Search NASA⌕ Search

SEARCH · Search NASA

Results for “Implicit modeling”

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 487 records · Page 27
All sourcesNASA NTRSDOE OSTI
DOE OSTI2021

Variational, stable, and self-consistent coupling of 3D electromagnetics to 1D transmission lines in the time domain

This work presents a new multiscale method for coupling the 3D Maxwell's equations to the 1D telegrapher's equations. While Maxwell's equations are appropriate for modeling complex electromagnetics in arbitrary-geometry domains, simulation cost for many applications (e.g. pulsed power) can be dramatically reduced by representing less complex transmission line regions of the domain with a 1D model. By assuming a transverse electromagnetic (TEM) ansatz for the solution in a transmission line region, we reduce the Maxwell's equations to the telegrapher's equations. Here, we propose a self-consistent finite element formulation of the fully coupled system that uses boundary integrals to couple between the 3D and 1D domains and supports arbitrary unstructured 3D meshes. Additionally, by using a Lagrange multiplier to enforce continuity at the coupling interface, we allow for an absorbing boundary condition to also be applied to non-TEM modes on this boundary. We demonstrate that this feature reduces non-physical reflection and ringing of non-TEM modes off of the coupling boundary. By employing implicit time integration, we ensure a stable coupling, and we introduce an efficient method for solving the resulting linear systems. We demonstrate the accuracy of the new method on two verification problems, a transient O-wave in a rectilinear prism and a steady-state problem in a coaxial geometry, and show the efficiency and weak scalability of our implementation on a cold test of the Z-machine MITL and post-hole convolute.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2022

Efficient hybrid explicit-implicit learning for multiscale problems

Splitting method is a powerful method to handle application problems by splitting physics, scales, domain, and so on. Many splitting algorithms have been designed for efficient temporal discretization. Here, in this paper, our goal is to use temporal splitting concepts in designing machine learning algorithms and, at the same time, help splitting algorithms by incorporating data and speeding them up. We propose a machine learning assisted splitting scheme which improves the efficiency of the scheme meanwhile preserves the accuracy. We consider a recently introduced multiscale splitting algorithms, where the multiscale problem is solved on a coarse grid. To approximate the dynamics, only a few degrees of freedom are solved implicitly, while others explicitly. This splitting concept allows identifying degrees of freedom that need implicit treatment. In this paper, we use this splitting concept in machine learning and propose several strategies. First, the implicit part of the solution can be learned as it is more difficult to solve, while the explicit part can be computed. This provides a speed-up and data incorporation for splitting approaches. Secondly, one can design a hybrid neural network architecture because handling explicit parts requires much fewer communications among neurons and can be done efficiently. Thirdly, one can solve the coarse grid component via PDEs or other approximation methods and construct simpler neural networks for the explicit part of the solutions. We discuss these options and implement one of them by interpreting it as a machine translation task. This interpretation of the splitting scheme successfully enables us using the Transformer since it can perform model reduction for multiple time series and learn the connection between them. We also find that the splitting scheme is a great platform to predict the coarse solution with insufficient information of the target model: the target problem is partially given and we need to solve it through a known problem which approximates the target. Our machine learning model can incorporate and encode the given information from two different problems and then solve the target problems. We conduct four numerical examples and the results show that our method is stable and accurate.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2020

Spatiotemporally dynamic implicit large eddy simulation using machine learning classifiers

In this article, we utilize machine learning to dynamically determine if a point on the computational grid requires implicit numerical dissipation for large eddy simulation (LES). The decision making process is learnt through a priori training on quantities derived from direct numerical simulation (DNS) data. In particular, we compute eddy-viscosities obtained through the coarse-graining of DNS quantities and utilize their projection onto a Gaussian distribution to categorize areas that may require dissipation. If our learning determines that closure is necessary, an upwinded scheme is utilized for computing the non-linear Jacobian. In contrast, if it is determined that closure is unnecessary, a symmetric and second-order accurate energy and enstrophy preserving Arakawa scheme is utilized instead. This results in a closure framework that precludes the specification of any model-form for the small scale contributions of turbulence but deploys an appropriate numerical dissipation from explicit closure driven hypotheses. Here, this methodology is deployed for the Kraichnan turbulence test-case and assessed through various statistical quantities such as angle-averaged kinetic energy spectra and vorticity structure functions. Our framework thus establishes a link between the use of explicit LES ideologies for closure and numerical dissipation-based modeling of turbulence leading to improved statistical fidelity of a posteriori simulations.

42 ENGINEERING↗
DOE OSTI2020

Reaction Energetics and 13 C Fractionation of Alanine Transamination in the Aqueous and Gas Phases

The alanine transaminase enzyme catalyzes the transfer of an amino group from alanine to α-ketoglutarate to produce pyruvate and glutamate. Isotope fractionation factors (IFFs) for the reaction +H 3 NCH(CH 3 )COO – + – OOCCH 2 CH 2 C(O)COO – ↔ CH 3 C(O)COO – + + H 3 NCH(CH 2 CH 2 COO – )COO – (zwitterionic neutral alanine + doubly deprotonated α-ketoglutarate ↔ pyruvate + zwitterionic glutamate anion) were calculated from the partition functions of explicitly and implicitly solvated molecules at 298 K. Calculations were done for alanine (noncharge separated, zwitterion, deprotonated), pyruvic acid (neutral, deprotonated), glutamic acid (noncharge separated, zwitterion, deprotonated, doubly deprotonated), and α-ketoglutaric acid (neutral, deprotonated, doubly deprotonated). The computational results, calculated from gas phase- and aqueous-optimized clusters with explicit H 2 O molecules at the MP2/aug-cc-pVDZ and MP2/aug-cc-pVDZ/COSMO levels, respectively, predict that substitution of 13 C at the C2 position of alanine and pyruvic acid and their various forms leads to the C2 position of pyruvic acid/pyruvate being enriched in 13 C/ 12 C ratio by 9‰. Simpler approaches that estimate the IFFs based solely on changes in the zero-point energies (ZPEs) are consistent with the higher-level model. ZPE-based IFFs calculated for simple analogues formaldehyde and methylamine (analogous to the C2 positions of pyruvate and alanine, respectively) predict a 13 C enrichment in formaldehyde of 7–8‰ at the MP2/aug-cc-pVDZ and aug-cc-pVTZ levels. A simple predictive model using canonical functional group frequencies and reduced masses for 13 C exchange between R 2 C=O and R 2 CH–NH 2 predicted enrichment in R 2 C=O that is too large by a factor of two but is qualitatively accurate compared with the more sophisticated models. Our models are all in agreement with the expectation that pyruvate and formaldehyde will be preferentially enriched in 13 C because of the strength of their >C=O bond relative to that of ≡C–NH 2 in alanine and methylamine. 13 C/ 12 C substitution is also modeled at the methyl and carboxylic acid sites of alanine and pyruvic acid, respectively.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2021

Impact of Integration Scheme on Performance of Anisotropic Plasticity Models

Given the prevalent role of metals in a variety of industries, schemes to integrate corresponding constitutive models in finite element applications have long been studied. A number of formulations have been developed to accomplish this task; each with their own advantages and costs. Often the focus has been on ensuring the accuracy and numerical stability of these algorithms to enable robust integration. While important, emphasis on these performance metrics may often come at the cost of computational expense potentially neglecting the needs of individual problems. In the current work, the performance of two of the most common integration methods for anisotropic plasticity -- the convex cutting plane (CCP) and closest point projection (CPP) -- across a variety of metrics is assessed; including accuracy and cost. A variety of problems are considered ranging from single elements to large representative simulations including both implicit quasistatic and explicit transient dynamic type responses. The relative performance of each scheme in the different instances is presented with an eye towards guidance on when the different algorithms may be beneficial.

36 MATERIALS SCIENCE↗
DOE OSTI2024

Evaluation of a New Approach for Entrainment and Detrainment Rate Estimation

Entrainment and detrainment rates (ε and δ) constitute the most critical free parameters in mass flux schemes commonly employed for cumulus parameterizations. Recently, Zhu et al. (2021) introduced a new approach that utilizes aircraft observations to simultaneously estimate ε and δ for cumulus clouds, overcoming the limitation of other observation-based approaches that solely yield ε without offering insights into δ. This study aims to comprehensively evaluate the reliability of this new approach. First, evaluation using an Explicit Mixing Parcel Model demonstrates the capability of the new approach to back-calculate predetermined ε and δ based on the physical properties before and after the entrainment mixing. Second, evaluation using large-eddy simulations illustrates that the new approach yields consistent ε and δ profiles compared to the traditional approach. Sensitivity tests indicate a weak sensitivity of the estimated δ with the new approach to the entrained air source. A decrease in the proportion of cloudy air in the assumed detrained air leads to a reduction in the estimated δ, while ε remains unaffected. Finally, the most appropriate assumptions for entrained and detrained air are discussed. Estimating ε for cumulus parameterizations involves acquiring ambient air more than 500 m away from the cloud edge as entrained air. Due to implicit mean field approximations in the traditional approach, determining the optimal assumption for detrained air properties proves challenging. Finally, this study confirms the reliability of the new approach in estimating ε and δ, providing confidence in its application to extensive observational data and advancement in parameterization.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2025

Learning nuclear cross sections across the chart of nuclides with graph neural networks

We explore the use of deep learning techniques to learn how nuclear cross sections change as we add or remove protons and neutrons. As a proof of principle, we focus on the neutron-induced reactions in the fast energy regime. Our approach follows a two-stage learning framework. First, we apply representation learning to encode cross section data into a latent space using either variational autoencoders (VAEs) or implicit neural representations (INRs). Then, we train graph neural networks (GNNs) on the resulting embeddings to predict missing values across the nuclear chart by leveraging the topological structure of neighboring isotopes. We demonstrate accurate cross section predictions within a 9 × 9 block of missing nuclei. We also find that the optimal GNN training strategy depends on the type of latent representation used, with VAE embeddings performing best under end-to-end optimization in the original space, while INR embeddings achieve better results when the GNN is trained only in the latent space. Furthermore, using clustering algorithms, we map groups of latent vectors into regions of the nuclear chart and show that VAEs and INRs can discover some of the neutron magic numbers. These findings suggest that deep-learning models based on the representation encoding of cross sections combined with graph neural networks hold significant potential in augmenting nuclear theory models, e.g., by providing reliable estimates of covariances of cross sections, including cross-material covariances.

Machine learning↗
DOE OSTI2024

Site heterogeneity and broad surface-binding isotherms in modern catalysis: Building intuition beyond the Sabatier principle

Learning the science of heterogeneous catalysis and electrocatalysis always starts with the simple case of a flat, uniform surface with an ideal adsorbate. It has of course been recognized for a century that real catalysts are more complicated. For the increasingly complex catalysts of the 21st century, this Perspective argues that surface heterogeneity and non-ideal binding isotherms are central features, and their implications need to be incorporated in current thinking. A variety of systems are described herein where catalyst complexity leads to broad, non-Langmuirian surface isotherms for the binding of hydrogen atoms – and this occurs even for ideal, flat Pt(111) surfaces. Modern catalysis employs nanoscale materials whose surfaces have substantial step, edge, corner, impurity, and other defect sites, and they increasingly have both metallic and non-metallic elements M n X m , including metal oxides, chalcogenides, pnictides, carbides, doped carbons, etc. The surfaces of such catalysts are often not crystal facets of the bulk phase underneath, and they typically have a variety of potential active sites. Catalytic surfaces in operando are often non-stoichiometric, amorphous, dynamic, and impure, and often vary from one part of the surface to another. Understanding of the issues that arise at such nanoscale, multi-element catalysts is just beginning to emerge. Yet these catalysts are widely discussed using Brønsted/Bell-Evans-Polanyi (BEP) relations, volcano plots, Tafel slopes, the Butler-Volmer equation, and other linear free energy relations (LFERs), which all depend on the implicit assumption that the active sites are “similar” and that surface adsorption is close to ideal. These assumptions underly the ubiquitous intuition based on the Sabatier Principle, that the fastest catalysis will occur when key intermediates have free energies of adsorption that are not too strong nor too weak. Current catalysis research often aims to minimize the complexity of non-ideal isotherms through experimental and computational design (e.g., the use of single crystal surfaces), and these studies are the foundation of the field. In contrast, this Perspective argues that the heterogeneity of binding sites and binding energies is an inherent strength of these catalysts. Here, this diversity makes many nanoscale catalysts inherently a high-throughput screen wrapped in a tiny package. Only by making the heterogeneity part of the foundation of catalysis models, sorting the types of active sites and dissecting non-ideal binding isotherms, will modern catalysis learn to harness the inherent diversity of real catalysts. Controlling and exploiting diversity rather than avoiding it will help to optimize complex modern catalysts and catalytic conditions.

Mayer, James M.↗
DOE OSTI2024

Development and application of hybrid AIMD/cDFT simulations for atomic-to-mesoscale chemistry

Many important chemical processes involve reactivity and dynamics in complex solutions. Gaining a fundamental understanding of these reaction mechanisms is a challenging goal that requires advanced computational and experimental approaches. However, important techniques such as molecular simulation have limitations in terms of scales of time, length, and system complexity. Furthermore, among the currently available solvation models, there are very few designed to describe the interaction between the molecular scale and the mesoscale. To help address this challenge, here, we establish a novel hybrid approach that couples first-principles plane-wave density functional theory with classical density functional theory (cDFT). In this approach, a region of interest described by ab initio molecular dynamics (AIMD) interacts with the surrounding medium described using cDFT to arrive at a self-consistent ground state. cDFT is a robust but efficient mesoscopic approach to accurate thermodynamics of bulk electrolyte solutions over a wide concentration range (up to 2M concentrations). Benchmarking against commonly used continuum models of solvation, such as SMD, as well as experiments, demonstrates that our hybrid AIMD–cDFT method is able to produce reasonable solvation energies for a variety of molecules and ions. With this model, we also examined the solvent effects on a prototype S N 2 reaction of the nucleophilic attack of a chloride ion on methyl chloride in the solution. The resulting reaction pathway profile and the solution phase barrier agree well with experiment, showing that our AIMD/cDFT hybrid approach can provide insight into the specific role of the solvent on the reaction coordinate.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2020

Exponential time differencing for the tracer equations appearing in primitive equation ocean models

The tracer equations are part of the primitive equations used in ocean modeling and describe the transport of tracers, such as temperature, salinity or chemicals, in the ocean. Depending on the number of tracers considered, several equations may be added to and coupled to the dynamics system. In many relevant situations, the time-step requirements of explicit methods imposed by the transport and mixing in the vertical direction are more restrictive than those for the horizontal, and this may cause the need to use very small time steps if a fully explicit method is employed. To overcome this issue, we propose an exponential time differencing (ETD) solver where the vertical terms (transport and diffusion) are treated with a matrix exponential, whereas the horizontal terms are dealt with in an explicit way. In this work, we investigate numerically the computational speed-ups that can be obtained over other semi-implicit methods, and we analyze the advantages of the method in the case of multiple tracers.

42 ENGINEERING↗
DOE OSTI2021

Density functional theory calculations of the thermodynamic and kinetic properties of point defects in β -U

Density functional theory (DFT) calculations of the thermodynamic and kinetic properties of point defects in the β phase of uranium are reported. Defect energies and entropies were calculated using 2 x 2 x 2 supercells and the Generalized Gradient Approximation (GGA) of Perdew-Burke-Ernzerhof (PBE) for the exchange-correlation potential. Due to computational cost, calculations of the vibrational properties governing entropies were performed by only displacing atoms within (roughly) the 3rd or 4th nearest neighbor shell of the defect, which implicitly assumes that atoms beyond this distance are unaffected by the defect. Migration barriers were estimated by nudged elastic band (NEB) calculations. The low symmetry of the β-U phase (the unit cell is tetragonal and contains 30 atoms) results in many point defect configurations and even more migration pathways. A connectivity map, starting from the most stable point defects, was developed in order to identify the rate-limiting step controlling the net diffusion rate in each crystallographic direction. The uranium self-diffusivity tensor was calculated by combining the defect formation energies, entropies, migration barriers and attempt frequencies. The fastest diffusion rate was determined to be a vacancy mechanism in the z crystallographic direction. The predicted uranium self-diffusivity for this mechanism agrees well with available experimental data. Finally, the diffusion mechanisms and rates identified in this study will inform fuel performance models of swelling and gas evolution.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗
DOE OSTI2020

MASTODON: An Open-Source Software for Seismic Analysis and Risk Assessment of Critical Infrastructure

Seismic analysis and risk assessment of safety-critical infrastructure like hospitals, nuclear power plants, dams, and facilities handling radioactive materials involve computationally intensive numerical models and coupled multiphysics scenarios. They are also performed in a strict regulatory environment that requires high software quality assurance standards, and in the case of safety-related nuclear facilities, a conformance to the American Society of Mechanical Engineers Nuclear Quality Assurance (NQA-1) standard. This paper introduces the open-source finite-element software, MASTODON (Multi-hazard Analysis of Stochastic Time-Domain Phenomena), which implements state-of-the-art seismic analysis and risk assessment tools in a quality-controlled environment. MASTODON is built on MOOSE (Multi-physics Object-Oriented Simulation Environment), which is a highly parallelizable, NQA-1 conforming, coupled multiphysics, finite-element framework developed at Idaho National Laboratory. MASTODON is capable of fault rupture and source-to-site wave propagation using the domain reduction method, nonlinear site response, and soil-structure interaction analysis, implicit and explicit time integration, automated stochastic simulations, and seismic probabilistic risk assessment. When coupled with other MOOSE applications, MASTODON can also solve strongly and weakly coupled multiphysics problems. This paper presents a summary of the capabilities of MASTODON and some demonstrative examples.

42 ENGINEERING↗
DOE OSTI2024

NUMERICAL MODELING OF A SOLID OXIDE FUEL CELL FOR USE IN REAL-TIME SIMULATION AND CYBER-PHYSICAL SYSTEMS

Cyber-physical systems provide a mechanism with which to investigate the physical phenomena and behavior of traditionally cost-prohibitive or otherwise fragile equipment. For the National Energy Technology Laboratory (NETL), this approach resulted in the Hybrid Performance (Hyper) facility which features a gas turbine-SOFC hybrid cycle utilizing real turbomachinery and a simulated SOFC stack. This allows for the investigation of combined cycle performance and control strategies, in an exhaustive manner, both without fear of destroying delicate state-of-the-art fuel cells, and with the full accuracy of real-world turbomachinery. Issues arose between the transient response of the SOFC model being limited to a sample time of 80 milliseconds, due to the calculation time of the SOFC model taking on average 40 milliseconds to calculate for a given timestep with spikes in calculation time reaching the 80 millisecond threshold. In order to be able to match the speed of transients from the turbomachinery and likewise better discern transient behavior, it was determined that the SOFC model must be optimized to operate at a sample time of 5 milliseconds. Therefore, it is necessary to optimize the SOFC model in order to decrease the calculation time from around 40 milliseconds, down to at the most 5 milliseconds. To do this, both the electrochemical algorithm and the thermal algorithm used to simulate the physical behavior of the SOFC are investigated to determine where improvements can be made. To this end the rootfinding numerical recipes of the electrochemical algorithm are investigated as the complex electrochemistry requires a highly iterative nested dual convergence loop to resolve the voltage-current relationship, and likewise the temporal discretization of the thermal algorithm is modified for the sake of higher accuracy and stability. Ultimately the new electrochemical algorithm featuring higher order rootfinding schemes proves to be efficient enough to reach the sub 5 millisecond target, signifying an order of magnitude reduction in calculation time, and when coupled with the new temporal discretization similar calculation time characteristics show that a fully implicit, higher order temporal discretization can also successfully be used if desired. Ultimately this result means that the cyber-physical simulation system can operate at higher sample rates, and resolve transient events at significantly higher resolution and fidelity.

Arias, Jesus↗
DOE OSTI2025

Preserving nonlinear constraints in variational flow filtering data assimilation

Data assimilation aims to estimate the states of a dynamical system by optimally combining sparse and noisy observations of the physical system with uncertain forecasts produced by a computational model. The states of many dynamical systems of interest obey nonlinear physical constraints, and the corresponding dynamics is confined to a certain sub-manifold of the state space. Standard data assimilation techniques applied to such systems yield posterior states lying outside the manifold, violating the physical constraints. This work focuses on particle flow filters which use stochastic differential equations to evolve state samples from a prior distribution to samples from an observation-informed posterior distribution. The variational Fokker-Planck (VFP)—a generic particle flow filtering framework—is extended to incorporate non-linear, equality state constraints in the analysis. To this end, two algorithmic approaches that modify the VFP stochastic differential equation are discussed: (i) VFPSTAB, to inexactly preserve constraints with the addition of a stabilizing drift term, and (ii) VFPDAE, to exactly preserve constraints by treating the VFP dynamics as a stochastic differential-algebraic equation (SDAE). Additionally, an implicit-explicit time integrator is developed to evolve the VFPDAE dynamics. The strength of the proposed approach for constraint preservation in data assimilation is demonstrated on three test problems: the double pendulum, Korteweg-de-Vries, and the incompressible Navier-Stokes equations.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2026

Structure-aware Initialization via Numerical Continuation and Informed Priors

Scientific machine learning (SciML) often operates in ill-conditioned, weakly identifiable regimes due to limited data or indirect observations. In such settings, optimization and inference are highly sensitive to the starting point, making initialization--often under-reported--a consequential degree of freedom. Random initialization is not a neutral default as it induces an implicit prior over candidate solutions and can systematically bias the result, producing large run-to-run variability. Here, we formalize this view by treating initialization as a hidden confounder in SciML and develop a unifying theory for structure-aware initialization via numerical continuation, constructing warm starts from related problem instances. Across representative tasks, including physics-informed neural networks, maximum likelihood estimation, and variational inference, warm starts have been shown to consistently reduce optimization effort and improve reliability.

Data integrity↗
DOE OSTI2021

An FFT-based approach for Bloch wave analysis: application to polycrystals

A method based on the Fast Fourier Transform is proposed to obtain the dispersion relation of acoustic waves in heterogeneous periodic media with arbitrary microstructures. The microstructure is explicitly considered using a voxelized Representative Volume Element (RVE). The dispersion diagram is obtained solving an eigenvalue problem for Bloch waves in Fourier space. To this aim, two linear operators representing stiffness and mass are defined through the use of differential operators in Fourier space. The smallest eigenvalues are obtained using the implicitly restarted Lanczos and the subspace iteration methods, and the required inverse of the stiffness operator is done using the conjugate gradient with a preconditioner. The method is used to study the propagation of acoustic waves in elastic polycrystals, showing the strong effect of crystal anistropy and polycrystaline texture on the propagation. It is shown that the method combines the simplicity of classical Fourier series analysis with the versatility of Finite Elements to account for complex geometries proving an efficient and general approach which allows the use of large RVEs in 3D.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2023

Impact of solvation on the GW quasiparticle spectra of molecules

First-principles calculations for electrochemistry require accurate treatment of both electronic structure and solvation. The perturbative GW approximation starting from density functional theory (DFT) calculations accurately models materials systems with varying dimensionality. Continuum solvation models enable efficient treatment of solvation effects in DFT calculations, but their applications with beyond-DFT electronic structure methods such as GW have been limited. Here, we introduce the frequency-dependent liquid polarizability from a nonlocal continuum solvation model in the screened Coulomb interaction of full-frequency GW calculations with a solvated DFT starting point. In this study, we show that the liquid screening contributions substantially reduce the HOMO–LUMO gap of molecules by 3–5 eV, while solvent effects on the DFT starting point negligibly impact the GW gap. The resulting framework facilitates the simultaneous electronic and solvation accuracy needed for first-principles electrochemistry.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2026

Agentic Diagrammatica: Towards Autonomous Symbolic Computation in High Energy Physics

We present Diagrammatica, a symbolic computation extension to the HEPTAPOD agentic framework, which enables LLM agents to plan and execute multi-step theoretical calculations. Symbolic computation poses a distinctive reliability challenge for LLM agents, as correctness is governed by implicit mathematical conventions that are not encoded in a form that can be easily checked in the computational backend. We identify two complementary remedies, tool-constrained computation and targeted knowledge grounding, and pursue the first as the primary architecture. Concretely, we concentrate the agent's action distribution onto tool calls with convention-fixing semantics, in which the agent specifies a compact, human-auditable diagram specification and a trusted backend performs the symbolic or numerical manipulations exactly. The toolkit provides two complementary calculation paths consuming a shared diagram specification: Naive Dimensional Analysis (NDA) for order-of-magnitude rate estimates and Exact Diagrammatic Analysis (EDA) for tree-level symbolic calculations via automatic FeynCalc code generation, both supplemented by automatic Feynman diagram enumeration and a navigable theory knowledge base. The architecture is validated on two benchmarks: (1) an exhaustive catalog of all tree-level, single-vertex $1\to 2$ partial decay widths across scalar, fermion, and vector parents, with complete massless and threshold limits and Standard Model validation; and (2) an NDA sensitivity study of the muon decay multiplicity $μ^+ \to ν_μ\barν_e + n(e^+e^-) + e^-$, determining the maximum observable $n$ at current and planned muon experiments.

Menzo, Tony [Alabama U.; Fermilab] (ORCID:00000002↗