to target heterogeneous architectures, our development efforts have been refocused on Tiramisu. Most high-performance computer systems today are complex and increasingly heterogeneous; they may have CPUs, GPUs and FPGAs. Achieving best performance requires taking full advantage of all these different architectures. To address this issue, we have designed Tiramisu, an optimization framework that enables Halide (and other DSLs) to target heterogeneous architectures. Tiramisu is an optimization framework that takes as input a high level, architecture-independent representation of code and a set of scheduling and data mapping commands that guide code transformation. The input can either be generated by a domain-specific language (DSL) compiler such as Halide or directly written by a programmer. Tiramisu then applies the user-specified code and data-layout transformations and generates an architecture-specific, low-level intermediate representation (IR) that takes advantage of modern architectural features such as multicore parallelism, non-uniform memory (NUMA) hierarchies, clusters, and accelerators like GPUs and FPGAs. We integrated Tiramisu within Halide and implemented a representative set of benchmarks to evaluate this integration. Tiramisu is now open source and is available for public use (http://tiramisu-compiler.org/). A paper about Tiramisu was published, it shows that Tiramisu extends Halide with many new capabilities and that Tiramisu can generate efficient code for multicores, GPUs, FPGAs and distributed heterogeneous systems. The performance of code generated by the Tiramisu backends matches or exceeds hand optimized reference implementations. For example, the multicore backend matches the highly optimized Intel MKL library on many kernels and shows speedups reaching 4x over the original Halide. In addition to making Tiramisu more robust, we have used Tiramisu to implement a set of representative tensor operation for constructing baryon building blocks required for multi baryon contractions in LQCD. In order to implement this code, we needed to generalize Tiramisu in two ways: first we needed to support indirect array accesses, and second, we needed to add support for complex numbers to Tiramisu. The code generated by Tiramisu is 6x faster than the reference code. Our efforts towards an MPI based multi-node version of tiramisu have matured and the resulting code scales well on multiple nodes (tests up to 512 KNL nodes have been undertaken).

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗
DOE OSTI2025

Edge ML for CAN bus intrusion detection in AVs

Autonomous Vehicles (AVs) are revolutionizing transportation, but their reliance on interconnected cyber-physical systems exposes them to unprecedented cybersecurity risks. This study addresses the critical challenge of detecting real-time cyber intrusions in self-driving vehicles by leveraging a dataset from the Udacity self-driving car project. We simulate four high-impact attack vectors, Denial of Service (DoS), spoofing, replay, and fuzzy attacks, by injecting noise into spatial features (e.g., bounding box coordinates) to replicate adversarial scenarios. We develop and evaluate two lightweight neural network architectures (NN-1 and NN-2) alongside a logistic regression baseline (LG-1) for intrusion detection. The models achieve exceptional performance, with NN-2 attaining an AUC score of 93.15% and 93.15% accuracy, demonstrating their suitability for edge deployment in AV environments. Through explainable AI techniques, we uncover unique forensic fingerprints of each attack type, such as spatial corruption in fuzzy attacks and temporal anomalies in replay attacks, offering actionable insights for feature engineering and proactive defense. Visual analytics, including confusion matrices, ROC curves, and feature importance plots, validate the models' robustness and interpretability. This research sets a new benchmark for AV cybersecurity, delivering a scalable, field-ready toolkit for Original Equipment Manufacturers (OEMs) and policymakers. By aligning intrusion fingerprints with SAE J3061 automotive security standards, we provide a pathway for integrating machine learning into safety-critical AV systems. Our findings underscore the urgent need for security-by-design AI, ensuring that AVs not only drive autonomously but also defend autonomously. This work bridges the gap between theoretical cybersecurity and life-preserving engineering, offering a leap toward safer, more secure autonomous transportation.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2024

Summary Report of the FY24 DOE Contributions to the GIF VHTR CMVB

The Generation-IV Forum (GIF) Very-High-Temperature Reactor-Computational Methods Validation and Benchmark (VHTR-CMVB) initiative, involving organizations from Korea Atomic Energy Research Institute (KAERI) (South Korea), Institute of Nuclear and New Energy Technology of Tsinghua University (INET) (China), U.S. Department of Energy (DOE) (U.S.), Joint Research Centre (JRC) (Europe), and Japan Atomic Energy Agency (JAEA) (Japan), is dedicated to the verification and validation of tools for High-Temperature Gas-Cooled Reactors (HTGRs) analysis, using data shared by Computational Methods Validation and Benchmark (CMVB) signatories. For FY24, the US DOE CMVB has committed to several critical activities. Under WP1, led by the US, the integration of the High Temperature Gas Cooled Reactor - Pebble-Bed Module (HTR-PM) Phenomena Identification and Ranking Table (PIRT) into the comparative PIRT is progressing, with a new draft of the comparison tables issued earlier this year and currently being utilized by INET for their contribution. Neutronic validation efforts under WP3 include the preparation of the burnup analysis benchmark, preliminary calculations, and the development of reference models and results. In WP2, a validation exercise for hot gas mixing in the lower plenum of HTR-PM is in progress, using experimental data from INET (China) to validate modeling approaches. A model of the experimental facility has been developed using StarCCM+, with initial calculations slated for presentation at the GIF CMVB meeting this fall. Another WP2 activity focuses on validating numerical models for air-cooled Reactor Cavity Cooling System (RCCS) with experimental data from the Wisconsin Madison RCCS facility. A high-fidelity model, developed using NEK-RS, is currently being validated with available data from a low power forced convection test. These efforts are aimed at enhancing and confirming the accuracy of HTGR analysis tools, ensuring their alignment with experimental data and regulatory requirements.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗
DOE OSTI2022

Analysis of SCALE Criticality and Sensitivity Calculations for Reflected HEU Cylinders [Slides]

The SCALE code package offers multiple nuclear data libraries supporting Monte Carlo transport, with sensitivity and uncertainty methods derived from MC transport solutions. Several libraries are multigroup, which introduce bias differing by system. Previous work has shown poor S/U results in several reflector materials: ICSBEP benchmark HMF-084 was selected to analyze biases and S/U method applicability to a variety of reflectors. Prior and ongoing work found inaccuracies in CSAS and TSUNAMI results, which were further investigated utilizing the HEU-MET-FAST-084 ICSBEP critical benchmark, chosen for its geometrical simplicity and variety of reflector materials. Perturbation of reflector thickness across various reflector materials allowed for an assortment of materials is to be tested swiftly for each sequence and method. Observation was an increasing bias of MG $k_{eff}$ relative to CE, in both direction and magnitude. IFP produced extremely reliable results. >85% of CLUTCH cases were found satisfactory.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗
DOE OSTI2024

Expanding the representation of aerosol, cloud, and precipitation processes with graph network-based simulators

We explored a novel framework for simulating the small-scale processes that drive the evolution of aerosol, cloud, and precipitation particles, which are a critical gap in the predictive understanding of weather and climate. Particle-based methods have emerged as an effective tool for modeling aerosol-cloud-precipitation interactions, but existing particle-based models are computationally too expensive to simulate the large domains relevant for the atmosphere or to represent the full suite of relevant processes. The lack of a comprehensive and efficient reference model is a critical bottleneck in our understanding of cloud and precipitation processes and our ability to parameterize these processes for regional- and global-scale simulations. To address this need, we explored an approach to accelerate and expand particle-based models using a new machine learning approach, graph network-based simulators (GNS). Rather than modeling the evolution of the system by numerically integrating continuity equations, the GNS represents dynamics through learned message passing. Our aim was to develop fast and accurate surrogate models for particle-based simulations. We explored applying GNS to simulate cloud droplet transport, growth, and evaporation under turbulent conditions, but we found the GNS over-smoothed the simulations. We then applied the GNS to simulate aerosol dynamics through gas condensation and found the GNS was able to reproduce the benchmark, physics-based simulation with high accuracy.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2024

Integrated Steady-State System Package for Nuclear Thermal Propulsion Analysis Using Multi-Dimensional Thermal Hydraulics and Dimensionless Turbopump Treatment

Nuclear thermal propulsion is an evolving technology that can be utilized for long-distance space travel. This technology yields the advantage of a high thrust and specific impulse, but requires an examination of the potential design adjustments necessary to enhance its feasibility. The development of nuclear thermal propulsion requires a comprehensive understanding of the system-level behavior during transient and steady-state operation. This paper extends our previous research by including the proper handling of turbomachinery with multi-channel thermal hydraulic simulations only for steady-state solutions. The system-level approach presented here enables the treatment of the turbopump components through non-dimensional analysis that eliminates the assumption of constant efficiencies. All the other components within the system (e.g., reflector and core) can be discretized to multiple channels and layers, in which the full thermal hydraulic solution is established. The approach chosen here enables the realistic modeling of the propellant flow within the expander cycle by capturing the pressure losses, mass flow rate splits, and enthalpy gain for various operational conditions. The verification of the package is completed through point comparisons of previous investigations into similar system designs. Furthermore, sensitivity studies are used to benchmark the capabilities of the package and investigate solution variations due to the perturbation of operational conditions and regimes. The sensitivity studies performed here are important to capture variation in flow characteristics (e.g., temperature, pressure, mass flow rates) for different design objectives such as the thrust and specific impulse. This work demonstrates that system-level simulations lacking multi-channel capability and proper turbomachinery treatment may yield higher uncertainties in understanding the engine’s response and characteristics to changing various requirements. This is extremely important when screening the design space of such propulsion systems and when transient simulations are required.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗
DOE OSTI2025

Development of a River Dynamical Core for E3SM to simulate compound flooding on Exascale-class heterogeneous supercomputers

Flooding events pose significant risk to human life, property, and infrastructure. Physically-consistent quantification of altered flood risks in global models requires hyper-resolution (~1 km) or fine flood simulations using two-dimensional (2D) physics schemes, both of which are unavailable in the current generation Earth System Models. Here, in this work, we have developed the River Dynamical Core (RDycore), which is an open-source, 2D shallow water equation (SWE) library for the U.S. Department of Energy's Energy Exascale Earth System Model (E3SM). RDycore uses PETSc and libCEED libraries that allows it to run efficiently on CPUs and GPUs, as well as select a time-integration algorithm at runtime without requiring any code modifications. RDycore achieves spatial error convergence rates for problems with analytical and manufactured solutions similar to those reported previously in the literature, or consistent with the implemented first-order spatial discretization scheme. RDycore's accuracy in predicting flooding for a well-studied dam break problem is comparable to existing SWE models. For a problem with 471 million grid cells, RDycore achieves a speedup of 6.6x and 7.6x on GPUs compared to CPUs when using 320 compute nodes on DOE's Perlmutter and Frontier supercomputers, respectively. The one-way coupling of the RDycore library within E3SM is demonstrated by performing multiple 5-day flooding simulations during Hurricane Harvey driven by five precipitation datasets. The E3SM--RDycore simulations at 30 m spatial resolution accurately simulate maximum water height during the hurricane when benchmarked against a previously published study and achieve a speedup of 15x (Perlmutter) and 21x (Frontier) on GPUs relative to CPUs. The work presented here is the foundational step in providing hardware and algorithmic portability framework for simulating kilometer-scale river dynamics within E3SM.

Flood Simulation↗
DOE OSTI2026

Modeling the behavior of concentrated aqueous HNO 3 using machine learning interatomic potentials

We develop two multi-defect machine learning interatomic potentials (MLIPs) trained at the BLYP-D2 and PBE-D3 density functional theories using the DeepMD-kit, allowing for the investigation of structural and thermodynamic properties of nitric acid over a wide range of concentrations via molecular dynamics (MD) simulations. We directly compute the degree of dissociation, α, and pK a from MD simulations, revealing that HNO 3 behaves as a weaker acid at higher concentrations, noting that our standard-state pK a value is in excellent agreement with the experimental one. In general, good agreement is observed with experimental results such as α and density outside the training dataset, with only modest deviations at low-to-medium concentrations. We benchmark our custom multi-defect DeepMD MLIPs against foundational models MACE-MP0 and MACE-OFF23. The foundation models capture some aspects of HNO 3 /NO 3 − solvation in concentrated nitric acid but show noticeable density errors and miss subtle structural features relevant to spectroscopy, whereas the bespoke DeepMD MLIPs yield more compact solvation shells, reproduce density-concentration trends, and run ∼12–15× faster than MACE-MP0. Although classical FFs are still more efficient and match experimental densities better, they lack chemical reactivity and thus cannot predict α or pK a , underscoring the need for system-specific reactive MLIPs beyond universal MLIPs.

Dinpajooh, Mohammadhasan [Pacific Northwest Nation↗