SEARCH · Search NASA

Results for “simulation research”

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.

SourceResearch connectionFrom yearThrough year

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

At least 19 records

All sources NASA NTRS DOE OSTI

DOE OSTI2025

FAIR Surrogate Benchmarks Supporting AI and Simulation Research (Final Report)

Computational Science is being revolutionized by integrating AI and simulation and, in particular, by deep learning surrogate models that can replace all or part of traditional large‐scale HPC computations. Such surrogates can achieve remarkable performance improvements, as much as several orders of magnitude, and save both compute time and energy. The Surrogate Benchmark Initiative (SBI) project creates a community repository and FAIR (Findable, Accessible, Interoperable, and Reusable) data ecosystem for HPC application surrogate benchmarks. The SBI team comes from Argonne National Laboratory (ANL), Indiana University (IU), Rutgers University, the University of Tennessee, Knoxville (UTK), and the University of Virginia (UVA). SBI repositories include data, code, and all relevant collateral artifacts that the science and engineering community need to use and reuse these data sets and surrogates. SBI repositories generate active research from both the participants in SBI and the broad community of AI and domain scientists. This project develops surrogates that use several different neural nets to learn and quickly infer the results of simulations and data systems and captures them as surrogate benchmarks with a rich set of metadata covering: Data; Model; Metrics specification; Machine specification; and Science, Speed, and Power Results. We research FAIR metadata for these benchmarks. We develop application surrogate examples as benchmarks across many fields (ANL, UTK, IU, UVA). We also study non-Surrogate benchmarks that have many common features and similar issues as regards FAIRness. We work with MLCommons (UVA, UTK), which is a major machine learning benchmarking activity where we get metadata ontologies, software, and benchmarks, Benchmarks have datasets, models, and metadata and they need a technical framework developed by UTK and Rutgers and deployed by UVA. We study features of Surrogates including performance, training set size, and uncertainty quantification (Rutgers, UVA and IU).

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2025

DOE FAIR Surrogate Benchmarks Supporting AI and Simulation Research (SBI Surrogate Benchmark Initiative) (Final Report)

Computational Science is being revolutionized by integrating AI and simulation and, in particular, by deep learning surrogate models that can replace all or part of traditional large‐scale HPC computations. Such surrogates can achieve remarkable performance improvements, as much as several orders of magnitude, and save both compute time and energy. The Surrogate Benchmark Initiative (SBI) project creates a community repository and FAIR (Findable, Accessible, Interoperable, and Reusable) data ecosystem for HPC application surrogate benchmarks. The SBI team comes from Argonne National Laboratory (ANL), Indiana University (IU), Rutgers University, the University of Tennessee, Knoxville (UTK), and the University of Virginia(UVA). SBI repositories include data, code, and all relevant collateral artifacts, that the science and engineering community needs to use and reuse these data sets and surrogates. SBI repositories generate active research from both participants in SBI and the broader AI and domain science communities. This project develops surrogates that use several different neural nets to learn and quickly infer the results of simulations and data systems and capture them as surrogate benchmarks with a rich set of metadata, covering. Data; Model; Metrics specification; Machine specification; Science, Speed, Power Results, We research FAIR metadata for these benchmarks. We develop application surrogate examples as benchmarks across many fields (ANL, UTK, IU, UVA). We also study non Surrogate benchmarks that have many common features and similar issues regarding FAIRness. We work with MLCommons (UVA, UTK), which is a major machine learning benchmarking activity where we get metadata ontologies, software, and benchmarks, benchmarks have datasets, models, and metadata, and they need a technical framework developed by UTK and Rutgers and deployed by UVA. We study features of Surrogates, including performance, training set size, and uncertainty quantification (Rutgers, UVA and IU).

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2023

Validating Connected, Automated, and Electric Vehicle Models and Simulation - Research Performance Progress Report

The objective of this project is to test connected and automated vehicles with both electrified and internal combustion engine powertrains to support updates and validation of modeling and simulation tools. This includes the development of the components and network architecture to execute and collect empirical data for multiple scenarios and traffic interactions. Specific program objectives include: • Translate Lab algorithms into vehicle and infrastructure controls • Conduct physical testing at realistic scale • Evaluate system performance • Improve models using empirical data • Improve control algorithms from lessons learned • Identify system and algorithm assumptions which need refinement It is important to note that the objective of this project was not to demonstrate the efficacy of the selected algorithms to improve energy efficiency but rather to validate and improve modeling and simulation tools using empirical data. While it is a desirable outcome to concurrently demonstrate improved energy efficiency through use of these algorithms, and in most cases that was the outcome, the success of this project was not predicated on the performance of the algorithm towards improving energy efficiency across all scenarios and test matrices.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

DOE OSTI2022

Evolutionary Game and Simulation Research of Blockchain-Based Co-Governance of Emergency Supply Allocation

Recently, with the spread of COVID-19 pandemic, emergency supply allocation system is drawing more and more social attention. Emergency supply allocation system is an important part of emergency governance system. It reflects social organizations’ credibility, public safety, and the modernization level of social governance. However, emergency supply allocation system still has some problems, such as information asymmetry, different desires of participants, unreasonable allocation, and so on. At present, it is widely accepted that the advantage of blockchain in co-governance could be of great help in solving above problems. And in order to distinguish the effect of blockchain to the emergency supply allocation, the paper builds a tripartite evolutionary game model among the government, social organizations, and the public to analyse the impact of blockchain platform on emergency supply allocation. The simulation analysis shows the following: (1) The strategy choices of the government have a crucial impact on the evolution and stability of social organizations’ strategies. It needs a long-term process to guide social organizations practicing active allocation, and the government should accelerate to build the blockchain platform to promote this process. (2) With the help of blockchain platform, the increment of penalty intensity of the government is conductive to increasing the probabilities that social organizations practice active allocation and the government practices strict supervision. (3) Blockchain platform has a significant impact on social organizations’ choice for active allocation in many aspects, such as the positive and negative effects of social organizations, effect’s increasing multiple, and the cost of the public informing. In the end, some suggestions are presented to improve the co-governance of emergency supply allocation.

Zhao, Huawei↗

DOE OSTI2020

Thermal chains and entrainment in cumulus updrafts, Part 2: Analysis of idealized simulations

Research has suggested that the structure of deep convection often consists of a series of rising thermals, or “thermal chain”, which contrasts with existing conceptual models that are used to construct cumulus parameterizations. In this work, simplified theoretical expressions for updraft properties obtained in Part 1 of this study are used to develop a hypothesis explaining why this structure occurs. In this hypothesis, cumulus updraft structure is strongly influenced by organized entrainment below the updraft’s vertical velocity maximum. In a dry environment, this enhanced entrainment can locally reduce condensation rates and increase evaporation, thus eroding buoyancy. For moderate-to-large initial cloud radius R, this breaks up the updraft into a succession of discrete pulses of rising motion (i.e., a thermal chain). For small R, this leads to the structure of a single, isolated rising thermal. In contrast, moist environments are hypothesized to favor plume-like updrafts for moderate-to-large R. In a series of axisymmetric numerical cloud simulations, R and environmental relative humidity (RH) are systematically varied to test this hypothesis. Vertical profiles of fractional entrainment rate, passive tracer concentration, buoyancy, and vertical velocity from these runs agree well with vertical profiles calculated from the theoretical expressions in Part 1. Analysis of the simulations supports the hypothesized dependency of updraft structure on R and RH, that is, whether it consists of an isolated thermal, a thermal chain, or a plume, and the role of organized entrainment in driving this dependency. Additional 3-dimensional (3-D) turbulent cloud simulations are analyzed, and the behavior of these 3-D runs is qualitatively consistent with the theoretical expressions and axisymmetric simulations.

54 ENVIRONMENTAL SCIENCES↗

DOE OSTI2024

Low Purge and Mercury Recovery Testing with Sludge Batch 10 Tank 40 Simulant

Researchers at the Savannah River National Laboratory (SRNL) have completed testing as requested by Savannah River Mission Completion (SRMC) to perform experiments to determine the impact of using a lower or inert purge in Sludge Batch (SB) 10 processing under the Nitric-Glycolic Acid (NGA) flowsheet. A key objective of this testing was also to determine the mercury speciation and recovery during each experiment. The testing was performed as part of the SB10 Technical Task Request (TTR) and Task Technical and Quality Assurance Plan (TTQAP). Two sets of tests were performed, and a Run Plan was approved prior to each set of experiments to document the planned testing. Three initial experiments were completed to determine whether a low air purge would be beneficial to CPC processing at higher acid stoichiometry (110%) based on the Koopman minimum acid (KMA) equation (116% Hsu). One of the tests, an inert nitrogen purge experiment, was also completed to demonstrate that excluding oxygen did not introduce any new hazards. The experiments were designed to be identical except for the change in purge gas and purge flowrate from run to run. After reviewing the results from the initial three experiments, six additional tests were proposed by SRNL to support the lower purge study and to look for processing alternatives for improving mercury recovery. These additional six tests were all completed at a very low acid stoichiometry to mimic the pH experienced during processing in DWPF (~7). DWPF is processing SB10 sludge at an acid stoichiometry of 90% based on the Hsu equation. The additional SRNL experiments were performed at an acid stoichiometry of 62.5% KMA stoichiometry (66.3% Hsu) to produce a Sludge Receipt and Adjustment Tank (SRAT) product with a pH of about 7. All experiments used simulants of both SWPF streams, although no entrained solvent was added during any of the experiments

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗

DOE OSTI2025

Closing the Accuracy Gap in Tandem Photovoltaic Testing: An Accessible and Efficient Spectral Tuning Method Using LED-Based Simulators for Research Laboratories and Industry

Accurate performance calibration of multijunction (MJ) solar cells is critical for advancing this technology toward large-scale terrestrial application, yet existing testing methods developed by photovoltaics calibration laboratories remain prohibitively complex and resource intensive for most research laboratories. Current approaches rely on expensive multisource simulators and/or intricate spectral tuning algorithms, limiting accessibility and hindering standardized comparisons of emerging MJ technologies such as perovskite-based tandems. This paper introduces a streamlined spectral irradiance adjustment method for light-emitting diode (LED)-based solar simulators, which are increasingly adopted in the photovoltaics community due to their versatility and cost-effectiveness. The method we present bins LED channels into groups based on the number of junctions of the MJ photovoltaic device under test (DUT) and their corresponding band gaps and incorporates an automated tuning algorithm that eliminates the need to adjust each channel’s irradiance individually. This tuning algorithm requires the relative spectral responsivities of both the DUT and a broadband reference cell, as well as a calibrated spectroradiometer. Measurement validation across various MJ devices, including III-V and perovskite/silicon tandems, demonstrates excellent agreement within 1% with well-established xenon-tungsten multisource simulators and associated tuning algorithms. By enabling precise spectral tuning with readily available equipment and a simple tuning algorithm, our approach bridges the measurement accuracy gap between research laboratories and accredited testing facilities, fostering more reliable device comparisons and accelerating the translation of MJ technologies into real-world energy systems.

14 SOLAR ENERGY↗

DOE OSTI2022

Advanced modeling and simulation of research reactors using dynamic mode decomposition

Full text of publication follows. Due to the ever-increasing safety requirements, the current trend of nuclear reactor analysis is shifting towards high-fidelity multi-physics models, which have a very high computational cost and modelling complexity. As the cost of even a single model run makes it impossible to analyse the behaviour and performance of these models on large-scale commercial plants, it has become even more significant to provide suitable benchmarks to validate and test them extensively. In this sense, research reactors offer a promising solution for the initial validation of high-fidelity models, as they are significantly smaller than commercial reactors and their characteristics are well known. In particular, the reactors of the TRIGA family have been used to assess and validate models and methods for Generation-IV designs, as they have some similar features (such as the dominance of natural convection as cooling mechanism and the difficulties in performing sub-channel analysis using standard codes). Still, the computational requirements of high-fidelity models make them unsuitable for real-time analysis, even following their assessment on research reactors. In this sense, Model Order Reduction (MOR) techniques give an additional strategy to reduce the computational cost of high-fidelity models (whilst preserving sufficient accuracy). In particular, this work focuses on Dynamic Mode Decomposition (DMD), a non-intrusive MOR technique that aims at representing models with explicit temporal dynamics by extracting the time-varying characteristics and the governing structures based only on a set of available data, thus without needing any underlying knowledge of the governing equations. In addition, DMD also computes a low-dimensional surrogate of the dynamic matrix of the system, making it suited for stability analysis and real-time evaluations. This work focuses on the application and validation of the DMD method on the Computational Fluid-Dynamics (CFD) model TRIGA Mark II reactor, also discussing in detail the potentiality of this algorithm as an advanced modelling tool for nuclear reactor analysis. (author)

21 SPECIFIC NUCLEAR REACTORS AND ASSOCIATED PLANTS↗

DOE OSTI2022

Collaborative Research: Particle Simulation of Collisionless Magnetic Reconnection Under Finite Guide Field

Magnetic reconnection is believed to be one of the most fundamental physical processes in laboratory and space plasmas. Fundamental physics of collisionless reconnection under general conditions of finite guide magnetic field, realistic electron-to-ion mass ratio, and in 3D geometries is still not adequately understood. The major goal of this project is to study the physics of collisionless magnetic reconnection in a current sheet under a finite guide field and with a realistic ion-to-electron mass ratio mi/me. Under the support of this DoE grant, we have made significant progress in (1) the investigation of linear and nonlinear current sheet instabilities under a wide range of guide field using the 3-D gyro kinetic electron and fully kinetic ion (GeFi) particle simulation model and (2) the investigation of fully nonlinear reconnection process with a strong guide field with the GeFi simulations. Furthermore, (3) we have extended the study to the realistic geometry of the magnetopause using 3D global hybrid simulations. This study, from local to global scales and for 3-D physics at plasma boundaries, has provides advanced theoretical understanding of the fundamental physics of magnetic reconnection in laboratory and space plasma regimes on multi-scales.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

DOE OSTI2023

Collaborative Research: Particle Simulation of Collisionless Magnetic Reconnection Under Finite Guide Field (Final Technical Report)

The project investigated 3-D linear instabilities of Harris current sheet under a realistic ion-to-electron mass ratio and a finite guide field. The excitation and 3-D properties of lower hybrid drift instability (LHDI), along with two new instabilities were demonstrated. The report presents the nonlinear properties of current sheet instabilities by employing Gyrokinetic electron and Fully kinetic ion (GeFi) particle simulation code. The GeFi simulation results are compared with fully kinetic PIC simulations.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

DOE OSTI2023

An Investigation of Time Distributions for Task Primitives to Support the HUNTER Dynamic Human Reliability Analysis

As an effort to support data collection for dynamic human reliability analysis (HRA), this study investigates time distributions for task primitives defined in the Goals, Operators, Methods, and Selection rules (GOMS)–Human Reliability Analysis (HRA) method. GOMS-HRA was developed to provide cognition-based time and human error probability information for dynamic HRA calculation in the Human Unimodel for Nuclear Technology to Enhance Reliability (HUNTER) framework. HUNTER is a framework to support the dynamic modelling of human error in conjunction with other modelling tools. In this paper, we investigate time distributions using experimental data collected from the Simplified Human Error Experimental Program (SHEEP) study, which suggests an HRA data collection framework to complement full-scope simulator research as well as collect input data for dynamic HRA using simplified simulators such as the Rancor Microworld Simulator. In this study, time required for GOMS-HRA task primitives to satisfy thirteen statistical distributions is investigated. Then, the time distributions from student operators and professional operators are compared and discussed. As a result, this study identified several time distributions on five GOMS-HRA task primitives at a statistically significant level. According to analyses to date, a greater number of significant time distributions was found in abnormal or emergency operating procedures rather than standard operating procedures. In the future, it is expected that the result of this study can provide objective reference on elapsed time data for task primitives as well as help to realistically simulate scenarios within dynamic HRA.

99 GENERAL AND MISCELLANEOUS↗

DOE OSTI2024

Validation of Phasor-Domain Transmission and Distribution Co-simulation Against Electromagnetic Transient Simulation

The rapid deployment of renewable energy resources has led to the widespread use of power electronics in modern power systems. As these systems transition from being dominated by large synchronous machines to increasingly incorporating inverter-based resources (IBRs), traditional methods are becoming inadequate. Addressing this challenge, this paper introduces a scalable phasor-domain T\&D co-simulation framework based on open-source software. It focuses on the framework's validation against the PSCAD Electromagnetic Transient (EMT) analysis tool. The validation results demonstrate the framework's high-fidelity and a computational time speed-up of 60 to 100 times, marking a pioneering validation effort in T\&D co-simulation research.

Inverter-based resources, co-simulation, Electroma↗

DOE OSTI2025

Time Distribution Analysis for Task Primitives to Support Dynamic Human Reliability Analysis

To support data collection for dynamic human reliability analysis (HRA), this study investigates time distributions for task primitives defined in the Goals, Operators, Methods, and Selection rules (GOMS)–Human Reliability Analysis (HRA) method and Human Reliability data EXtraction (HuREX). GOMS-HRA was developed to provide cognition-based time and human error probability (HEP) information for dynamic HRA calculations within the Human Unimodel for Nuclear Technology to Enhance Reliability (HUNTER) framework, while HuREX is a comprehensive HRA data collection method developed by the Korea Atomic Energy Research Institute (KAERI). In this paper, we examine time distributions by using experimental data collected from the Simplified Human Error Experimental Program (SHEEP) study, which proposes an HRA data collection framework to complement full-scope simulator research and gather input data for dynamic HRA by using simplified simulators such as the Rancor Microworld simulator. This paper investigates whether the time required for GOMS-HRA and HuREX task primitives fits 13 statistical distributions. Additionally, we compare and discuss the time distributions obtained from both student operators and professional operators. The result was that this study identified several time distributions for five GOMS-HRA and four HuREX task primitives. In the future, the results of this study are expected to provide objective reference data on the elapsed time for task primitives and aid in realistically simulating scenarios within dynamic HRA.

Dynamic Human Reliability Analysis↗

DOE OSTI2021

Enabling Load Following Capability in the Transatomic Power MSR

This project is dedicated towards designing a fuel processing system that enables liquid-fueled molten salt reactors (MSR) to load follow by removing the dissolved xenon in the fuel salt. As one of the Gen-IV nuclear reactor concepts, the molten salt reactor receives increasing development interests in the recent years. One distinguishing feature of the liquid-fueled molten salt reactor is its improved ability to operate in a load-following mode by including the unique online fission product removal system. Load-following means that the reactor changes its power output based on the demand on the grid. Most of the current operating nuclear reactors have limited load-following ability and operate as the base load on the grid. Due to the rapid increase of solar energy, the requirement on load-following capacity is significantly increased because of the varying power output of the solar panels, yet the traditional load-following capacity is expected to decrease as the decarbonization of the grid continues. Therefore, the ability to perform load-following operation for the nuclear reactors will greatly enhance the resilience of the grid and make nuclear energy more economically competitive. This load-following feature is included in many commercial molten salt reactor designs, such as the designs by Transatomic Power, Terrestrial Energy, and Flibe Energy. Unfortunately, detailed analysis of the fuel processing system for commercial scale MSRs is still lacking, as well as how the fuel processing quantitively impacts the load-following operation. Moreover, experimental data for many of the underlying physics of fuel processing is limited. This project aims to pave the way for the fuel processing technology to advance to the commercial stage by performing combined experimental and simulation research. During the project period, four interconnected aspects of the development of the fuel processing system in liquid-fueled molten salt reactors are investigated. These aspects are the simulation and analysis of the fission product removal system, the fuel cycle simulation, the coupled reactor neutronics and thermal hydraulics transient simulation, and the gaseous fission product removal experiment. Multiphase CFD simulations are performed for components of the processing systems, and simplified air-water experiments are carried out to provide validation data. It is concluded that the CFD simulation can satisfactorily predict the system level performance of the components, and engineering models are constructed based on this success. Fuel cycle analysis is performed for two representative MSR design, the MSBR and the Transatomic Power MSR. Open-source code SaltProc is developed to incorporate the unique fuel processing system of the MSRs. It is concluded that the removal of xenon is essential for load-following operation in thermal spectrum MSR and Molten Salt Breeder Reactor. For the Transatomic Power MSR, the xenon poisoning effect is negligible due to its relatively fast neutron spectrum, though the overall fuel cycle economics still benefits from the removal of xenon. Coupled reactor neutronics and thermal hydraulics transient simulation is performed specifically for the Transatomic Power MSR. It is concluded that the reactor core design could perform power ramping fast enough to satisfy load-following operation. Combining the findings from each aspect, it is concluded that the load-following operation of a thermal neutron MSR is dependent upon the removal of xenon, which could be achieved for a commercial sized reactor using continuous inert gas sparging in a separate system with reasonable dimensions.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗

DOE OSTI2022

High Rate Sodium Storage Mechanisms in Non-Graphitic Carbons

Lithium ion batteries (LIBs) have been commercialized as electrical energy storage devices in many common applications. Unlike lithium, sodium (Na) is both easy to find and is inexpensive, with wide supplies of precursors available on land and from salt water through desalination. Hybrid sodium ion capacitors and dedicated high power sodium ion batteries (NIBs) are emerging extremely fast charge time systems that employ two-dimensional carbon electrodes to store the charge. They are finding use in regenerative braking energy storage for cars, busses and public rail. To date, fast charge storage mechanisms in two-dimensional carbons are not understood. This combined experimental – simulation research will provide new fundamental insight into these unexplored but essential aspects of Na storage. A range of carbon structures and chemistries will be analyzed using advanced methods, including neutron scattering and first principles simulation. This work will yield the first new series of scientific insights on where in the carbon structure the Na ions reside and on their transport characteristics.

25 ENERGY STORAGE↗

DOE OSTI2025

Modeling detonation with CartaBlanca simulations

The accurate modeling of high explosive (HE) detonation and the con- sequent large solid deformation, failure, plastic flow, porosity growth, and shock wave propagation is important because simulations can capture spatial and temporal features that experimental diagnostics cannot capture. However, the simulation of the explosive event poses challenges to a computational scientist. These include the accurate modeling of ductile damage, crack formation, plastic deformation, as well as physical and nu- merical instabilities. The material response can be history-dependent and subject to large material deformation. Our research simulates the impact of a high explosive (Detasheet) onto a tantulum metal plate. We performed the simulations using CartaBlanca at different mesh resolutions. We decided the study would be impactful if we perform the simulations with the Material Point Method. Differences were observed at the different mesh resolutions in velocity and nodal stress magnitude, so increased mesh resolutions may be required. In addition, the Discontinuous Galerkin method would be needed to account for the large gas expansion.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2025

Integration and Demonstration of Monitoring, Modeling, and Prediction of DV-1 Amendment Performance at the Bench Scale: DV-1 Amendment Demonstration

During fiscal years 2024 and 2025, the U.S. Department of Energy’s Hanford Field Office commissioned Pacific Northwest National Laboratory to conduct applied research aimed at reducing the cost, time, and uncertainty associated with in situ treatment of vadose zone contaminants at the Hanford Site. This report outlines the integration of three key research efforts into a meso-scale demonstration designed to advance field-scale solutions that aim to (1) optimize the delivery of chemical amendments to contaminated soils, (2) reduce uncertainty in amendment delivery performance assessment using advanced monitoring techniques, and (3) provide real-time insights into when and where amendment-induced precipitation reactions occur in the subsurface. To achieve these objectives, the tank-scale (~ 1 cubic meter) Geophysical Imaging of Flow and Transport (GIFT) system was developed. GIFT enables experimental testing of amendment delivery while incorporating automated multi-modal monitoring approaches, including pressure measurements, direct fluid sampling, and remote time-lapse geophysical imaging. The data generated from these monitoring techniques will serve as inputs for a generative artificial-intelligence-driven digital twin – a numerical simulation model designed to honor observed data while quantifying uncertainty in simulation accuracy. Using this simulator, researchers will refine an amendment injection strategy to maximize delivery efficiency within a low-permeability soil zone. Monitoring data will be interpreted through simulated outputs to enhance understanding of the injection process. The efficacy of this integrated approach will be evaluated through direct sampling at the conclusion of the experiment.

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗

DOE OSTI2026

From Rules to Reasoning: A Survey of Large Language Model-Based Approaches to Scientific Hypothesis and Idea Generation

Scientific hypothesis generation represents a fundamental challenge in contemporary research due to exponentially expanding literature volumes and increasing disciplinary specialization. Large language models (LLMs) have emerged as transformative tools for automated scientific discovery, moving beyond traditional rule-based and literature-mining approaches. Four paradigmatic approaches define current LLM-driven hypothesis generation: direct prompting and fine-tuning methods, knowledge-enhanced frameworks integrating retrieval-augmented generation (RAG), multi-agent collaborative systems simulating research teams, and reasoning-focused approaches implementing cognitive architectures. Domain-specific applications demonstrate statistical equivalence to human expert performance in social psychology, experimental validation in biomedical research, and near-expert quality in astronomy. Evaluation methodologies encompass human expert assessment, LLM-as-judge frameworks, and comprehensive benchmarking systems. Technical challenges include hallucination management, knowledge integration limitations, and balancing novelty with feasibility. Future directions emphasize hybrid neural-symbolic architectures and sophisticated human-AI collaboration models for responsible scientific discovery acceleration.

AI-driven discovery↗