Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Robust optimization”

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 1,531 records · Page 85
All sourcesNASA NTRSDOE OSTI
DOE OSTI2024

Robustness of topological persistence in knowledge distillation for wearable sensor data

Topological data analysis (TDA) has shown great success in various applications involving wearable sensor data. However, there are difficulties in leveraging topological features in machine learning and wearable sensors because of the large time consumption and computational resources required to extract the features. To address this problem, knowledge distillation (KD) is utilized to generate a small model and accommodate topological features with persistence image (PI) representations from the raw time series data. Deploying topological knowledge in KD enables the student to achieve better performance compared to the one trained solely on raw time series data. However, it is not yet known if there are coherent characteristics for topological features in PI, which can aid in improving the performance during KD. In this paper, we investigate the suitability and challenges of utilizing topological features in KD for wearable sensor data, thereby contributing to the advancement of the field. Our study explores the impact of transferred topological features by comparing the Teacher-to-Student framework with Multiple Teachers-to-Student where teachers utilize both time series data and persistence images obtained by TDA as inputs. Additionally, we conduct a rigorous examination of topological knowledge effects by testing under various corruptions, knowledge types, and learning strategies in the context of human activity recognition tasks. Our analysis of topological features in KD presents the optimal strategy for incorporating these features. This study includes datasets of varying scales, window lengths, and activity classes, providing a comprehensive evaluation. Our results demonstrate that leveraging topological features in KD to enhance performance across databases.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2020

Selection of optimal wavelengths for optical soiling modelling and detection in photovoltaic modules

Soiling impacts the photovoltaic (PV) module performance by decreasing the amount of light reaching the photovoltaic cells and by changing their external spectral response. Currently, the soiling monitoring market is moving toward optical sensors that measure transmittance or reflectance, rather than directly measuring the impact of soiling on the performance of photovoltaic modules. These sensors, which use a single optical measurement, are not able to correct the soiling losses that depend on the solar irradiance spectra and on the spectral response of the monitored PV material. This work investigates methods that can improve the optical detection of soiling by extracting the full soiling spectrum profiles using only two or three monochromatic measurements. Spectral transmittance data, measured with a spectrophotometer and collected during a 46-week experimental soiling study carried out in Jaén, Spain, was analysed in this work. The use of a spectral profile for the hemispherical transmittance of soiled PV glass is found to significantly improve the soiling detection, returning the lowest errors independently of the PV materials and irradiance conditions. In addition, this work shows that it is also possible to select the measurement wavelengths to minimize the soiling loss detection error depending on the monitored PV semiconductor material (silicon, CdTe, a-Si, CIGS and a representative perovskite). The approaches discussed in this work are also found to be more robust to potential measurement errors compared to single wavelength measurement techniques.

14 SOLAR ENERGY↗
DOE OSTI2025

Integrating Safety, Security, and Nuclear Operations for Advanced Reactors

The traditional separation between safety, security, and operations teams has created significant barriers to achieving optimal outcomes. When security considerations are introduced late in the design process, they often conflict with already-established architectural, operational, or engineering parameters. Retrofitting security measures can lead to increased costs, schedule delays, and compromises in security effectiveness. For instance, the need to retrofit physical barriers or surveillance systems often results in trade-offs that could have been avoided with earlier input from security professionals. Delayed integration can also affect regulatory processes and result in licensing delays. Security reviews conducted at later stages frequently identify gaps that necessitate significant redesign efforts, impacting not only scope, schedule, and budget, but also adding risk and lowering stakeholder confidence in the project. This paper aims to address these challenges by identifying practical opportunities for integrating security considerations seamlessly with design and operations teams throughout the entire lifecycle of nuclear facilities—from conceptual design to commissioning and beyond. The research emphasizes the value of early and continuous collaboration among stakeholders to ensure that security measures are robust, operationally effective, and cost-efficient. By examining case studies, analyzing past incidents, and leveraging best practices from other high-security industries, this study highlights actionable strategies for bridging the gap between safety, security, and operations teams.

Zineddin, Dr. Z. [ORNL] (ORCID:0009000848740725)↗
DOE OSTI2023

Lithium-Ion Battery Diagnostics Using Electrochemical Impedance via Machine-Learning

Diagnosing battery states such as health, state-of-charge, or temperature is crucial for ensuring the safety and reliability of electrochemical energy storage systems. While some states, such as temperature, may be measured using cheap sensors, accurate diagnosis of battery health metrics usually requires time-consuming performance measurements, making them infeasible for use in real-world operation. These health metrics can be measured during lab-testing and then estimated on-line using predictive life models or via state observer algorithms such as Kalman filters, but these predictive methods should be supplemented by actual measurement of battery health whenever possible to ensure reliability. Rapid measurement of battery health may be done by various types of fast diagnostic techniques such as electrochemical impedance spectroscopy (EIS), which can be performed in only a few minutes and require only a fraction of the energy and power needed for a full charge and discharge measurement. But there is a substantial challenge for estimating battery health using EIS data, as EIS is sensitive to cell temperature, state-of-charge, current, and resting time in addition to health. Thus, utilizing EIS data to predict battery capacity requires correcting for all these additional variables, a task that is extremely difficult to handle analytically. This talk utilizes machine-learning methods to estimate the effectiveness of battery capacity prediction from EIS data, leveraging a data set of hundreds of EIS measurements recorded at varying temperature and state-of-charge throughout a 500-day aging study of 32 commercial, large-format NMC-Graphite lithium-ion batteries. Using EIS as input to machine-learning models is complicated by the nonlinear response of impedance to battery health, temperature, and state-of-charge, as well as the collinearity between the impedance response at neighboring frequencies, which can easily lead to overfit models. To train robust models, features from EIS data need to be extracted from the data or some subset of critical frequencies selected. Many approaches for extracting and selecting features from EIS data from electrochemical analysis and machine-learning fields were identified for analysis: using the entire raw spectra; selection of one, two, or many frequencies from the entire spectra; selecting interesting points from the EIS measurement using domain knowledge; fitting EIS with an equivalent-circuit model; calculating statistics on the raw impedance values; and reducing the dimensionality of the data using unsupervised linear (principal component analysis) and non-linear (uniform manifold approximation and projection) methods. These approaches were rigorously compared using a machine-learning pipeline approach, training linear, Gaussian process, and random forest regression models and quantifying performance using cross-validation as well as a held-out test set. An artificial neural network model trained on the raw spectra was also tested. Promising pipelines were fine-tuned via Bayesian hyperparameter optimization using cross-validation loss and training with class-specific weights to counter data set imbalance. The most reliable method for utilizing impedance in this work was the selection of two optimal frequencies through an exhaustive search, resulting in about 2% mean absolute error on test data for both Gaussian process and random forest model architectures. Interrogation of a variety of models reveals critical frequencies of 100 Hz and 103 Hz for this data set, though the optimal set of frequencies is not necessarily intuitive, i.e., the best performing models are not simply those that use impedance at frequencies that have the highest correlation to the relative discharge capacity. The best performing model is an ensemble model, which is able to predict battery capacity with 1.9% mean absolute error for unseen cells using impedance recorded at a variety of temperatures and states-of-charge.

battery↗
DOE OSTI2023

Synergistic Co-Ir/Ru Composite Electrocatalysts Impart Efficient and Durable Oxygen Evolution Catalysis in Acid

Exploring highly active and robust catalysts, which have low precious metal content, to boost the kinetically sluggish oxygen evolution reaction (OER) is a key concern for hydrogen production via proton exchange membrane water electrolysis (PEMWE). Here, in this work, rational engineering of the morphology and the local geometric ligand environment of Ir and Ru catalysts are presented by using defect-rich, lanthanum- and lithium-doped Co 3 O 4 nanofiber (LLCF) as substrate that promotes the electrocatalytic OER. Two catalysts, IrCoOx@LLCF and RuCoOx@LLCF, achieve mass activities of 1013.5 A g Ir –1 and 1911.4 A g Ru –1 in 0.1 M HClO 4 at 300 mV overpotential, respectively, which are 26 and 50 times higher than that of commercial IrO 2 and RuO 2 . Operando X-ray absorption spectroscopy unveils the reversible structure of IrCoOx during the OER and the suppression of over-oxidation of Co and Ir, giving rise to high stability. Density functional theory calculations reveal that the local geometric ligand engineering optimizes the binding of oxygenated species to the active sites, resulting in strongly enhanced catalytic activity.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2022

GIF Supercritical Water Cooled Reactor: Proliferation Resistance and Physical Protection White Paper

This report is part of a series of six white papers, prepared jointly by the Proliferation Resistance and Physical Protection Working Group (PRPPWG) and the six System Steering Committees (SSCs) and provisional System Steering Committees (pSSCs). This publication is an update to a similar series published in 2011 presenting the status of Proliferation Resistance & Physical Protection (PR&PP) characteristics for each of the six systems selected by the Generation IV International Forum (GIF) for further research and development, namely: the Sodium-cooled fast Reactor (SFR), the Very high temperature reactor (VHTR), the gas-cooled fast reactor (GFR), the Molten salt reactor (MSR) and the Supercritical water–cooled reactor (SCWR). This white paper represents the status of Proliferation Resistance and Physical Protection (PR&PP) characteristics for the Supercritical Water-cooled Fast reactor (SCFR) reference designs selected by the Generation IV International Forum (GIF) SCWR System Steering Committee (SSC). The intent is to generate preliminary information about the PR&PP features of the SCWR reactor technology and to provide insights for optimizing their PR&PP performance for the benefit of SCWR system designers. It updates the SCWR analysis published in the 2011 report “Proliferation Resistance and Physical Protection of the Six Generation IV Nuclear Energy Systems”, prepared Jointly by the Proliferation Resistance and Physical Protection Working Group (PRPPWG) and the System Steering Committees and provisional System Steering Committees of the Generation IV International Forum, taking into account the evolution of both the systems, the GIF R&D activities, and an increased understanding of the PR&PP features. The white paper, prepared jointly by the GIF PRPPWG and the GIF SCWR SSC, follows the highlevel paradigm of the GIF PR&PP Evaluation Methodology to investigate the PR&PP features of the eight proposed GIF SCWR designs. Two small modular reactors, the Canadian SSR and the Canada/China/Europe ECC-SMART are also mentioned. An overview of the fuel cycles for the GIF designs are provided. For PR, the document analyses and discusses the proliferation resistance aspects in terms of robustness against State-based threats associated with diversion of materials, misuse of facilities, breakout scenarios, and production in clandestine facilities. Similarly, for PP, the document discusses the robustness against theft of material and sabotage by non-State actors. The document follows a common template adopted by all the white papers in the updated series.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗
DOE OSTI2024

1. Physical Security Engineering by Design for Nuclear Facilities; 2. Nuclear Power Plant Site Security Management – A Security Strategy; 3. The UAE Women in Nuclear Energy Security

1. Security by design, or SeBD, is a comprehensive approach that integrates the physical protection system of a nuclear plant into every stage of its existence. This includes planning, designing, constructing, commissioning, and operating the facility, using a combination of analytical, physical, technological, and procedural measures. Essentially, SeBD involves intentionally applying and incorporating security into all aspects of design and operation throughout the entire lifecycle of a facility. By implementing this methodology throughout various phases such as program development, process implementation, staff training and procedures management in conjunction with plant equipment, facilities can be optimized to minimize security risks without compromising functional design requirements. This ultimately improves the overall security posture of the site and reduces the need for costly modifications or additional security resources post-design. 2. A site security strategy is a living document that is revised on a periodic or event-driven basis, ensuring that site security operations and corresponding procedures provide long-term, effective protection for the entire nuclear power plant (NPP) site. A site security strategy aims to mitigate threats across the entire NPP site via in-depth defense approaches and mutual support; therefore, if a layer is omitted or altered, then the effect across all layers must be re-evaluated. Therefore, the aim of the site security strategy is to provide an appropriate, scalable security regime that deters, denies, delays, and detects incidents and, equally importantly, reassures legitimate users and the regulator that due diligence and regulatory compliance have been achieved, ensuring that the site is safe and secure. Robust access control for vehicles and pedestrians is at the heart of the strategy. Vehicle and pedestrian searching and screening are seen as the strongest mitigation methods against vehicle- and pedestrian-borne attacks. The security strategy must also be supported through comprehensive staff training and the development of robust processes, procedures, and planning. If all these measures are to be effective, then training must be implemented during each phase of construction, partial operation/commissioning, and full operation. No single element of site security is completely isolated from the influence of other elements. Ideally, consideration of all key elements will result in a security strategy that is integrated and proportional to the threat and that does not over specify individual security solutions through the application of isolated measures but rather applies a holistic, all-encompassing approach. 3. When women enter the labor force, numerous positive outcomes emerge, including increased GDP, educational gains, and decreased maternal mortality. Despite these benefits, women's employment rates and equality vary significantly worldwide as does support for women in the workforce. This paper will explore the multifaceted benefits of women's employment, the factors influencing labor force participation rates, and the urgency to achieve gender equality as outlined in the 2015 United Nations Sustainable Development Goals (SDGs). It will then examine the emerging presence of women in the traditionally male-dominated nuclear field, specifically within the United Arab Emirates (UAE) as a testament to their resilience and determination to break social norms and advance gender equality.

Zineddin, Dr. Z.↗
DOE OSTI2023

Data-Driven Affinely Adjustable Robust Volt/VAr Control

Recent years have seen the increasing proliferation of distributed energy resources with intermittent power outputs, posing new challenges to the voltage management in distribution networks. To this end, this paper proposes a data-driven affinely adjustable robust Volt/VAr control (AARVVC) scheme, which modulates the smart inverter’s reactive power in an affine function of its active power, based on the voltage sensitivities with respect to real/reactive power injections. To achieve a fast and accurate estimation of voltage sensitivities, we propose a data-driven method based on deep neural network (DNN), together with a rule-based bus-selection process using the bidirectional search method. Our method only uses the operating statuses of selected buses as inputs to DNN, thus significantly improving the training efficiency and reducing information redundancy. Finally, a distributed consensus-based solution, based on the alternating direction method of multipliers (ADMM), for the AARVVC is applied to decide the inverter’s reactive power adjustment rule with respect to its active power. Only limited information exchange is required between each local agent and the central agent to obtain the slope of the reactive power adjustment rule, and there is no need for the central agent to solve any (sub)optimization problems. Finally, numerical results on the modified IEEE-123 bus system validate the effectiveness and superiority of the proposed data-driven AARVVC method.

24 POWER TRANSMISSION AND DISTRIBUTION↗
DOE OSTI2024

PIP-II Linac Cryogenic Distribution System Design Challenges

The PIP-II linac Cryogenic Distribution System (CDS) is characterized by extremely small heat inflows and robust mechanical design. It consists of a Distribution Valve Box (DVB), Intermediate Transfer Line, Tunnel Transfer Line, comprising 25 Bayonet Cans, and ends with a Turnaround Can. Multiple helium streams, each characterized by distinct helium parameters, flow through each of these elements. The CDS geometry allows maintaining an acceptable pressure drop for each helium stream, considering the planned flows and helium parameters in different operation modes. This is particularly crucial for the return line of helium vapors, which return from cryomodules to the cold compressors and thus have very restrictive pressure drop requirements. On both sides of the DVB there are fixed supports for process pipes. One of the DVB design challenges was to route the process pipes in such a way that their shape provided sufficient compensation for thermal shrinkage. This ensures th at the forces resulting from thermal shrinkage acting on the cryogenic valves remain at a level acceptable to the manufacturer. The required thermal budget of the CDS was achieved by thermo-mechanical optimization of its components, like process pipes fixed supports in Bayonet Cans.

43 PARTICLE ACCELERATORS↗
DOE OSTI2023

Reinforcement Learning Approach to Cybersecurity in Space (RELACSS)

Securing satellite groundstations against cyber-attacks is vital to national security missions. However, these cyber threats are constantly evolving. As vulnerabilities are discovered and patched, new vulnerabilities are discovered and exploited. In order to automate the process of discovering existing vulnerabilities and the means to exploit them, a reinforcement learning framework is presented in this report. We demonstrate that this framework can learn to successfully navigate an unknown network and detect nodes of interest despite the presence of a moving target defense. The agent then exfiltrates a file of interest from the node as quickly as possible. This framework also incorporates a defensive software agent that learns to impede the attacking agents progress. This setup allows for the agents to work against each other and improve their abilities. We anticipate that this capability will help uncover unforeseen vulnerabilities and the means to mitigate them. The modular nature of the framework enables users to swap out learning algorithms and modify the reward functions in order to adapt the learning tasks to various use cases and environments. Several algorithms, viz., tabular Q learning, deep Q networks, proximal policy optimization, advantage actor-critic, generative adversarial imitation learning, are explored for the agents and the results highlighted. The agent learns to solve the tasks in a light-weight abstract environment. Once the agent learns to perform sufficiently well, it can be deployed in a minimega virtual machine environment (or a real network) with wrappers that map abstract actions to software commands. The agent also uses a local representation of the actions called a ‘slot-mechanism’. This allows the agent to learn in a certain network and generalize it to different networks. The defensive agent learns to predict the actions taken by an offensive agent and uses that information to anticipate the threat. This information can then either be used to raise an alarm or to take actions to thwart the attack. We believe that with the appropriate reward design, a representative environment, and action set, this framework can be generalized to tackle other cybersecurity tasks. By sufficiently training these agents, we can anticipate vulnerabilities leading to robust future designs. We can also deploy automated defensive agents that can help secure satellite groundstation and their vital national security missions.

45 MILITARY TECHNOLOGY, WEAPONRY, AND NATIONAL DEF↗
DOE OSTI2025

Heterovalent Substitution of K 2 SrP 2 O 7 :Cr 3+ to Achieve Anti-Thermal-Quenching Broadband Near-Infrared Luminescence

Broadband near-infrared (NIR) light sources based on phosphor-converted light-emitting diodes are highly desirable for biochemical analysis and medical diagnosis applications. However, thermal quenching remains a demanding challenge for developing efficient NIR phosphors. Herein, we report the enhancement of both quantum efficiency and thermal stability in Cr 3+ -activated K 2 SrP 2 O 7 phosphors through a heterovalent substitution strategy by replacing Sr 2+ with Al 3+ in K 2 Sr 1–x Al x P 2 O 7 (0.05 ≤ x ≤ 0.2) to obtain optimized broadband NIR emission. Structural modulation via Al 3+ substitution leads to the optimized composition, K 2 Sr 0.88 Al 0.1 P 2 O 7 :0.02Cr 3+ , which emits across a broad NIR range of 650–1100 nm peaking at 807 nm with a full width at half-maximum of ∼130 nm under 448 nm excitation. Remarkably, its emission intensity at 150 °C remains 120% of the initial value at room temperature, demonstrating a rare antithermal-quenching behavior. Temperature-dependent XRD studies further reveal that Al 3+ substitution effectively suppresses lattice expansion at elevated temperatures, indicating enhanced lattice stability under thermal excitation. Detailed structural and spectral analyses show that the substitution enhances local site symmetry, reduces electron–phonon coupling, increases thermally induced absorption probability, and fortifies energetic barriers against nonradiative transitions. These synergistic effects collectively endow this NIR phosphor with a superior thermal stability. Furthermore, NIR light-emitting diodes fabricated with this phosphor exhibit strong potential for applications in information identification, nondestructive detection, and night vision technologies. This study demonstrates a local structure engineering strategy for designing thermally robust Cr 3+ -activated NIR phosphors, offering valuable insights into material discovery and NIR spectroscopy device development.

Cr3+↗
DOE OSTI2020

Cognition at the Point of Sensing

Over the last 15 years, compressive sensing techniques have been developed which have the potential to greatly reduce the amount of data collected by systems while preserving the amount of information obtained. A cost of this efficiency is that a computationally-intensive optimization routine must be used to put the sensed data into a form that a person can interpret. At the same time, machine learning techniques have experienced tremendous growth as well. Machines have demonstrated the ability learn how to effectively perform tasks such as detection and classification at speeds much faster than humanly possible. Our goal in this project was to study the feasibility of using compressive sensing systems "at the edge." That is, how can compressive sensing sensors be deployed such that information is created at the remote sensor rather than sending raw data to a central processing location? Studies were performed to analyze whether machine learning could be done on the compressively sensed data in its raw form. If a machine is performing the task, is it possible to do so without putting the data into a human interpretable form? We show that this is possible for some systems, in particular a compressive sensing snapshot imaging spectrometer. Machine learning tasks were demonstrated to be more effective and more robust to noise when the machine learning algorithm worked on data in its raw form. This system is shown to outperform a traditional spectrometer. Techniques for reducing the complexity of the reconstruction routine were also analyzed. Techniques for such as data regularization, deep neural networks, and matrix completion were studied and shown to have benefits over traditional reconstruction techniques. In this project we showed that compressive sensing sensors are indeed feasible at the edge. As always, sensors and algorithms must be carefully tuned to work in the constrained environment. In this project we developed tools and techniques to enable those analyses.

47 OTHER INSTRUMENTATION↗
DOE OSTI2022

Combined Imaging and RNA-Seq on a Microfluidic Platform for Viral Infection Studies

The goal of this work was to pioneer a novel, low-overhead protocol for simultaneously assaying cell-surface markers and intracellular gene expression in a single mammalian cell. The purpose of developing such a method is to be able to understand the mechanisms by which pathogens engage with individual mammalian cells, depending on their cell surface proteins, and how both host and pathogen gene expression changes are reflective of these mechanisms. The knowledge gained from such analyses of single cells will ultimately lead to more robust pathogen detection and countermeasures. Our method was aimed at streamlining both the upstream cell sample preparation using microfluidic methods, as well as the actual library making protocol. Specifically, we wanted to implement a random hexamer-based reverse transcription of all RNA within a single cell (as opposed to oligo dT-based which would only capture polyadenylated transcripts), and then use a CRISPR-based method called scDash to deplete ribosomal DNAs (since ribosomal RNAs make up the majority of the RNA in a mammalian cell). After significant troubleshooting, we demonstrate that we are able to prepare cDNA from RNA using the random hexamer primer, and perform the rDNA depletion. We also show that we can visualize individually stained cells, setting up the pipeline for connecting surface markers to RNA-sequencing profiles. Finally, we test a number of devices for various parts of the pipeline, including bead generation, optical barcoding and cell dispensing, and demonstrate that while some of these have potential, more work is needed to optimize this part of the pipeline.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2022

GIF Gas Cooled Fast Reactor: Proliferation Resistance and Physical Protection White Paper

This white paper represents the status of Proliferation Resistance and Physical Protection (PR&PP) characteristics for the Gas-cooled Fast reactor (GFR) reference designs selected by the Generation IV International Forum (GIF) GFR System Steering Committee (SSC). The intent is to generate preliminary information about the PR&PP features of the GFR reactor technology and to provide insights for optimizing their PR&PP performance for the benefit of GFR system designers. It updates the GFR analysis published in the 2011 report “Proliferation Resistance and Physical Protection of the Six Generation IV Nuclear Energy Systems”, prepared Jointly by the Proliferation Resistance and Physical Protection Working Group (PRPPWG) and the System Steering Committees and provisional System Steering Committees of the Generation IV International Forum, taking into account the evolution of both the systems, the GIF R&D activities, and an increased understanding of the PR&PP features. The white paper, prepared jointly by the GIF PRPPWG and the GIF GFR SSC, follows the high-level paradigm of the GIF PR&PP Evaluation Methodology to investigate the PR&PP features of the GIF GFR 2400 MWth reference design. The ALLEGRO reactor is also described. The EM2 and HEN MHR reactor are mentioned. An overview of fuel cycle for the GFR reference design and for the ALLEGRO reactor are provided. For PR, the document analyses and discusses the proliferation resistance aspects in terms of robustness against State-based threats associated with diversion of materials, misuse of facilities, breakout scenarios, and production in clandestine facilities. Similarly, for PP, the document discusses the robustness against theft of material and sabotage by non-State actors. The document follows a common template adopted by all the white papers in the updated series.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗
DOE OSTI2025

Accurate and uncertainty-aware multi-task prediction of HEA properties using prior-guided deep Gaussian processes

Surrogate modeling techniques have become indispensable in accelerating the discovery and optimization of high-entropy alloys (HEAs), especially when integrating computational predictions with sparse experimental observations. This study systematically evaluates the training and testing performance of four prominent surrogate models—conventional Gaussian processes (cGP), Deep Gaussian processes (DGP), encoder-decoder neural networks for multi-output regression and eXtreme Gradient Boosting (XGBoost)—applied to a hybrid dataset of experimental and computational properties of the 8-component HEA system Al-Co-Cr-Cu-Fe-Mn-Ni-V. We specifically assess their capabilities in predicting correlated material properties, including yield strength, hardness, modulus, ultimate tensile strength, elongation, and average hardness under dynamic/quasi-static conditions, alongside auxiliary computational properties. The comparison highlights the strengths of hierarchical deep modeling approaches in handling heteroscedastic, heterotopic, and incomplete data commonly encountered in materials science. Our findings illustrate that combined surrogate models such as DGPs infused with machine-learned priors outperform other surrogates by effectively capturing inter-property correlations and by assimilating prior knowledge. This enhanced predictive accuracy positions the combined surrogate models as powerful tools for robust and data-efficient materials design.

36 MATERIALS SCIENCE↗
DOE OSTI2024

Catabolism of β-5 linked aromatics by Novosphingobium aromaticivorans

ABSTRACT Aromatic compounds are an important source of commodity chemicals traditionally produced from fossil fuels. Aromatics derived from plant lignin can potentially be converted into commodity chemicals through depolymerization followed by microbial funneling of monomers and low molecular weight oligomers. This study investigates the catabolism of the β-5 linked aromatic dimer dehydrodiconiferyl alcohol (DC-A) by the bacterium Novosphingobium aromaticivorans . We used genome-wide screens to identify candidate genes involved in DC-A catabolism. Subsequent in vivo and in vitro analyses of these candidate genes elucidated a catabolic pathway composed of four required gene products and several partially redundant dehydrogenases that convert DC-A to aromatic monomers that can be funneled into the central aromatic metabolic pathway of N. aromaticivorans . Specifically, a newly identified γ-formaldehyde lyase, PcfL, opens the phenylcoumaran ring to form a stilbene and formaldehyde. A lignostilbene dioxygenase, LsdD, then cleaves the stilbene to generate the aromatic monomers vanillin and 5-formylferulate (5-FF). We also showed that the aldehyde dehydrogenase FerD oxidizes 5-FF before it is decarboxylated by LigW, yielding ferulic acid. We found that some enzymes involved in the β-5 catabolism pathway can act on multiple substrates and that some steps in the pathway can be mediated by multiple enzymes, providing new insights into the robust flexibility of aromatic catabolism in N. aromaticivorans . A comparative genomic analysis predicted that the newly discovered β-5 aromatic catabolic pathway is common within the order Sphingomonadales. IMPORTANCE In the transition to a circular bioeconomy, the plant polymer lignin holds promise as a renewable source of industrially important aromatic chemicals. However, since lignin contains aromatic subunits joined by various chemical linkages, producing single chemical products from this polymer can be challenging. One strategy to overcome this challenge is using microbes to funnel a mixture of lignin-derived aromatics into target chemical products. This approach requires strategies to cleave the major inter-unit linkages of lignin to release monomers for funneling into valuable products. In this study, we report newly discovered aspects of a pathway by which the Novosphingobium aromaticivorans DSM12444 catabolizes aromatics joined by the second most common inter-unit linkage in lignin, the β-5 linkage. This work advances our knowledge of aromatic catabolic pathways, laying the groundwork for future metabolic engineering of this and other microbes for optimized conversion of lignin into products.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2023

Hierarchical Modeling to Enhance Spectrophotometry Measurements—Overcoming Dynamic Range Limitations for Remote Monitoring of Neptunium

A robust hierarchical model has been demonstrated for monitoring a wide range of neptunium concentrations (0.75–890 mM) and varying temperatures (10–80 °C) using chemometrics and feature selection. The visible–near infrared electronic absorption spectrum (400–1700 nm) of monocharged neptunyl dioxocation (Np(V) = NpO2+) includes many bands, which have molar absorption coefficients that differ by nearly 2 orders of magnitude. The shape, position, and intensity of these bands differ with chemical interactions and changing temperature. These challenges make traditional quantification by univariate methods unfeasible. Measuring Np(V) concentration over several orders of magnitude would typically necessitate cells with varying path length, optical switches, and/or multiple spectrophotometers. Alternatively, the differences in the molar extinction coefficients for multiple absorption bands can be used to quantify Np(V) concentration over 3 orders of magnitude with a single optical path length (1 mm) and a hierarchical multivariate model. In this work, principal component analysis was used to distinguish the concentration regime of the sample, directing it to the relevant partial least squares regression submodels. Each submodel was optimized with unique feature selection filters that were selected by a genetic algorithm to enhance predictions. Through this approach, the percent root mean square error of prediction values were ≤1.05% for Np(V) concentrations and ≤4% for temperatures. This approach may be applied to other nuclear fuel cycle and environmental applications requiring real-time spectroscopic measurements over a wide range of conditions.

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

Towards Precision Photometric Type ia Supernova Cosmology With Machine Learning

The revolutionary discovery of dark energy and accelerating cosmic expansion was made with just 42 type Ia supernovae (SNe Ia) in 1999. Since then, large synoptic surveys, e.g., Dark Energy Survey (DES), have observed thousands more SNe Ia and the upcoming Rubin Legacy Survey of Space and Time (LSST) and Roman Space Telescope promise to deliver millions in the next decade. This unprecedented data volume can produce the required precision to unambiguously test concordance cosmology which could represent a monumental shift in our understanding of dark energy and its role in cosmic history. However, extracting a pure SN Ia sample with accurate redshifts for such a large dataset will be a challenge. Specifically, spectroscopic classification will not be possible for the vast majority of discovered objects, and only ~25% will have spectroscopic redshifts. This thesis presents a series of observational and methodological studies designed to address the questions associated with this new era of photometric SN Ia cosmology. First, we present a machine learning method for photometric classification of SNe, Supernova Classification with a COnvolutional Neural Network (SCONE). Photometric classification enables SNe with no spectroscopic information to be confidently categorized, and is a critical component of current and future analysis pipelines. SCONE achieves >99% accuracy distinguishing simulated SNe Ia from non-Ia SNe, and has been integrated into DES, LSST, and Roman analysis pipelines. We also demonstrate the efficacy of SCONE on early-time photometric classification, which will be vital for optimal allocation of spectroscopic resources. We show that SCONE can distinguish between 6 SN types with 75% accuracy on the night of initial discovery, comparable to results in the literature for full-phase SNe. Next, we study current methods for estimating SN Ia redshifts and propose a machine learning alternative that uses SN photometry alone to extract redshift information. Most SNe Ia inherit redshift information from their host galaxy, but the process of matching SNe to the correct host galaxy can be challenging. We systematically analyze the impact of incorrect redshifts from host galaxy mismatch on 5 years of DES SN data, and conclude that improved host matching or redshift estimation methods can reduce our systematic errors by ~10%. In response to this finding, we present a SN photometry-only method for estimating redshifts independent of host galaxy information, Photo-zSNthesis. We show that Photo-zSNthesis redshift estimates are accurate to within 2% across the full redshift range of LSST, a first in the literature. Finally, we focus on the robustness of machine learning (ML) algorithms for real-world and scientific applications. ML models generalize poorly beyond their training set and often experience severe performance degradation when deployed on new data. We demonstrate a general method for improving robustness that achieves new state-of-the-art results on astronomical object classification, wildlife identification, and tumor detection.

79 ASTRONOMY AND ASTROPHYSICS↗