Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Network simulation”

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 1,369 records · Page 76

Structural relaxation and self-repair behavior in nano-scaled Zr-Cu metallic glass under cyclic loading: Molecular-dynamics simulations

Bulk metallic glasses are generally regarded as highly brittle materials at room temperature, with deformation localized within a few principal shear bands. In this simulation work, it is demonstrated that when the Zr-Cu metallic glass is in a small size-scale, it can deform under cyclic loading in a semi-homogeneous manner without the occurrence of pronounced mature shear bands. Instead, the plastic deformation in simulated samples proceeds via the network-like shear-transition zones (STZs) by the reversible and irreversible structure-relaxations during cyclic loading. Dynamic recovery and reversible/irreversible structure rearrangements occur in the current model, along with annihilation/creation of excessive free volumes. This behavior would in-turn retard the damage growth of metallic glass. Current studies can help to understand the structural relaxation mechanism in metallic glass under loading. The results also imply that the brittle bulk metallic glasses can become ductile with the sample size being reduced. The application of metallic glasses in the form of thin film or nano pieces in micro-electro-mechanical systems (MEMS) could be promising.

36 MATERIALS SCIENCE↗

Modeling of the Coral Microbiome: the Influence of Temperature and Microbial Network

Host-associated microbial communities are shaped by extrinsic and intrinsic factors to the holobiont organism. Environmental factors and microbe-microbe interactions act simultaneously on the microbial community structure, making the microbiome dynamics challenging to predict. The coral microbiome is essential to the health of coral reefs and sensitive to environmental changes. Here, we develop a dynamic model to determine the microbial community structure associated with the surface mucus layer (SML) of corals using temperature as an extrinsic factor and microbial network as an intrinsic factor. The model was validated by comparing the predicted relative abundances of microbial taxa to the relative abundances of microbial taxa from the sample data. The SML microbiome from Pseudodiploria strigosa was collected across reef zones in Bermuda, where inner and outer reefs are exposed to distinct thermal profiles. A shotgun metagenomics approach was used to describe the taxonomic composition and the microbial network of the coral SML microbiome. By simulating the annual temperature fluctuations at each reef zone, the model output is statistically identical to the observed data. The model was further applied to six scenarios that combined different profiles of temperature and microbial network to investigate the influence of each of these two factors on the model accuracy. The SML microbiome was best predicted by model scenarios with the temperature profile that was closest to the local thermal environment, regardless of the microbial network profile. Our model shows that the SML microbiome of P. strigosa in Bermuda is primarily structured by seasonal fluctuations in temperature at a reef scale, while the microbial network is a secondary driver. Coral microbiome dysbiosis (i.e., shifts in the microbial community structure or complete loss of microbial symbionts) caused by environmental changes is a key player in the decline of coral health worldwide. Multiple factors in the water column and the surrounding biological community influence the dynamics of the coral microbiome. However, by including only temperature as an external factor, our model proved to be successful in describing the microbial community associated with the surface mucus layer (SML) of the coral P. strigosa. The dynamic model developed and validated in this study is a potential tool to predict the coral microbiome under different temperature conditions.

59 BASIC BIOLOGICAL SCIENCES↗

An Investigation of LES Wall Modeling for Rayleigh–Bénard Convection via Interpretable and Physics-Aware Feedforward Neural Networks with DNS

Abstract The traditional approach of using the Monin–Obukhov similarity theory (MOST) to model near-surface processes in large-eddy simulations (LESs) can lead to significant errors in natural convection. In this study, we propose an alternative approach based on feedforward neural networks (FNNs) trained on output from direct numerical simulation (DNS). To evaluate the performance, we conduct both a priori and a posteriori tests. In the a priori (offline) tests, we compare the statistics of the surface shear stress and heat flux, computed from filtered DNS input variables, to the stress and flux obtained from the filtered DNS. Additionally, we investigate the importance of various input features using the Shapley additive explanations value and the conditional average of the filter grid cells. In the a posteriori (online) tests, we implement the trained models in the System for Atmospheric Modeling (SAM) LES and compare the LES-generated surface shear stress and heat flux with those in the DNS. Our findings reveal that vertical velocity, a traditionally overlooked flow quantity, is one of the most important input features for determining the wall fluxes. Increasing the number of input features improves the a priori test results but does not always improve the model performance in the a posteriori tests because of the differences in input variables between the LES and DNS. Last, we show that physics-aware FNN models trained with logarithmic and scaled parameters can well extrapolate to more intense convection scenarios than in the training dataset, whereas those trained with primitive flow quantities cannot. Significance Statement The traditional near-surface turbulence model, based on a shear-dominated boundary layer flow, does not represent near-surface turbulence in natural convection. Using a feedforward neural network (FNN), we can construct a more accurate model that better represents the near-surface turbulence in various flows and reveals previously overlooked controlling factors and process interactions. Our study shows that the FNN-generated models outperform the traditional model and highlight the importance of the near-surface vertical velocity. Furthermore, the physics-aware FNN models exhibit the potential to extrapolate to convective flows of various intensities beyond the range of the training dataset, suggesting their broader applicability for more accurate modeling of near-surface turbulence.

54 ENVIRONMENTAL SCIENCES↗

Photon Reconstruction in the Belle II Calorimeter Using Graph Neural Networks

We present the study of a fuzzy clustering algorithm for the Belle II electromagnetic calorimeter using Graph Neural Networks. We use a realistic detector simulation including simulated beam backgrounds and focus on the reconstruction of both isolated and overlapping photons. We find significant improvements of the energy resolution compared to the currently used reconstruction algorithm for both isolated and overlapping photons of more than 30% for photons with energies E γ < 0.5 GeV and high levels of beam backgrounds. Overall, the GNN reconstruction improves the resolution and reduces the tails of the reconstructed energy distribution and therefore is a promising option for the upcoming high luminosity running of Belle II.

Calorimeter↗

The landscape of disc outflows from black hole–neutron star mergers

ABSTRACT We investigate mass ejection from accretion discs formed in mergers of black holes (BHs) and neutron stars (NSs). The third observing run of the LIGO/Virgo interferometers provided BH–NS candidate events that yielded no electromagnetic (EM) counterparts. The broad range of disc configurations expected from BH–NS mergers motivates a thorough exploration of parameter space to improve EM signal predictions. Here we conduct 27 high-resolution, axisymmetric, long-term hydrodynamic simulations of the viscous evolution of BH accretion discs that include neutrino emission/absorption effects and post-processing with a nuclear reaction network. In the absence of magnetic fields, these simulations provide a lower limit to the fraction of the initial disc mass ejected. We find a nearly linear inverse dependence of this fraction on disc compactness (BH mass over initial disc radius). The dependence is related to the fraction of the disc mass accreted before the ouflow is launched, which depends on the disc position relative to the innermost stable circular orbit. We also characterize a trend of decreasing ejected fraction and decreasing lanthanide/actinide content with increasing disc mass at fixed BH mass. This trend results from a longer time to reach weak freezout and an increasingly dominant role of neutrino absorption at higher disc masses. We estimate the radioactive luminosity from the disc outflow alone available to power kilonovae over the range of configurations studied, finding a spread of two orders of magnitude. For most of the BH–NS parameter space, the disc outflow contribution is well below the kilonova mass upper limits for GW190814.

79 ASTRONOMY AND ASTROPHYSICS↗

Long-lived neutron-star remnants from asymmetric binary neutron star mergers: element formation, kilonova signals and gravitational waves

We present 3D general-relativistic neutrino-radiation hydrodynamics simulations of two asymmetric binary neutron star mergers producing long-lived neutron stars remnants and spanning a fraction of their cooling time scale. The mergers are characterized by significant tidal disruption with neutron rich material forming a massive disc around the remnant. The latter develops one-armed dynamics that is imprinted in the emitted kilo-Hertz gravitational waves. Angular momentum transport to the disc is initially driven by spiral-density waves and enhanced by turbulent viscosity and neutrino heating on longer timescales. The mass outflows are composed by neutron-rich dynamical ejecta of mass ∼10 −3 –10 −2 M ⊙ followed by a persistent spiral-wave/neutrino-driven wind of ≳ 10 −2 M ⊙ with material spanning a wide range of electron fractions, ∼0.1–0.55. Dynamical ejecta (winds) have fast velocity tails up to ∼0.8 (∼0.4) c. The outflows are further evolved to days timescale using 2D ray-by-ray radiation-hydrodynamics simulations that include an online nuclear network. We find complete r-process yields and identify the production of 56 Ni and the subsequent decay chain to 56 Co and 56 Fe. Synthetic kilonova light curves predict an extended (near-) infrared peak a few days postmerger originating from r-process in the neutron-rich/high-opacity ejecta and UV/optical peaks at a few hours (ten minutes) postmerger originating from weak r-process (free-neutron decay) in the faster ejecta components. Additionally, the fast tail of tidal origin generates kilonova afterglows potentially detectable in radio and X band on a few to ten years time scale. Quantitative effects originating from the tidal disruption merger dynamics are reflected in the multimessenger emissions.

abundances↗

Assessing Constitutive Models for Prediction of High-Temperature Flow Behavior with a Perspective of Alloy Development

The utility of different constitutive models describing high-temperature flow behavior has been evaluated from the perspective of alloy development. Strain compensated Arrhenius model, modified Johnson–Cook (MJC) model, model D8A and artificial neural network (ANN) have been used to describe flow behavior of different model alloys. These alloys are four grades of SS 316LN with different nitrogen contents ranging from 0.07 to 0.22%. Grades with 0.07%N and 0.22%N have been used to determine suitable material constants of the constitutive equations and also to train the ANN model. While the ANN model has been developed with chemical composition as a direct input, the MJC and D8A models have been amended to incorporate the effect of nitrogen content on flow behavior. The prediction capabilities of all models have been validated using the experimental data obtained from grades containing 0.11%N and 0.14%N. The comparative analysis demonstrates that ‘N-amended D8A’ and ‘N-amended MJC’ are preferable to the ANN model for predicting flow behavior of different grades of 316LN. The work provides detailed insights into the usual statistical error analysis technique and frames five additional criteria which must be considered when a model is analyzed from the perspective of alloy development.

36 MATERIALS SCIENCE↗

Sensitivity Analysis of Drivers Water Shortage in the Los Angeles Region During Drought

The code and detailed step-by-step instructions for generating the model output data, processing results, and analysis and plotting are provided at https://github.com/IMMM-SFA/Ferencz_et_al_2026_ER_Water. The PyArtes model is a python adaptation of the Artes model. PyArtes uses many of the same input data and optimization model architecture as Artes. Documentation for the PyArtes model is provided in the Supplement to the paper. The primary data product are simulated monthly water shortages for indoor and outdoor demand under a large ensemble of drought scenarios (>13,000). The droughts are hypothetical and are not based on historical time series data of supply sources - though historical data did help inform ranges explored for supply parameters. Demands are informed by recent 2017-2021 water supply data. Demands used for the model can be accessed at https://github.com/IMMM-SFA/Ferencz_et_al_2026_ER_Water. Simulations resolve demand for over 90 water providers in the study region. The results report 36 months of water shortage data for each indoor and outdoor demand node. The study also developed a multilayer perceptron (MLP) neural network trained on a subset of the simulated shortage ensemble to emulate worst annual water shortage for a given set of parameter multipliers -- provided the parameter values fall within the ranges sampled in the ensemble. Emulated water shortages for synthetic ensembles are in the MLP-generated shortages folder. The MLP model was used to generate larger ensembles to support Sobol analysis that would have been extremely computationally expensive to simulate. Datasets provided in this repository*: Simulated shortages. These results are used for the analysis for Figures 5, 8, and 9 in the paper, and also to train the MLP emulator. .zip file containing outputs for the 13,312 scenario ensemble. Separate .csv files for indoor and outdoor shortage for each scenario. Rows = demand ids (~100), Columns = months (36) Units = acre-feet/month of shortage (shortage = monthly demand - supply). 1 acft = 1233.48 m^3 .csv files of aggregated shortages derived from the 13,312 ensemble Rows = scenarios (13,312), Columns = demand ids (~100) Units = acre-feet/year (either worst annual shortage or total shortage over the 3-year drought) .csv file of the parameter multipliers scenarios for the ensemble .csv file of the parameter ranges and baseline values the multipliers were applied to MLP-generated shortages. These results are used for Figures 4, 6, and 7 in the paper. mwd higher folder: scenario ensembles, emulated worst year total shortages (acft), and Sobol results Emulated shortages. Rows = scenarios, columns = demand ids, units acft Sobol results. Rows = demand ids, columns Sobol (S1, ST, or 95% confidence interval) value for each parameter mwd lower folder: scenario ensembles, emulated worst year total shortages (acft), and Sobol results same organization as mwd higher MLP performance: performance metrics (R^2, RMSE, BIAS, MAPE) for the testing subset (20% or 2,662 scenarios) and simulated vs emulated worst year shortage (acre-feet/year) for every demand node, MWD wholesale regions, and the entire study region (LAC). Supporting data for figures. Figure plotting scripts in the associated GitHub repo. These files support analysis and visualization. Geospatial Data used for plotting simulated water shortages and Sobol results. Dictionary of full names for demand nodes in the model and estimates of water supply by source type informed by Artes input files and California Urban Water Management Planning data: https://water.ca.gov/Programs/Water-Use-And-Efficiency/Urban-Water-Use-Efficiency/Urban-Water-Management-Plans *Readme files provided for each folder.

drought↗

Deep Generative Models for Fast Photon Shower Simulation in ATLAS

The need for large-scale production of highly accurate simulated event samples for the extensive physics programme of the ATLAS experiment at the Large Hadron Collider motivates the development of new simulation techniques. Building on the recent success of deep learning algorithms, variational autoencoders and generative adversarial networks are investigated for modelling the response of the central region of the ATLAS electromagnetic calorimeter to photons of various energies. The properties of synthesised showers are compared with showers from a full detector simulation using GEANT4 . Both variational autoencoders and generative adversarial networks are capable of quickly simulating electromagnetic showers with correct total energies and stochasticity, though the modelling of some shower shape distributions requires more refinement. This feasibility study demonstrates the potential of using such algorithms for ATLAS fast calorimeter simulation in the future and shows a possible way to complement current simulation techniques.

97 MATHEMATICS AND COMPUTING↗

Symbolic diagnostics to interpret and analyze neural network models

Embedded machine-learned models (EMLMs) have the promise to improve the predictive accuracy of engineering simulators in environments of national interest. EMLMs often comprise complex input-output maps (e.g., neural networks), which make them unamenable to rigorous analysis and generally difficult to interpret. In the face of decades of theory, this lack of interpretability is a significant barrier to building confidence in these models. This work outlines an approach to interpret EMLMs using sparse polynomial regression for comparison with theoretical understanding. To do so, we build on the concept of Locally Interpretable Model-agnostic Explanations (LIME) using physics-informed clustering, prototype selection, and library construction. While general, we demonstrate our method on tensor-basis neural networks used in Reynolds-Averaged Navier-Stokes simulations of hypersonic fluid flows. Results are presented for a simulated toy model and for direct numerical simulations (DNS) of turbulent flows over a flat plate.

97 MATHEMATICS AND COMPUTING↗

DGs for Service Restoration to Critical Loads in a Secondary Network

During a major outage in a secondary network distribution system, distributed generators (DGs) connected to the primary feeders as well as the secondary network can be used to serve critical loads. This paper proposed a resilience-oriented method to determine restoration strategies for secondary network distribution systems after a major disaster. Technical issues associated with the restoration process are analyzed, including the operation of network protectors, inrush currents caused by the energization of network transformers, synchronization of DGs to the network, and circulating currents among DGs. A look-ahead load restoration framework is proposed, incorporating technical issues associated with secondary networks, limits on DG capacity and generation resources, dynamic constraints, and operational limits. The entire outage duration is divided into a sequence of periods. Restoration strategies can be adjusted at the beginning of each period using the latest information. Numerical simulation of the modified IEEE 342-node low voltage networked test system is performed to validate the effectiveness of the proposed method.

Xu, Yin (ORCID:0000000209459024)↗

Rapid neutron and gamma-ray source localization using machine learning

Rapid localization of radiation sources is critical for applications including nuclear emergency response, safeguards, and security. However, conventional imaging systems such as neutron scatter cameras and Compton cameras depend on rare coincidence events, which often result in long acquisition times. In this work, we address the challenge of rapid source localization by developing a machine learning approach to predict the direction of a single radiation source using only count rates from an array of neutron and gamma-ray detectors. The proposed model is a fully connected neural network (FCNN) trained using Monte Carlo simulation data from a 252 Cf source. The model hyperparameters are optimized with a small set of routine 252 Cf measurements. We benchmarked the performance of the trained and optimized machine learning model using additional 252 Cf , 137 Cs , and PuBe measurements under laboratory conditions with varying source-detector configurations. For these measurements, the machine learning model achieved a mean localization error smaller than 30° with 3 x 10 3 system counts, corresponding to 8 s measurement time for the imaging system used in this work. In this low-statistics regime, the method outperformed traditional scatter-based imaging by more than 75% in localization accuracy for the evaluated measurement configurations. These results demonstrate that a machine learning-based approach can significantly reduce the time required for accurate single-source localization, providing a robust and computationally efficient alternative to traditional imaging systems in time-critical nuclear security and emergency response scenarios.

Gamma-ray imaging↗

THE SPIN AND ORIENTATION OF DARK MATTER HALOS WITHIN COSMIC FILAMENTS

Clusters, filaments, sheets, and voids are the building blocks of the cosmic web. Forming dark matter halos respond to these different large-scale environments, and this in turn affects the properties of galaxies hosted by the halos. It is therefore important to understand the systematic correlations of halo properties with the morphology of the cosmic web, as this informs both about galaxy formation physics and possible systematics of weak lensing studies. In this study, we present and compare two distinct algorithms for finding cosmic filaments and sheets, a task which is far less well established than the identification of dark matter halos or voids. One method is based on the smoothed dark matter density field and the other uses the halo distributions directly. We apply both techniques to one high-resolution N-body simulation and reconstruct the filamentary/sheet like network of the dark matter density field. We focus on investigating the properties of the dark matter halos inside these structures, in particular, on the directions of their spins and the orientation of their shapes with respect to the directions of the filaments and sheets. We find that both the spin and the major axes of filament halos with masses approx<10{sup 13} h {sup -1} M{sub sun} are preferentially aligned with the direction of the filaments. The spins and major axes of halos in sheets tend to lie parallel to the sheets. There is an opposite mass dependence of the alignment strength for the spin (negative) and major (positive) axes, i.e. with increasing halo mass the major axis tends to be more strongly aligned with the direction of the filament, whereas the alignment between halo spin and filament becomes weaker with increasing halo mass. The alignment strength as a function of the distance to the most massive node halo indicates that there is a transit large-scale environment impact: from the two-dimensional collapse phase of the filament to the three-dimensional collapse phase of the cluster/node halo at small separation. Overall, the two algorithms for filament/sheet identification investigated here agree well with each other. The method based on halos alone can be easily adapted for use with observational data sets.

79 ASTRONOMY AND ASTROPHYSICS↗

HOLISMOKES IX. Neural network inference of strong-lens parameters and uncertainties from ground-based images

Modeling of strong gravitational lenses is a necessity for further applications in astrophysics and cosmology. With the large number of detections in current and upcoming surveys, such as the Rubin Legacy Survey of Space and Time (LSST), it is pertinent to investigate automated and fast analysis techniques beyond the traditional and time-consuming Markov chain Monte Carlo sampling methods. Building upon our (simple) convolutional neural network (CNN), we present here another CNN, specifically a residual neural network (ResNet), that predicts the five mass parameters of a singular isothermal ellipsoid (SIE) profile (lens center x and y, ellipticity e x and e y , Einstein radius θ E ) and the external shear (γ ext, 1 , γ ext, 2 ) from ground-based imaging data. In contrast to our previous CNN, this ResNet further predicts the 1σ uncertainty for each parameter. To train our network, we use our improved pipeline to simulate lens images using real images of galaxies from the Hyper Suprime-Cam Survey (HSC) and from the Hubble Ultra Deep Field as lens galaxies and background sources, respectively. We find very good recoveries overall for the SIE parameters, especially for the lens center in comparison to our previous CNN, while significant differences remain in predicting the external shear. From our multiple tests, it appears that most likely the low ground-based image resolution is the limiting factor in predicting the external shear. Given the run time of milli-seconds per system, our network is perfectly suited to quickly predict the next appearing image and time delays of lensed transients. Therefore, we use the network-predicted mass model to estimate these quantities and compare to those values obtained from our simulations. Unfortunately, the achieved precision allows only a first-order estimate of time delays on real lens systems and requires further refinement through follow-up modeling. Nonetheless, our ResNet is able to predict the SIE and shear parameter values in fractions of a second on a single CPU, meaning that we are able to efficiently process the huge amount of galaxy-scale lenses expected in the near future.

79 ASTRONOMY AND ASTROPHYSICS↗

Combining molecular dynamics with mesoscopic Green’s function reaction dynamics simulations

In many reaction-diffusion processes, ranging from biochemical networks, catalysis, to complex self-assembly, the spatial distribution of the reactants and the stochastic character of their interactions are crucial for the macroscopic behavior. The recently developed mesoscopic Green’s Function Reaction Dynamics (GFRD) method enables efficient simulation at the particle level provided the microscopic dynamics can be integrated out. Yet, many processes exhibit non-trivial microscopic dynamics that can qualitatively change the macroscopic behavior, calling for an atomistic, microscopic description. We propose a novel approach that combines GFRD for simulating the system at the mesoscopic scale where particles are far apart, with a microscopic technique such as Langevin dynamics or Molecular Dynamics (MD), for simulating the system at the microscopic scale where reactants are in close proximity. This scheme defines the regions where the particles are close together and simulated with high microscopic resolution and those where they are far apart and simulated with lower mesoscopic resolution, adaptively on the fly. The new multi-scale scheme, called MD-GFRD, is generic and can be used to efficiently simulate reaction-diffusion systems at the particle level.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

MASSIVELY PARALLEL FULLY COUPLED IMPLICIT MODELING OF COUPLED THERMAL-HYDROLOGICAL-MECHANICAL PROCESSES FOR ENHANCED GEOTHERMAL SYSTEM RESERVOIRS

Development of enhanced geothermal systems (EGS) will require creation of a reservoir of sufficient volume to enable commercial-scale heat transfer from the reservoir rocks to the working fluid. A key assumption associated with reservoir creation/stimulation is that sufficient rock volumes can be hydraulically fractured via both tensile and shear failure, and more importantly by reactivation of naturally existing fractures (by shearing) to create the reservoir. The advancement of EGS greatly depends on our understanding of the dynamics of the intimately coupled rock-fracture-fluid system and our ability to reliably predict how reservoirs behave under stimulation and production. In order to increase our understanding of how reservoirs behave under these conditions, we have developed a physics-based rock deformation and fracture propagation simulator by coupling a discrete element model (DEM) for fracturing with a continuum multiphase flow and heat transport model. In DEM simulations, solid rock is represented by a network of discrete elements (often referred as particles) connected by various types of mechanical bonds such as springs, elastic beams or bonds that have more complex properties (such as stress-dependent elastic constants). Fracturing is represented explicitly as broken bonds (microcracks), which form and coalesce into macroscopic fractures when external load is applied. DEM models have been applied to a very wide range of fracturing processes from the molecular scale (where thermal fluctuations play an important role) to scales on the order of 1 km or greater. In this approach, the continuum flow and heat transport equations are solved on an underlying fixed finite element grid with evolving porosity and permeability for each grid cell that depends on the local structure of the discrete element network (such as DEM particle density). The fluid pressure gradient exerts forces on individual elements of the DEM network, which therefore deforms and fractures. Such deformation/fracturing in turn changes the permeability, which again changes the evolution of fluid pressure, coupling the two phenomena. The intimate coupling between fracturing and fluid flow makes the meso-scale DEM simulations necessary, as these methods have substantial advantages over conventional continuum mechanical models of elastic rock deformation. The challenges that must be overcome to simulate EGS reservoir stimulation, preliminary results, progress to date and near future research directions and opportunities will be discussed.

15 GEOTHERMAL ENERGY↗

An algorithm for physics informed scan path optimization in additive manufacturing

Site specific microstructure control is a critical research area within the field of additive manufacturing due to its potential to revolutionize part performance. One way to achieve site specific microstructure control is through control of the solidification conditions via the construction of intricate scan paths; however, the search space for such a problem is large. Previous attempts only considered the solidification conditions at the top surface while also requiring either lots of manual-fine tuning or large amounts of computational resources. This paper introduces a general method for scan path optimization which considers the solidification conditions in the bulk of the material without an increase in computational expense. This method consists of three core components:1. A heat transfer model for simulating the temperature field at a given time.2. A surrogate model which takes scan pattern information and temperature data and predicts the solidification conditions of the bulk as well as the meltpool depths for a spot melt.3. A decision algorithm to decide which spot melt should be printed next based on the outputs of the surrogate model.Each of these components can be changed without changing the overall method. Within this work, this method is applied in the creation of an algorithm containing a semi-analytic heat transfer model to simulate the temperature field, a fully convolutional neural network (FCNN) as the surrogate model, and a greedy decision algorithm. The resulting algorithm produced complex scan patterns which gave strong results for simulated microstructure control.

36 MATERIALS SCIENCE↗

Toward machine-learning-assisted PW-class high-repetition-rate experiments with solid targets

We present progress in utilizing a machine learning (ML) assisted optimization framework to study the trends in a parameter space defined by spectrally shaped, high-intensity, petawatt-class (8 J, 45 fs) laser pulses interacting with solid targets and give the first simulation-based overview of predicted trends. A neural network (NN) incorporating uncertainty quantification is trained to predict the number of hot electrons generated by the laser–target interaction as a function of pulse shaping parameters. The predictions of this NN serve as the basis function for a Bayesian optimization framework to navigate this space. For post-experimental evaluation, we compare two separate neural network (NN) models. One is based solely on data from experiments, and the other is trained only on ensemble particle-in-cell simulations. Reviewing the predicted and observed trends across the experiment-capable laser parameter search space, we find that both ML models predict a maximal increase in hot electron generation at a level of approximately 12%–18%; however, no statistically significant enhancement was observed in experiments. On direct comparison of the NN models, the average discrepancy is 8.5%, with a maximum of 30%. Since shot-to-shot fluctuations in experiments affect the observations, we evaluate the behavior of our optimization framework by performing virtual experiments that vary the number of repeated observations and the noise levels. Here, we discuss the implications of such a framework for future autonomous exploration platforms in high-repetition-rate experiments.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗