Search NASA⌕ Search

SEARCH · Search NASA

Results for “time segmentation”

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 109 records · Page 6
All sourcesNASA NTRSDOE OSTI
DOE OSTI2025

Impact of Time Dependent Reactor and Sensor Physics on Core Power Synthesis

Online synthesis of the power distribution is critical in the operation and control of nuclear power reactors to ensure that the core is operating within safety margins, and to provide essential knowledge associated with the burnup of the fuel. In light water reactors (LWRs), power synthesis is achieved by using some a priori knowledge of the state of the reactor core and updating based on the signals coming from in-core sensors—namely, self-powered neutron detectors (SPNDs). This report aims to study the effects of fuel burnup and sensor degradation on the ability to accurately synthesize the power distribution in a LWR. Several modeling tools were used to simulate power synthesis based on the responses of SPNDs, with emitters made out of Rh or V. A representative pressurized water reactor low-enriched uranium (LEU) core was modeled using the Polaris/Purdue Advanced Reactor Core Simulator (PARCS) approach. The Monte Carlo N-Particle Transport 6 (MCNP6) code was used, as well, to calculate response functions between different segments of fuel to individual SPNDs; this is a crucial parameter for power synthesis. The Oak Ridge Isotope GENeration (ORIGEN) package in the Standardized Computer Analyses for Licensing Evaluation (SCALE) code was used to model the time-dependent isotopic transmutation in the SPND emitters. All these data were fed into a custom code that enacted the point-based iterative (PBI) method to simulate power synthesis. Developmental work was also performed on high-fidelity SPND models in the GEometry ANd Tracking 4 (Geant4) code, which enables higher-accuracy modeling of the current responses from SPNDs. In this work, five sets of time-dependent power synthesis test cases were conducted. In these test cases, systematic changes in the input conditions enabled an analysis of the effect of (1) slightly inaccurate a priori power distribution assumptions with respect to fuel burnup, (2) highly inaccurate a priori assumptions with respect to fuel burnup (such that burnup is not included in the a priori assumed distribution), and (3) differences between Rh and V SPNDs in terms of downstream consequences of the transmutation in the emitters. The authors discovered that one may permissibly have slightly inaccurate a priori assumptions of the fuel burnup (such that the level of burnup may be slightly under- or over-approximated by the accumulated burnup in approximately 9.3 full power days), but to not account for burnup at all in the a priori assumption leads to severe levels of error, approaching 25% at maximum. The authors also discovered that V SPNDs are extraordinarily robust in the low-enriched uranium fuel cycle considered in this modeling work, whereas Rh SPNDs undergo significant transmutation that can result in large errors in the synthesized power distribution.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗
DOE OSTI2026

Intelligent Experiments through Real-Time AI: Fast Data Processing and Autonomous Detector Control for High-Energy Nuclear Experiments

The aim of this project is to develop software and hardware for fast real-time data processing and autonomous detector control and calibration for the sPHENIX and the future EIC experiments. Below summarizes Georgia Tech team efforts in the past year: 1. We developed a real-time clustering algorithm and FPGA-based pipeline architecture for processing fired pixel data from ALPIDE sensors in sPHENIX experiments. Our Columnar Clustering Co-Design introduces a hardware-aware, stream-friendly approach that segments pixel data by column pairs using a Column Pair Clustering (CPC) strategy, followed by Cluster Stitching to merge adjacent subclusters. Implemented in Vitis HLS, the pipeline comprises five stages—read-in, subclustering, stitching, analysis, and write-out—connected by tagged HLS streams with custom end-of-event signaling for robust synchronization. We designed a pipelined dataflow model optimized for throughput, low latency, and minimal buffering, enabling scalable clustering across events of arbitrary size. Our system maintains spatial precision via center-of-mass and shape key extraction and efficiently handles edge cases such as fragmented or nested clusters. Compared against DBSCAN in both software and hardware, our approach demonstrates competitive performance under FPGA constraints. 2. We also conducted a comprehensive algorithm-to-hardware co-design of connected component analysis tailored for sPHENIX experiments, focusing on real-time, low-latency processing using FPGAs and High-Level Synthesis (HLS). Starting from a Python-based particle tracking pipeline, the team translated the core logic—graph traversal via DFS and Union-Find—into an HLS-compatible C++ model, replacing dynamic memory and recursion with static arrays and pipelined control flow. The final design includes a fully streamed and dataflow-compatible Union-Find kernel optimized across five iterations, incorporating loop pipelining, array partitioning, AXI/FIFO interface tuning, and function flattening. Experimental results show up to 14.8× speedup over the CPU baseline, reducing per-graph latency to 1.58 μs and demonstrating strong resource efficiency with only ~7k LUTs and zero BRAM usage. The design maintains functional correctness against the Python reference using a Python-based C-simulation framework and Mean Squared Error metrics. This work validates the potential of HLS-driven FPGA designs for edge-level HEP data acquisition, laying a scalable foundation for future integration with real-time detector pipelines and multi-graph processing systems.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗
DOE OSTI2025

Are there differences in public interest toward automated vehicles’ ownership in burdened and non-burdened communities?

This study examines differences in public interest toward automated vehicle (AV) adoption between burdened communities (BCs) and non-burdened communities (non-BCs). The study uniquely captures travel behavior by integrating the census tract-level community indicators with the 2017–2019 Puget Sound Household Travel Survey. Public interest in AV ownership is modeled as a five-level ordered outcome, ranging from ‘Not interested at all’ to ‘Very interested.’ Segmented ordered logit models and Random Forest are used to identify key drivers and barriers to AV adoption across BCs and non-BCs. Results show that longer commute times, lack of personal vehicles, and concerns about AV performance in adverse weather conditions are significantly associated with higher interest levels toward owning AVs in BCs. In conclusion, the findings offer insights for developing targeted policies to promote AV adoption and address differences between BCs and non-BCs in accessing emerging transportation technologies.

Automated vehicles↗
DOE OSTI2025

Active Learning Meets Foundation Models: Fast Remote Sensing Data Annotation for Object Detection

Object detection in remote sensing demands extensive, high-quality annotations—a process that is both labor-intensive and time-consuming. In this work, we introduce a real-time active learning and semi-automated labeling framework that leverages foundation models to streamline dataset annotation for object detection in remote sensing imagery. For example, by integrating a Segment Anything Model (SAM), our approach generates mask-based bounding boxes that serve as the basis for dual sampling: (a) uncertainty estimation to pinpoint challenging samples, and (b) diversity assessment to ensure broad data coverage. Furthermore, our Dynamic Box Switching Module (DBS) addresses the well-known cold start problem for object detection models by replacing its suboptimal initial predictions with SAM-derived masks, thereby enhancing early-stage localization accuracy. Extensive evaluations on multiple remote sensing datasets plus a real-world user study, demonstrate that our framework not only reduces annotation effort, but also significantly boosts detection performance compared to traditional active learning sampling methods. The code for training and the user interface will be made available.

Burges, Marvin [ORNL] (ORCID:0000000312690769)↗
DOE OSTI2026

Impact of Time-Dependent Reactor and Sensor Physics on Core Power Synthesis (Rev.1)

Online synthesis of power distribution is critical in the operation and control of nuclear power reactors to ensure that the core is operating within safety margins and to provide essential knowledge associated with the burnup of the fuel. In light-water reactors, power synthesis is achieved by using some a priori knowledge of the state of the reactor core and updating based on the signals coming from in-core sensors—namely, self-powered neutron detectors (SPNDs). This report examines the effects of fuel burnup and sensor degradation on the ability to accurately synthesize the power distribution in a pressurized water reactor (PWR), considering the typical low-enriched uranium (LEU, 3%-5% enrichment) fuel cycle as well as the higher enrichment LEU+ (5%-8% enrichment) fuel cycle. Several modeling tools were used to simulate power synthesis based on the responses of SPNDs, with emitters made out of Rh or V. A representative PWR LEU core was modeled using the Polaris/Purdue Advanced Reactor Core Simulator (PARCS) approach. The Monte Carlo N-Particle Transport 6 (MCNP6) code was used as well to calculate response functions between different segments of fuel to individual SPNDs; this is a crucial parameter for power synthesis. The Oak Ridge Isotope GENeration (ORIGEN) package in the Standardized Computer Analyses for Licensing Evaluation (SCALE) code was used to model the time-dependent isotopic transmutation in the SPND emitters. All these data were fed into a custom code that enacted the point-based iterative method to simulate power synthesis. Developmental work was also performed on high-fidelity SPND models in the GEometry ANd Tracking 4 (Geant4) code, which enables higher-accuracy modeling of the current responses from SPNDs. In this work, five sets of time-dependent power synthesis test cases were conducted. In these test cases, systematic changes in the input conditions enabled an analysis of the effect of (1) slightly inaccurate a priori power distribution assumptions with respect to fuel burnup, (2) highly inaccurate a priori power distribution assumptions with respect to fuel burnup (such that burnup is not included in the a priori assumed distribution), and (3) differences between Rh and V SPNDs in terms of downstream consequences of the transmutation in the emitters and the extended nature of the LEU+ fuel cycle in comparison with LEU. The authors discovered that one may permissibly have slightly inaccurate a priori assumptions of the fuel burnup (such that the level of burnup may be slightly underapproximated or overapproximated by the accumulated burnup in approximately 9.3 full power days), but to not account for burnup at all in the a priori assumptions leads to severe levels of error, approaching 25% at maximum (for LEU). The authors also discovered that V SPNDs are extraordinarily robust in both the LEU and LEU+ fuel cycles considered in this modeling work, whereas Rh SPNDs undergo significant transmutation that can result in large errors in the synthesized power distribution.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗
DOE OSTI2026

Creating a Training Dataset for Semantic Segmentation of Canal Networks for Irrigation Modernization

Canal infrastructure has provided critical irrigation water to the western United States for over a century. To continue providing vital water resources to the semi-arid West, irrigation systems must undergo maintenance and modernization. Many canal companies are resource-constrained, and because funding opportunities often require detailed knowledge of existing infrastructure, they can struggle to secure financial capital. We address this problem by creating training data for a semantic segmentation deep learning model to map canal networks throughout the western United States. To create a diverse and robust training dataset, we labelled 1-m NAIP imagery with the locations of no canals, wet canals, and dry/vegetated canals. Since creating these datasets is time consuming, we first developed a preprocessing methodology to identify canals within our four study areas. We used NAIP imagery and provided canal centerline data to buffer, standardize, and cluster the imagery, automating the labeling process as much as possible. However, this still required manual cleaning and manual classification of canal type. Challenges arose when canals were interrupted (e.g., road culverts or piped sections) or when nearby features shared similar characteristics (e.g., irrigated fields, trees, and shadows). Combining automated preprocessing with manual refinement produced four detailed canal masks to be used in the semantic segmentation model developed by Richard Tapia.

13 - HYDRO ENERGY↗
DOE OSTI2025

Strip-based Scintillation Detector for Dual-readout High-granularity Calorimetry

New calorimeter technology is being developed for future collider experiments. We are developing a calorimeter that integrates two technologies, a high-granularity calorimeter and a dual-readout calorimeter, and has high time resolution at a picosecond level. The proposed calorimeter is a sampling calorimeter based on scintillation and Cherenkov detectors with high-granularity readout. The Cherenkov detector has a picosecond level time resolution. For the high-granularity scintillation detector, scintillator strips of 300mm × 30mm ×3mm-thick are aligned horizontally and vertically to realize an effective 30mm-square-cell segmentation. Prototype strips with different designs and scintillator materials were tested to compare the performance.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗
DOE OSTI2025

Sequence modeling of higher-order wave modes of quasi-circular, spinning, non-precessing binary black hole mergers

Higher-order gravitational wave modes from quasi-circular, spinning, non-precessing binary-black-hole (BBH) mergers encode rich information about the nonlinear dynamics of strong-field gravity. We present a transformer-based sequence-completion surrogate that, given an early-inspiral segment, forecasts the subsequent late inspiral, merger, and ringdown. The intended applications are (i) patching or completing expensive or interrupted numerical-relativity (NR) simulations and (ii) providing late-time cross-checks and rapid hybridization studies. The training set is built from the NRHybSur3dq8 surrogate, which provides spherical-harmonic modes up to $\ell$ ≤ 4 (excluding (4, 0) and (4,±1), and including (5, 5)) for mass ratios q ≤ 8, dimensionless spin components s$^{z}_{1,2}$ ϵ[–0.8, 0.8], and inclination angles θ ϵ [0, π]. Waveforms are supplied on the interval t ϵ [–5000M, –100 M) and the model autoregressively generates the plus and cross polarizations (h + , h x ) on t ϵ [–100 M, 130M]. Training on the Delta supercomputer with 16 NVIDIA A100 GPUs required ~15 h on more than 14 million hybrid waveforms. Evaluation on a held-out test set of 840,000 samples yields mean and median overlaps of 0.996 and 0.997, respectively, with respect to the surrogate ground truth.

black-hole merger↗
DOE OSTI2025

Development of High-Granularity Dual-Readout Calorimetry with psec Timing

Dual-readout and particle flow algorithm (PFA) are technologies proposed for precise jet energy measurement in future colliders. While PFA requires highly granular calorimeters, dual-readout has mainly been used with fiber-based calorimeters that do not have highly segmented capabilities. It is still non-trivial to combine these two technologies in one calorimeter system because of the use of fibers in most of the dualreadout calorimeters, which is not compatible with the high granularity requirement of PFA technologies. The aim of this study is to develop a novel calorimetry that combines dual-readout and PFA by adopting a highly segmented tile-based configuration. This paper compares the improvement of energy resolution using dual-readout approach across several configurations of highly granular hadron calorimeters through simulation. Results indicate that setups with fine sampling and close placement of scintillators and Cherenkov detectors improve dual-readout performance.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗
DOE OSTI2026

Real-Time GPU-Accelerated OFDR With an Integrated Auxiliary Interferometer

A GPU-accelerated optical frequency domain reflectometry (OFDR) system with an improved integrated auxiliary interferometer is proposed. Unlike conventional approaches that require separate auxiliary interferometers and multiple detection channels, the proposed OFDR system embeds this functionality directly into the signal via an intentional beat component. This enables self-calibration of laser nonlinearity while maintaining a cost-effective hardware configuration. Building on this simplified configuration, the system leverages GPU acceleration with an NVIDIA RTX 4070 Ti to achieve real-time performance, delivering high-throughput signal processing for continuous OFDR interrogation. The signal processing pipeline comprises signal capture, resampling for nonlinearity compensation, and frequency shift computation, all optimized for parallel execution. Hardware benchmarking demonstrates substantial acceleration over CPU implementations, achieving up to a 45× speedup for resampling and frequency shift computations and enabling processing latencies below 30 ms. Thermal response validation is conducted under two complementary scenarios: localized heating using a water bath and cryogenic-temperature conditions using liquid nitrogen. Under localized heating, the system achieves an accuracy of 0.249 °C with a thermal sensitivity of 5.971 GHz/°C, while cryogenic-temperature validation demonstrates a frequency shift response with a sensitivity of 2.383 GHz/°C and an accuracy of 2.04 °C. The high acceleration of the proposed GPU-accelerated OFDR system and its accuracy are achieved by exploiting CUDA-based stride indexing, enabling efficient parallel segmentation and processing of large datasets without additional memory copies. The benchmarking results confirm the robustness, accuracy, and deployability of the proposed OFDR system across a wide temperature range, establishing it as a practical platform for real-time distributed fiber sensing in structurally dynamic environments.

Harb, Salah [Lawrence Berkeley National Laboratory↗
DOE OSTI2024

Direct cross-section measurement of the weak r -process Sr 88 ( α , n ) Zr 91 reaction in ν -driven winds of core-collapse supernovae

About half of the heavy elements beyond iron are known to be produced by the rapid neutron capture process, known as the r process. However, the astrophysical site producing the r process is still uncertain. Chemical abundances observed in several cosmic sites indicate that different mechanisms should be at play. For instance, the abundances around silver measured in a subset of metal-poor stars indicate the presence of a weak r process. This process may be active in neutrino-driven winds of core collapse supernovae where (α,n) reactions dominate the synthesis of Ζ ≈ 40 elements in the expelled materials. Scarcely measured, the rates of (α,n) reactions are determined from statistical Hauser-Feshbach calculations with α-optical-model potentials, which are still poorly constrained. Further, the uncertainties of the (α,n) reaction rates therefore make a significant contribution to the uncertainties of the abundances determined from stellar modeling. In this work, the 88 Sr (α,n)⁢ 91 Zr reaction which impacts the weak r-process abundances has been probed at astrophysics energy for the first time; directly measuring the total cross sections at astrophysical energies of 8.37–13.09 MeV in the center of mass (3.8–7.5 GK). Two measurements were performed at ATLAS with the electrically segmented ionization chamber MUSIC, in inverse kinematics, while following the active target technique. The cross sections of this α-induced reaction on 88 Sr, located at the shell closure N = 50, have been found to be lower than expected, by a factor of 3, despite recent statistical calculations validated by measurements on neighboring nuclei. This result encourages more experimental investigations of (α,n) reactions, at Ν = 50 and towards the neutron-rich side, to further test the predictive power and reliability of such calculations.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗
DOE OSTI2024

Multi deep learning-based stochastic microstructure reconstruction and high-fidelity micromechanics simulation of time-dependent ceramic matrix composite response

A multi deep learning-based framework is developed for efficient, automated microstructure reconstruction and generation of stochastic representative volume elements (SRVEs) with periodic boundary conditions (PBCs) for accurate modeling of ceramic matrix composite (CMC) response. The methodology comprises a convolutional neural network coupled with regression layers to act as a vanilla regression network for semantic segmentation of the microstructure, allowing accurate characterization of the phases and their distributions at the microscale. Scanning electron microscope and confocal microscope are used to obtain C/SiNC and SiC/SiNC CMCs micrographs for vanilla regression testing. Microstructure variability in terms of fiber volume fraction and porosity are quantified through the output regression layer, ensuring accurate representation of material variability in SRVE construction. Generative adversarial network (GAN) and its variants are designed to produce high-fidelity SRVE, spanning CMCs microstructure variability space. A circular padding algorithm is developed to generate SRVEs with PBCs during training of GANs. The accuracy of the generated SRVEs is established through micromechanics simulations, where an efficient formulation of the high-fidelity generalized methods of cells (HFGMC) approach is used to compute the effective mechanical properties. Furthermore, an iterative algorithm is implemented in the HFGMC solver to simulate time-dependent deformation of SiC/SiNC subjected to creep loading conditions.

36 MATERIALS SCIENCE↗
DOE OSTI2026

Upcycling waste PET into functional multiblock copolymers through controlled macromolecular design

Poly(ethylene terephthalate) (PET) oligomers derived from glycolysis depolymerization were converted into multiblock copolymers through diisocyanate-mediated coupling with dihydroxy-terminated oligomers, enabling precise control over copolymer sequence distribution, connectivity, and mechanical performance. Here, we demonstrate that telechelic PET oligomers isolated directly from depolymerized consumer waste can serve as reactive building blocks for the formation of segmented multiblock copolymers, eliminating the need to revert to monomeric feedstocks. Dihydroxy-terminated PET oligomers (Mw ≈ 8 kg mol−1) were coupled with poly(ethylene oxide) (PEO, Mw ≈ 4 kg mol−1) to form PET-PEO multiblock copolymers with high molar mass (Mw ≈ 160 kg mol−1). We further show that the timing of end-capping reactions provides a key control parameter that governs the competition between chain extension and termination, thereby dictating the resulting multiblock architecture and molecular-weight evolution. Evaluation of their mechanical properties reveals that virgin PET exhibits high modulus (∼3 GPa) and strength (43 MPa) but limited ductility (<10% elongation). In contrast, PET–PEO multiblock copolymers retain comparable tensile strength (44 MPa), albeit while exhibiting dramatically enhanced ductility (>90% elongation), forming tougher materials with efficient stress transfer between the rigid PET domains and the flexible PEO segments. When incorporated into PET/PEO blends at low loadings, the multiblock copolymers serve as effective compatibilizers, yielding materials with an intermediate modulus (1.1–1.2 GPa) and improved elongation compared to uncompatibilized blends. Furthermore, the presence of PEO blocks increases water uptake and gas permeability relative to virgin PET, reflecting the tunability of molecular transport through the copolymeric blocks. To our knowledge, this represents the first report of PET–PEO multiblock copolymers derived from post-consumer PET for gas transport applications. These results demonstrate that multiblock copolymer formation from telechelic PET oligomers provides a versatile platform for tailoring the mechanical and transport behavior of polyester-based materials through controlled macromolecular design and establishes a generalizable strategy for transforming consumer plastic waste into functional segmented polymers without requiring complete depolymerization to monomers.

Watson-Sanders, Shelby [Department of Chemistry, U↗
DOE OSTI2025

Optical image analysis of WSe 2 − thresholding for layer detection

The fast and reliable layer identification of two-dimensional transition metal dichalcogenide (TMD), such as WSe 2 , is essential to investigating their thickness-dependent electronic and optical properties. This article presents efficient optical image thresholding methodology designed to segment the mono, bi, and tri-layer regions of WSe 2 flakes mechanically exfoliated onto a SiO 2 /Si substrate. The optical images were first preprocessed to exclude the background effect and analyzed using the pixel medians and interquartile ranges for fundamental color channels—red, green, and blue (RGB). The analysis of red channel pixel intensities yielded three distinct ranges, serving as thresholds for layer segmentation: monolayer (111.0–118.0), bilayer (103.0–110.0), and tri-layer (93.0–103.0). Similarly, thresholds were established for each color channel, facilitating a comparative study of the segmentation performances. Further, the intersection-over-union ($IoU$) calculations revealed that the red and green channels demonstrated greater than 99 % and 90 % accuracy in differentiating each layer, respectively. This approach yields remarkable results without substantial data calibration that utilizes time-intensive heuristic techniques. Moreover, the proposed methodology offers the flexibility to compare performances across different color channels, expanding the applicability for other 2D material systems.

2D Materials↗
DOE OSTI2025

The Work of Mechanical Degradation in Elongating Polymer Melts

Molecular dynamics simulations are used to study the mechanical degradation of well-entangled polymer melts during uniaxial extensional flow. Simulations measure the transient rise in extensional stresses and relate them to the molecular alignment and scission of chain backbones. Intermolecular entanglements couple chain scission in space and time, making degradation sensitive to deformation history and strain rate in ways not displayed by dilute polymer solutions. The rate of chain scission is nonmonotonic and peaks at strains corresponding to the maximum extensibility of entanglement segments but prior to the full extension of chain backbones. We measure a specific work per scission event w* and decompose it into separate contributions associated with chain alignment, chemical bond breaking, and scission-induced plasticity. We find chain scission in melts requires activating plastic dissipation that is multiple orders of magnitude larger than the chemical work required to break a covalent backbone bond. Our findings underscore the critical need to consider bulk polymer mechanics and rheology in designing efficient mechanical degradation and mechanochemical processes.

36 MATERIALS SCIENCE↗
DOE OSTI2025

Optimization of a Mixed Fleet of Aerial Drones for Medical Supplies: A Case Study of Blood Delivery Logistics

Aerial drones have emerged as an innovative solution for faster transportation of time-sensitive items (e.g., emergency medical supplies), potentially reducing the transmission of contagious diseases and enhancing healthcare availability through contactless autonomous delivery. We study fleet sizing and efficient scheduling of a mixed fleet of drones for delivering time-sensitive medical items having distinct release and due times to minimize the required fleet size and fleet composition, the required number of additional batteries, and the total energy consumption. We continuously track the remaining battery energy of drones to determine the optimal timing for battery replacement, rather than replacing the battery at each node. Using actual drone flight test data, we employed a machine learning (ML) method to estimate the energy consumption of different drone types during flight segments for different operating parameters. We present a novel mixed-integer programming model to efficiently formulate the problem that integrates the estimated energy consumption functions from ML. We propose a new greedy heuristic (GH) algorithm and a customized genetic algorithm (GA) for solving large-scale instances of this problem faster. Results demonstrate that the GH algorithm is substantially faster than the accelerated CPLEX and the GA, while sacrificing the solution quality by a small amount. Results based on an actual blood sample delivery case study from Pendleton, Oregon, United States, show that using a mixed fleet of drones reduces the total cost and total energy consumption up to 18.18% and 28.7%, respectively, compared to using a homogeneous fleet.

29 - ENERGY PLANNING, POLICY AND ECONOMY↗
DOE OSTI2024

Unpaired image translation to mitigate domain shift in liquid argon time projection chamber detector responses

Deep learning algorithms often are developed and trained on a training dataset and deployed on test datasets. Any systematic difference between the training and a test dataset may severely degrade the final algorithm performance on the test dataset—what is known as the domain shift problem . This issue is prevalent in many scientific domains where algorithms are trained on simulated data but applied to real-world datasets. Typically, the domain shift problem is solved through various domain adaptation (DA) methods. However, these methods are often tailored for a specific downstream task, such as classification or semantic segmentation, and may not easily generalize to different tasks. This work explores the feasibility of using an alternative way to solve the domain shift problem that is not specific to any downstream algorithm. The proposed approach relies on modern Unpaired Image-to-Image (UI2I) translation techniques, designed to find translations between different image domains in a fully unsupervised fashion. In this study, the approach is applied to a domain shift problem commonly encountered in Liquid Argon Time Projection Chamber (LArTPC) detector research when seeking a way to translate samples between two differently distributed LArTPC detector datasets deterministically. This translation allows for mapping real-world data into the simulated data domain where the downstream algorithms can be run with much less domain-shift-related performance degradation. Conversely, using the translation from the simulated data to a real-world domain can increase the realism of the simulated dataset and reduce the magnitude of any systematic uncertainties. To evaluate the quality of the translations, we use both pixel-wise metrics and a downstream task to measure the effectiveness of UI2I methods for mitigating the domain shift problem. We adapted several popular UI2I translation algorithms to work on scientific data and demonstrated the viability of these techniques for solving the domain shift problem with LArTPC detector data. To facilitate further development of DA techniques for scientific datasets, the ‘Simple Liquid-Argon Track Samples’ dataset used in this study is also published.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2024

Behavioral Segmentation and Clustering of Geospatial Trajectories

The rapid growth of global positioning system (GPS) devices has led to a corresponding increase in the size of GPS datasets. While these large GPS datasets contain a wealth of information about the behaviors of the moving objects in them, manual classification and anomaly detection are prohibitively time consuming. We utilize unsupervised machine learning techniques to first identify the behaviors for individual moving objects and then cluster those objects by their behavioral sequences. In this way, trajectories behaving unusually as well as common patterns of behavior are both detectable in large datasets without requiring an a priori definition of "unusual" or "common."

97 MATHEMATICS AND COMPUTING↗