Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “gradient methods”

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

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

At least 1,441 records · Page 80
All sourcesNASA NTRSDOE OSTI
DOE OSTI2020

Nonlinear global gyrokinetic delta- f turbulence simulations in a quasi-axisymmetric stellarator

We use the global gyrokinetic stellarator code XGC-S to look at turbulence physics in geometry of the proposed quasi-axisymmetric stellarator, QUASAR. In this article, the XGC-S code is first verified for nonlinear simulations of ion temperature gradient-driven turbulence by a successful benchmark with the established nonlinear tokamak code XGC1 in a CYCLONE-like tokamak case. We then apply the XGC-S code to QUASAR for nonlinear turbulence physics. We find that the highly poloidally localized linear mode gives way to a more evenly distributed nonlinear turbulent structure. Furthermore, the radial extent of the mode increases and occupies a large fraction of the volume, despite the initially localized temperature profile, as the profiles relax.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
DOE OSTI2023

Fluid structural interaction simulation of the MP-1 plate performance irradiated in the advanced test reactor

Due to the high neutron fluxes they generate and increased heat transfer performance, plate type fuels are used in the U.S. high-performance research reactors. During irradiation, a significant amount of fission energy (i.e., ~200 MeV per atom) is released by the U-235 chain reaction that is carried away by the coolant. Through thermal hydraulic analysis, the coolant’s heat transfer capability is investigated to ensure that the desired fuel temperature can be maintained. At high temperatures, the plate undergoes elastic/plastic deformation, creep, and swelling as a result of both the temperature gradients and the fission gas production within the fuel. These effects are studied via fuel performance analysis. When the plate deformation is small enough that the coolant flow remains relatively unchanged, conducting these two types of analyses independently will suffice. But at high fuel burnups, the swollen plates may encroach into the coolant channels that separate the fuel plates from each other and cause these channels to narrow. Large reductions in channel gap size imperil cooling performance, causing fuel temperatures to rise. Further, if the plate deformation is asymmetric, the fuel centerline will shift toward one side of the channel, causing an uneven reduction in coolability. In addition, a boehmite (oxide) layer will, over time, grow on the plate surface, further obstructing the heat transfer process. To precisely predict fuel temperatures/deformation, a complete coupled analysis that considers coolant flow, heat transfer, oxide growth, elastic/plastic deformation, creep, and swelling is needed; however, this type of analysis method is not available in the literature. To fill the gap, this research developed a fluid structure interaction (FSI) approach to the fuel plate analysis, then successfully applied it to the Mini-Plate (MP)-1 experiment, which was irradiated in the Advanced Test Reactor (ATR) for both one and two cycles. The complete analysis coupled STAR-CCM+, a computational fluid dynamic (CFD) software for calculating flow, with Abaqus, a finite element analysis code for calculating plate deformation. Improvements in the results were found when comparing the fully coupled analysis to the independently conducted analyses but they were not significant due to the miniature size of the plates and the relatively short irradiation time. In the future, the fully coupled approach presented herein will be applied to full-size fuel plates with longer irradiation cycles once additional experiments become available.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗
DOE OSTI2022

Quasiparticle self-consistent GW band structures and phase transitions of LiAlO 2 in tetrahedrally and octahedrally coordinated structures

In this work, a first-principles computational study is presented of various phases of LiAlO 2 . The relative total energies and equations of state of the α, β, and γ phases are determined after structural relaxation of each phase. The β and γ tetrahedral phases are found to be very close in energy and lattice volume with the γ phase having the lowest energy. The octahedral α phase is a high-pressure phase and the transition pressure from the γ and β phases to α is determined to be about 1 GPa. The electronic band structures of each phase at their own equilibrium volume are determined using the quasiparticle self-consistent (QS) GW method as well as using the 0.8 Σ approach in which the QS GW self-energy is reduced by a factor of 0.8 to correct for the underscreening of W in QS GW. The effective masses of the band edges and the nature of the band gaps are presented. The lowest energy γ phase is found to have a pseudodirect gap of 7.69 eV. The gap is direct at Γ but corresponds to a dipole forbidden transition. The imaginary part of the dielectric function and the absorption coefficient are calculated in the long-wavelength limit and the random phase approximation, without local field or electron-hole interaction effects for each phase, and their anisotropies are discussed. Si doping on the Al site is investigated as a possible n-type dopant in γ-LiAlO 2 using a 128-atom supercell corresponding to 3.125% Si on the Al sublattice in the generalized gradient approximation and a smaller 16-atom cell with 25% Si in the QS GW approximation. The Si is found to significantly perturb the conduction band and lower the gap but a clearly separated deep donor defect level is not found. However, the donor binding energy is still expected to be relatively deep, on the order of a few tenths eV in the hydrogenic effective mass approximation.

36 MATERIALS SCIENCE↗
DOE OSTI2025

Criticality analysis of nuclear binding energy neural networks

Machine learning methods, in particular deep learning methods such as artificial neural networks (ANNs) with many layers, have become widespread and useful tools in nuclear physics. However, these ANNs are typically treated as ‘black boxes’, with their architecture (width, depth, and weight/bias initialization) and the training algorithm and parameters chosen empirically by optimizing learning based on limited exploration. We test a non-empirical approach to understanding and optimizing nuclear physics ANNs by adapting a criticality analysis based on renormalization group flows in terms of the hyperparameters for weight/bias initialization, training rates, and the ratio of depth to width. This treatment utilizes the statistical properties of neural network initialization to find a generating functional for network outputs at any layer, allowing for a path integral formulation of the ANN outputs as a Euclidean statistical field theory. We use a prototypical example to test the applicability of this approach: a simple ANN for nuclear binding energies. We find that with training using a stochastic gradient descent optimizer, the predicted criticality behavior is realized, and optimal performance is found with critical tuning. However, the use of an adaptive learning algorithm leads to somewhat superior results without concern for tuning and thus obscures the analysis. Nevertheless, the criticality analysis offers a way to look within the black box of ANNs, which is a first step towards potential improvements in network performance beyond using adaptive optimizers.

artificial neural network↗
DOE OSTI2022

CoSHA: Code for Stellar Properties Heuristic Assignment—for the MaStar Stellar Library

We introduce CoSHA: a Code for Stellar properties Heuristic Assignment. In order to estimate the stellar properties, CoSHA implements a Gradient Tree Boosting algorithm to label each star across the parameter space (T eff , $\mathrm{log}g$, [Fe/H], and [α/Fe]). We use CoSHA to estimate the stellar atmospheric parameters of 22,000 unique stars in the MaNGA Stellar Library (MaStar). To quantify the reliability of our approach, we run internal tests, using both the Göttingen Stellar Library (a theoretical library) and the first data release of MaStar, and external tests, by comparing the resulting distributions in the parameter space with the APOGEE estimates of the same properties. In summary, our parameter estimates span the ranges T eff = [2900, 12,000] K, $\mathrm{log}g=[-0.5,5.6]$, [Fe/H] = [-3.74, 0.81], and αM = [-0.22, 1.17]. We report internal (external) uncertainties of the properties of ${\sigma }_{{T}_{\mathrm{eff}}}\sim 43(240)$ K, ${\sigma }_{\mathrm{log}g}\sim 0.2(0.4)$, σ [Fe/H] ~ 0.16(0.24), and σ [α/Fe] ~ 0.09(0.08). These uncertainties are comparable to those of other methods with similar objectives. Despite the fact that CoSHA is not aware of the spatial distributions of these physical properties in the Milky Way, we are able to recover the main trends known in the literature. The catalog of physical properties for MaStar can be accessed online.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2024

The magnetic gradient scale length explains why certain plasmas require close external magnetic coils

Abstract The separation between the last closed flux surface of a plasma and the external coils that magnetically confine it is a limiting factor in the construction of fusion-capable plasma devices. This plasma-coil separation must be large enough so that components such as a breeding blanket and neutron shielding can fit between the plasma and the coils. Plasma-coil separation affects reactor size, engineering complexity, and particle loss due to field ripple. For some plasmas it can be difficult to produce the desired flux surface shaping with distant coils, and for other plasmas it is infeasible altogether. Here, we seek to understand the underlying physics that limits plasma-coil separation and explain why some configurations require close external coils. In this paper, we explore the hypothesis that the limiting plasma-coil separation is set by the shortest scale length of the magnetic field as expressed by the ∇ B tensor. We tested this hypothesis on a database of > 40 stellarator and tokamak configurations. Within this database, the coil-to-plasma distance compared to the minor radius varies by over an order of magnitude. The magnetic scale length is well correlated to the coil-to-plasma distance of actual coil designs generated using the REGCOIL method (Landreman 2017 Nucl. Fusion 57 046003). Additionally, this correlation reveals a general trend that larger plasma-coil separation is possible with a small number of field periods.

Physics↗
DOE OSTI2020

Dynamic Structure and Phase Behavior of a Block Copolymer Electrolyte under dc Polarization

There is an important consideration when designing lithium battery electrolytes for advanced applications is how the electrolyte facilitates ion transport at fast charge and discharge rates. Large current densities are accompanied by large salt concentration gradients across the electrolyte. Nanostructured composite electrolytes have been proposed to enable the use of high energy density lithium metal anodes, but many questions about the interplay between the electrolyte morphology and the salt concentration gradient that forms under dc polarization remain unanswered. To address these questions, we use an in situ small-angle X-ray scattering technique to examine the nanostructure of a polystyrene- block -poly(ethylene oxide) copolymer electrolyte under dc polarization with spatial and temporal resolution. In the quiescent state, the electrolyte exhibits a lamellar morphology. The passage of ionic current in a lithium symmetric cell leads to the formation of concurrent phases: a disordered morphology near the negative electrode, lamellae in the center of the cell, and coexisting lamellae and gyroid near the positive electrode. The most surprising result of this study was obtained after the applied electric field was turned off: a current-induced gyroid phase grows in volume for 6 h in spite of the absence of an obvious driving force. We show that this reflects the formation of localized pockets of salt-dense electrolyte, termed concentration hotspots, under dc polarization. Our methods may be applied to understand the dynamic structure of composite electrolytes at appreciable current densities.

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

Projecting Permafrost Thaw of Sub-Arctic Tundra With a Thermodynamic Model Calibrated to Site Measurements

Northern circumpolar permafrost thaw affects global carbon cycling, as large amounts of stored soil carbon becomes accessible to microbial breakdown under a warming climate. The magnitude of carbon release is linked to the extent of permafrost thaw, which is locally variable and controlled by soil thermodynamics. Soil thermodynamic properties, such as thermal diffusivity, govern the reactivity of the soil-atmosphere thermal gradient, and are controlled by soil composition and drainage. In order to project permafrost thaw for an Alaskan tundra experimental site, we used seven years of site data to calibrate a soil thermodynamic model using a data assimilation technique. The model reproduced seasonal and interannual temperature dynamics for shallow (5–40 cm) and deep soil layers (2–4 m), and simulations of seasonal thaw depth closely matched observed data. The model was then used to project permafrost thaw at the site to the year 2100 using climate forcing data for three future climate scenarios (RCP 4.5, 6.0, and 8.5). Minimal permafrost thawing occurred until mean annual air temperatures rose above the freezing point, after which we measured over a 1 m increase in thaw depth for every 1 °C rise in mean annual air temperature. Under no projected warming scenario was permafrost remaining in the upper 3 m of soil by 2100. We demonstrated an effective data assimilation method that optimizes parameterization of a soil thermodynamic model. The sensitivity of local permafrost to climate warming illustrates the vulnerability of sub-Arctic tundra ecosystems to significant and rapid soil thawing.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2020

A micromirror array with annular partitioning for high-speed random-access axial focusing

Dynamic axial focusing functionality has recently experienced widespread incorporation in microscopy, augmented/virtual reality (AR/VR), adaptive optics and material processing. However, the limitations of existing varifocal tools continue to beset the performance capabilities and operating overhead of the optical systems that mobilize such functionality. The varifocal tools that are the least burdensome to operate (e.g. liquid crystal, elastomeric or optofluidic lenses) suffer from low (≈100 Hz) refresh rates. Conversely, the fastest devices sacrifice either critical capabilities such as their dwelling capacity (e.g. acoustic gradient lenses or monolithic micromechanical mirrors) or low operating overhead (e.g. deformable mirrors). Here, we present a general-purpose random-access axial focusing device that bridges these previously conflicting features of high speed, dwelling capacity and lightweight drive by employing low-rigidity micromirrors that exploit the robustness of defocusing phase profiles. Geometrically, the device consists of an 8.2 mm diameter array of piston-motion and 48-μm-pitch micromirror pixels that provide 2π phase shifting for wavelengths shorter than 1100 nm with 10–90% settling in 64.8 μs (i.e., 15.44 kHz refresh rate). The pixels are electrically partitioned into 32 rings for a driving scheme that enables phase-wrapped operation with circular symmetry and requires <30 V per channel. Optical experiments demonstrated the array’s wide focusing range with a measured ability to target 29 distinct resolvable depth planes. Overall, the features of the proposed array offer the potential for compact, straightforward methods of tackling bottlenecked applications, including high-throughput single-cell targeting in neurobiology and the delivery of dense 3D visual information in AR/VR.

42 ENGINEERING↗
DOE OSTI2022

FLOW Estimation and Rose Superposition (FLOWERS): an integral approach to engineering wake models

Abstract. Annual energy production (AEP) is often the objective function in wind plant layout optimization studies. The conventional method to compute AEP for a wind farm is to first evaluate power production for each discrete wind direction and speed using either computational fluid dynamics simulations or engineering wake models. The AEP is then calculated by weighted-averaging (based on the wind rose at the wind farm site) the power produced across all wind directions and speeds. We propose a novel formulation for time-averaged wake velocity that incorporates an analytical integral of a wake deficit model across every wind direction. This approach computes the average flow field more efficiently, and layout optimization is an obvious application to exploit this benefit. The clear advantage of this new approach is that the layout optimization produces solutions with comparable AEP performance yet is completed 2 orders of magnitude faster. The analytical integral and the use of a Fourier expansion to express the wind speed and wind direction frequency create a relatively smooth solution space for the gradient-based optimizer to excel in comparison to the existing weighted-averaging power calculation.

17 WIND ENERGY↗
DOE OSTI2022

Experimentally validated multiphysics modeling of fracture induced by thermal shocks in sintered UO 2 pellets

Uranium Dioxide (UO 2 ) fuel powers almost all commercial Nuclear Power Plants (NPPs) worldwide, generating carbon-free energy and contributing to the fight against climate change. UO 2 fuel incurs damage and fractures due to large thermal gradients that develop across the fuel pellet during normal and transient operating conditions. A comprehensive understanding of the underlying mechanisms by which these processes take place is still lacking. A combined experimental and computational approach is utilized here to quantify the behavior of UO 2 fuel fracture induced by thermal shock. Here, this work introduces both (1) an experimental study to understand the fuel fracturing behavior of sintered UO 2 pellets when exposed to thermal shock, and (2) a Multiphysics phase-field fracture model capable of simulating this process. Parametric studies were conducted to evaluate the effects of uncertainties in fracture properties on the fracture behavior of UO 2 due to thermal shocking. A set of energy release rate (or equivalently fracture toughness) and contract area (the part of the fuel pellet in direct contact with the cold bath) were able to capture the overall fracture trends of the corresponding experimental data. Our combined approach presents a new method for accounting for the effects of microstructure and sample size on the energy release rate/fracture toughness. The experimental data were collected from multiple experiments that exposed UO 2 pellets to high-temperature conditions (589–676 °C) followed by a quench in sub-zero water. This work demonstrates that joint experimental and computational efforts are able to advance the understanding of thermal fracture in the primary fuel source for existing and future NPPs.

36 MATERIALS SCIENCE↗
DOE OSTI2021

Galaxy morphological classification catalogue of the Dark Energy Survey Year 3 data with convolutional neural networks

ABSTRACT We present in this paper one of the largest galaxy morphological classification catalogues to date, including over 20 million galaxies, using the Dark Energy Survey (DES) Year 3 data based on convolutional neural networks (CNNs). Monochromatic i-band DES images with linear, logarithmic, and gradient scales, matched with debiased visual classifications from the Galaxy Zoo 1 (GZ1) catalogue, are used to train our CNN models. With a training set including bright galaxies (16 ≤ i < 18) at low redshift (z < 0.25), we furthermore investigate the limit of the accuracy of our predictions applied to galaxies at fainter magnitude and at higher redshifts. Our final catalogue covers magnitudes 16 ≤ i < 21, and redshifts z < 1.0, and provides predicted probabilities to two galaxy types – ellipticals and spirals (disc galaxies). Our CNN classifications reveal an accuracy of over 99 per cent for bright galaxies when comparing with the GZ1 classifications (i < 18). For fainter galaxies, the visual classification carried out by three of the co-authors shows that the CNN classifier correctly categorizes discy galaxies with rounder and blurred features, which humans often incorrectly visually classify as ellipticals. As a part of the validation, we carry out one of the largest examinations of non-parametric methods, including ∼100 ,000 galaxies with the same coverage of magnitude and redshift as the training set from our catalogue. We find that the Gini coefficient is the best single parameter discriminator between ellipticals and spirals for this data set.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2021

A Fire Community Observatory: Interdisciplinary, AI-informed Post-Fire Rapid Response for Improved Water Cycle Science at Watershed Scale

Wildfire is an ecological disturbance that disrupts the hydrological cycle. In the past few years, a record number of multiple-and-compounding fires have occurred across urban-wildland gradients in the Western United States. Changes to watershed hydrological partitioning in response to fires (infiltration, runoff, evapotranspiration) presents unprecedented challenges to “Water-in-the-West” through negative impacts to water supply and its quality, and is a direct threat to downstream communities, groundwater, and drinking water supply infrastructure. While much work is being done to advance Artificial Intelligence and Machine Learning (AI/ML) use during fires for emergency response (i.e. predict fire movement, direct evacuations), significant potential exists to use AI/ML to address three scientific grand challenges that are rarely addressed in a convergent science context: 1) how to enhance the potential resiliency of a landscape before fire(s), 2) how to cost-effectively and optimally monitor watershed changes after fires, and 3) how to predict future hydrological and biogeochemical trajectories in fire-impacted watershed given climate change. This whitepaper addresses DOE Focal Area 1: Data acquisition and assimilation enabled by machine learning, AI, and advanced methods.

58 GEOSCIENCES↗
DOE OSTI2024

Thermally Insulating Transparent Barrier (THINNER) coatings on single pane windows

Conventional silica aerogel monoliths can provide remarkable thermal insulation but the presence of large pores (> 30 nm) tends to scatter visible light and render the material opaque or translucent instead of transparent. In addition, they are prone to cracking during synthesis and handling which makes them difficult to integrate in products and in particular in window solutions. This project developed two new solgel synthesis methods using silica precursors or preformed silica nanoparticles and ambient drying to produce mesoporous organo-silica monoliths. The monoliths were (i) thermally insulating, (ii) optically transparent, (iii) flexible, and (iv) hydrophobic. They feature porosity ranging from 50% to 90% with narrow pore size distribution with pore less than 20 nm resulting in excellent optical clarity (haze < 2%) and very low thermal conductivity (< 30 mW/mK). Interestingly, not only porosity but also pore size and mass fractal dimension were found to affect the thermal conductivity of the mesoporous silica. The superior transparency of the monoliths was shown to be attributed to dependent scattering among silica nanoparticles. Flexibility was achieved through trimethylchlorosilane surface modification. Furthermore, process scale-up and integration of the ambigel monoliths into window solutions using optically clear adhesives were also demonstrated in 6”x6” double-pane windows. The aerogel and the window solution were shown to be durable to accelerated aging under UV, moisture, and/or temperature gradient and thermal cycling. Overall, the window solution achieved the technical performance and the cost target of $10/sqft set for the SHIELD program. However, scaling to industry relevant scale (> 10’x10’) remains a challenge due to requirements on the drying process and fume hood size and to the propensity of large aerogel slab to crack during drying.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗
DOE OSTI2025

Toward Complete Merger Identification at Cosmic Noon with Deep Learning

As we enter the era of large imaging surveys such as Roman, Rubin, and Euclid, a deeper understanding of potential biases and selection effects in optical astronomical catalogs created with the use of ML-based methods is paramount. This work focuses on a deeper understanding of the performance and limitations of deep learning-based classifiers as tools for galaxy merger identification. We train a ConvNeXT-Pico model on mock HST CANDELS images from the IllustrisTNG50 simulation. Our focus is on a more challenging classification of galaxy mergers and non-mergers at higher redshifts 1 < z < 1.5, including minor mergers and lower mass galaxies down to the stellar mass of 108M⊙. We demonstrate, for the first time, that a deep learning model, such as the one developed in this work, can successfully identify even minor and low mass mergers even at these redshifts. Our model achieves overall accuracy, purity, and completeness of over 73%. We show that some galaxy mergers can only be identified from certain observation angles, leading to a potential upper limit in overall accuracy. Using Grad-CAMs and UMAPs, we more deeply examine the performance and observe a visible gradient in the latent space with stellar mass and specific star formation rate, but no visible gradient with merger mass ratio or merger stage.

Schechter, Aimee L. [U. Colorado, Boulder]↗
DOE OSTI2023

Optimal design of chemoepitaxial guideposts for the directed self-assembly of block copolymer systems using an inexact Newton algorithm

Directed self-assembly (DSA) of block copolymers (BCPs) is one of the most promising developments in the cost-effective production of nanoscale devices. The process makes use of the natural tendency for BCP melts to form nanoscale structures upon phase separation. The phase separation can be directed through the use of chemically patterned substrates to promote the formation of morphologies that are essential to the production of semiconductor devices. Moreover, the design of substrate pattern can be formulated as an optimization problem for which we seek optimal substrate designs that effectively produce given target morphologies. In this paper, we adopt a phase field model given by a nonlocal Cahn–Hilliard partial differential equation (PDE) based on the minimization of the Ohta–Kawasaki free energy, and present an efficient PDE-constrained optimization framework for the optimal design problem. The design variables are the locations of circular- or strip-shaped guiding posts that are used to model the substrate chemical pattern. To solve the ensuing optimization problem, we propose a variant of an inexact Newton conjugate gradient algorithm tailored to this problem. Additionally, we demonstrate the effectiveness of our computational strategy on numerical examples that span a range of target morphologies. Owing to our second-order optimizer and fast state solver, the numerical results demonstrate five orders of magnitude reduction in computational cost over previous work. The efficiency of our framework and the fast convergence of our optimization algorithm enable us to rapidly solve the optimal design problem in not only two, but also three spatial dimensions.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2022

Assessing MP2 frozen natural orbitals in relativistic correlated electronic structure calculations

The high computational scaling with the basis set size and the number of correlated electrons is a bottleneck limiting applications of coupled cluster algorithms, in particular for calculations based on two- or four-component relativistic Hamiltonians, which often employ uncontracted basis sets. This problem may be alleviated by replacing canonical Hartree–Fock virtual orbitals by natural orbitals (NOs). Here, in this paper, we describe the implementation of a module for generating NOs for correlated wavefunctions and, in particular, second order Møller–Plesset perturbation frozen natural orbitals (MP2FNOs) as a component of our novel implementation of relativistic coupled cluster theory for massively parallel architectures [Pototschnig et al. J. Chem. Theory Comput. 17, 5509, (2021)]. Our implementation can manipulate complex or quaternion density matrices, thus allowing for the generation of both Kramers-restricted and Kramers-unrestricted MP2FNOs. Furthermore, NOs are re-expressed in the parent atomic orbital (AO) basis, allowing for generating coupled cluster singles and doubles NOs in the AO basis for further analysis. By investigating the truncation errors of MP2FNOs for both the correlation energy and molecular properties—electric field gradients at the nuclei, electric dipole and quadrupole moments for hydrogen halides HX (X = F–Ts), and parity-violating energy differences for H 2 Z 2 (Z = O–Se)—we find MP2FNOs accelerate the convergence of the correlation energy in a roughly uniform manner across the Periodic Table. It is possible to obtain reliable estimates for both energies and the molecular properties considered with virtual molecular orbital spaces truncated to about half the size of the full spaces.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2024

Optimization using pathwise algorithmic derivatives of electromagnetic shower simulations

Among the well-known methods to approximate derivatives of expectancies computed by Monte-Carlo simulations, averages of pathwise derivatives are often the easiest one to apply. Computing them via algorithmic differentiation typically does not require major manual analysis and rewriting of the code, even for very complex programs like simulations of particle-detector interactions in high-energy physics. However, the pathwise derivative estimator can be biased if there are discontinuities in the program, which may diminish its value for applications. This work integrates algorithmic differentiation into the electromagnetic shower simulation code HepEmShow based on G4HepEm, allowing us to study how well pathwise derivatives approximate derivatives of energy depositions in a sampling calorimeter with respect to parameters of the beam and geometry. We found that when multiple scattering is disabled in the simulation, means of pathwise derivatives converge quickly to their expected values, and these are close to the actual derivatives of the energy deposition. Additionally, we demonstrate the applicability of this novel gradient estimator for stochastic gradient-based optimization in a model example.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗