Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “probabilistic modeling”

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

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

At least 379 records · Page 21

Physics makes the difference: Bayesian optimization and active learning via augmented Gaussian process

Abstract Both experimental and computational methods for the exploration of structure, functionality, and properties of materials often necessitate the search across broad parameter spaces to discover optimal experimental conditions and regions of interest in the image space or parameter space of computational models. The direct grid search of the parameter space tends to be extremely time-consuming, leading to the development of strategies balancing exploration of unknown parameter spaces and exploitation towards required performance metrics. However, classical Bayesian optimization (BO) strategies based on the Gaussian process (GP) do not readily allow for the incorporation of the known physical behaviors or past knowledge. Here we explore a hybrid optimization/exploration algorithm created by augmenting the standard GP with a structured probabilistic model of the expected system’s behavior. This approach balances the flexibility of the non-parametric GP approach with a rigid structure of physical knowledge encoded into the parametric model. The fully Bayesian treatment of the latter allows additional control over the optimization via the selection of priors for the model parameters. The method is demonstrated for a noisy version of a standard univariate test function used to evaluate optimization algorithms and further extended to physical lattice models. This methodology is expected to be universally suitable for injecting prior knowledge in the form of physical models and past data in the BO framework.

42 ENGINEERING↗

Stochastic Model Predictive Control for Demand Response in a Home Energy Management System

This paper presents a chance constrained, model predictive control (MPC) algorithm for demand response (DR) in a home energy management system (HEMS). The HEMS optimally schedules controllable appliances given user preferences such as thermal comfort and energy cost sensitivity, and available residentially-owned power sources such as photovoltaic (PV) generation and home battery systems. The proposed control architecture ensures both the DR event and indoor thermal comfort are satisfied with a high probability given the uncertainty in available PV generation and the outdoor temperature forecast. The uncertainties are incorporated into the MPC formulation using probabilistic constraints instead of computationally limiting sampling-based approaches. Simulation results for various user preferences and probabilistic model parameters show the effectiveness of the HEMS algorithm response to DR requests.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

Stochastic Model Predictive Control for Demand Response in a Home Energy Management System: Preprint

This paper presents a chance constrained, model predictive control (MPC) algorithm for demand response (DR) in a home energy management system (HEMS). The HEMS optimally schedules controllable appliances given user preferences such as thermal comfort and energy cost sensitivity, and available residentially-owned power sources such as photovoltaic (PV) generation and home battery systems. The proposed control architecture ensures both the DR event and indoor thermal comfort are satisfied with a high probability given the uncertainty in available PV generation and the outdoor temperature forecast. The uncertainties are incorporated into the MPC formulation using probabilistic constraints instead of computationally limiting sampling-based approaches. Simulation results for various user preferences and probabilistic model parameters show the effectiveness of the HEMS algorithm response to DR requests.

47 OTHER INSTRUMENTATION↗

How New Mexico Leveraged a COVID-19 Case Forecasting Model to Preemptively Address the Healthcare Needs of the State: Quantitative Analysis

Prior to the COVID-19 pandemic, US hospitals relied on static projections of future trends for long-term planning and were only beginning to consider forecasting methods for short-term planning of staffing and other resources. With the overwhelming burden imposed by COVID-19 on the health care system, an emergent need exists to accurately forecast hospitalization needs within an actionable timeframe. Our goal was to leverage an existing COVID-19 case and death forecasting tool to generate the expected number of concurrent hospitalizations, occupied intensive care unit (ICU) beds, and in-use ventilators 1 day to 4 weeks in the future for New Mexico and each of its five health regions. We developed a probabilistic model that took as input the number of new COVID-19 cases for New Mexico from Los Alamos National Laboratory’s COVID-19 Forecasts Using Fast Evaluations and Estimation tool, and we used the model to estimate the number of new daily hospital admissions 4 weeks into the future based on current statewide hospitalization rates. The model estimated the number of new admissions that would require an ICU bed or use of a ventilator and then projected the individual lengths of hospital stays based on the resource need. By tracking the lengths of stay through time, we captured the projected simultaneous need for inpatient beds, ICU beds, and ventilators. We used a postprocessing method to adjust the forecasts based on the differences between prior forecasts and the subsequent observed data. Thus, we ensured that our forecasts could reflect a dynamically changing situation on the ground. Forecasts made between September 1 and December 9, 2020, showed variable accuracy across time, health care resource needs, and forecast horizon. Forecasts made in October, when new COVID-19 cases were steadily increasing, had an average accuracy error of 20.0%, while the error in forecasts made in September, a month with low COVID-19 activity, was 39.7%. Across health care use categories, state-level forecasts were more accurate than those at the regional level. Although the accuracy declined as the forecast was projected further into the future, the stated uncertainty of the prediction improved. Forecasts were within 5% of their stated uncertainty at the 50% and 90% prediction intervals at the 3- to 4-week forecast horizon for state-level inpatient and ICU needs. However, uncertainty intervals were too narrow for forecasts of state-level ventilator need and all regional health care resource needs. Real-time forecasting of the burden imposed by a spreading infectious disease is a crucial component of decision support during a public health emergency. Our proposed methodology demonstrated utility in providing near-term forecasts, particularly at the state level. This tool can aid other stakeholders as they face COVID-19 population impacts now and in the future.

60 APPLIED LIFE SCIENCES↗

THE MASS DISTRIBUTION OF COMPANIONS TO LOW-MASS WHITE DWARFS

Measuring the masses of companions to single-line spectroscopic binary stars is (in general) not possible because of the unknown orbital plane inclination. Even when the mass of the visible star can be measured, only a lower limit can be placed on the mass of the unseen companion. However, since these inclination angles should be isotropically distributed, for a large enough, unbiased sample, the companion mass distribution can be deconvolved from the distribution of observables. In this work, we construct a hierarchical probabilistic model to infer properties of unseen companion stars given observations of the orbital period and projected radial velocity of the primary star. We apply this model to three mock samples of low-mass white dwarfs (LMWDs; M ≲ 0.45 M {sub ☉}) and a sample of post-common-envelope binaries. We use a mixture of two Gaussians to model the WD and neutron star (NS) companion mass distributions. Our model successfully recovers the initial parameters of these test data sets. We then apply our model to 55 WDs in the extremely low-mass (ELM) WD Survey. Our maximum a posteriori model for the WD companion population has a mean mass μ{sub WD} = 0.74 M {sub ☉}, with a standard deviation σ{sub WD} = 0.24 M {sub ☉}. Our model constrains the NS companion fraction f {sub NS} to be <16% at 68% confidence. We make samples from the posterior distribution publicly available so that future observational efforts may compute the NS probability for newly discovered LMWDs.

79 ASTRONOMY AND ASTROPHYSICS↗

Engineering and Application of a Thermostable MHETase for PET Depolymerization

Enzymatic hydrolysis of poly(ethylene terephthalate) (PET) releases mono(2-hydroxyethyl) terephthalate (MHET) as a major product, the accumulation of which can prolong reactor residence times and complicate downstream monomer separations. The use of a MHETase enzyme can enable MHET hydrolysis to the monomers, terephthalic acid and ethylene glycol, but industrial PETases typically operate at thermophilic temperatures and the well-known MHETase from Ideonella sakaiensis is a mesophilic enzyme, thus warranting the development of thermophilic MHETases. Here, we characterize thermostable MHET-active enzymes from a natural diversity screen by applying a hidden Markov model based on the previously reported, archaeal ferulic acid esterase, PET46. We identified enzymes with higher thermostability than PET46 and quantified their MHETase activity in reactions at 70 °C. The crystal structure of MHT077, the homologue with the highest MHETase activity and an apparent melting temperature (T m,app ) of 94.6 °C, informed site saturation mutagenesis in the active site and lid-domain interface. MHT077 exhibited a ∼100-fold slower unfolding rate at 65 °C than PET46, indicating substantially greater kinetic stability. In parallel, we applied evolution-informed design, a probabilistic model that leverages coevolutionary patterns in large multiple sequence alignments, to improve the activity and thermostability of five ferulic acid esterases. One design, EV-MHT043–5 was identified with a comparable thermostability (T m,app = 96.1 °C) and a 3-fold improvement in its MHETase activity relative to the wildtype enzyme, MHT043. Combination variants of beneficial mutations were screened and afforded a variant, MHT077 LFK , which reduced MHET accumulation in bioreactor experiments with postconsumer PET waste. Overall, this study expands the known MHET-hydrolyzing protein scaffolds available for enzymatic PET recycling.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Integrated Risk-Informed Decision-Making for an ALMR PRISM

Decision-making is the process of identifying decision alternatives, assessing those alternatives based on predefined metrics, selecting an alternative (i.e., making a decision), and then implementing that alternative. The generation of decisions requires a structured, coherent process, or a decision-making process. The overall objective for this work is that the generalized framework is adopted into an autonomous decision-making framework and tailored to specific requirements for various applications. In this context, automation is the use of computing resources to make decisions and implement a structured decision-making process with limited or no human intervention. The overriding goal of automation is to replace or supplement human decision makers with reconfigurable decision-making modules that can perform a given set of tasks rationally, consistently, and reliably. Risk-informed decision-making requires a probabilistic assessment of the likelihood of success given the status of the plant/systems and component health, and a deterministic assessment between plant operating parameters and reactor protection parameters to prevent unnecessary trips and challenges to plant safety systems. The probabilistic portion of the decision-making engine of the supervisory control system is based on the control actions associated with an ALMR PRISM. Newly incorporated into the probabilistic models are the prognostic/diagnostic models developed by Pacific Northwest National Laboratory. These allow decisions to incorporate the health of components into the decision–making process. Once the control options are identified and ranked based on the likelihood of success, the supervisory control system transmits the options to the deterministic portion of the platform. The deterministic portion of the decision-making engine uses thermal-hydraulic modeling and components for an advanced liquid-metal reactor Power Reactor Inherently Safe Module. The deterministic multi-attribute decision-making framework uses various sensor data (e.g., reactor outlet temperature, steam generator drum level) and calculates its position within the challenge state, its trajectory, and its margin within the controllable domain using utility functions to evaluate current and projected plant state space for different control decisions. The metrics that are evaluated are based on reactor trip set points. The integration of the deterministic calculations using multi-physics analyses and probabilistic safety calculations allows for the examination and quantification of margin recovery strategies. This also provides validation of the control options identified from the probabilistic assessment. Thus, the thermalhydraulics analyses are used to validate the control options identified from the probabilistic assessment. Future work includes evaluating other possible metrics and computational efficiencies, and developing a user interface to mimic display panels at a modern nuclear power plant.

97 MATHEMATICS AND COMPUTING↗

Transition from Consultation to Monitoring-NRC's Increasingly Focused Review of Factors Important to F-Area Tank Farm Facility Performance - 13153

In consultation with the NRC, DOE issued a waste determination for the F-Area Tank Farm (FTF) facility in March 2012. The FTF consists of 22 underground tanks, each 2.8 to 4.9 million liters in capacity, used to store liquid high-level waste generated as a result of spent fuel reprocessing. The waste determination concluded stabilized waste residuals and associated tanks and auxiliary components at the time of closure are not high-level and can be disposed of as LLW. Prior to issuance of the final waste determination, during the consultation phase, NRC staff reviewed and provided comments on DOE's revision 0 and revision 1 FTF PAs that supported the waste determination and produced a technical evaluation report documenting the results of its multi-year review in October 2011. Following issuance of the waste determination, NRC began to monitor DOE disposal actions to assess compliance with the performance objectives in 10 CFR Part 61, Subpart C. To facilitate its monitoring responsibilities, NRC developed a plan to monitor DOE disposal actions. NRC staff was challenged in developing a focused monitoring plan to ensure limited resources are spent in the most cost-effective manner practical. To address this challenge, NRC prioritized monitoring areas and factors in terms of risk significance and timing. This prioritization was informed by NRC staff's review of DOE's PA documentation, independent probabilistic modeling conducted by NRC staff, and NRC-sponsored research conducted by the Center for Nuclear Waste Regulatory Analyses in San Antonio, TX. (authors)

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗

Data Analysis with Graphical Models: Software Tools

Probabilistic graphical models (directed and undirected Markov fields, and combined in chain graphs) are used widely in expert systems, image processing and other areas as a framework for representing and reasoning with probabilities. They come with corresponding algorithms for performing probabilistic inference. This paper discusses an extension to these models by Spiegelhalter and Gilks, plates, used to graphically model the notion of a sample. This offers a graphical specification language for representing data analysis problems. When combined with general methods for statistical inference, this also offers a unifying framework for prototyping and/or generating data analysis algorithms from graphical specifications. This paper outlines the framework and then presents some basic tools for the task: a graphical version of the Pitman-Koopman Theorem for the exponential family, problem decomposition, and the calculation of exact Bayes factors. Other tools already developed, such as automatic differentiation, Gibbs sampling, and use of the EM algorithm, make this a broad basis for the generation of data analysis software.

Buntine, Wray L.↗

A Review of Bayesian Networks for Spatial Data

We report Bayesian networks are a popular class of multivariate probabilistic models as they allow for the translation of prior beliefs about conditional dependencies between variables to be easily encoded into their model structure. Due to their widespread usage, they are often applied to spatial data for inferring properties of the systems under study and also generating predictions for how these systems may behave in the future. We review published research on methodologies for representing spatial data with Bayesian networks and also summarize the application areas for which Bayesian networks are employed in the modeling of spatial data. We find that a wide variety of perspectives are taken, including a GIS-centric focus on efficiently generating geospatial predictions, a statistical focus on rigorously constructing graphical models controlling for spatial correlation, as well as a range of problem-specific heuristics for mitigating the effects of spatial correlation and dependency arising in spatial data analysis. Special attention is also paid to potential future directions for integration of Bayesian networks with spatial processes.

97 MATHEMATICS AND COMPUTING↗

Risk assessment of turbine rotor failure using probabilistic ultrasonic non-destructive evaluations

The study presents a method and application of risk assessment methodology for turbine rotor fatigue failure using probabilistic ultrasonic nondestructive evaluations. A rigorous probabilistic modeling for ultrasonic flaw sizing is developed by incorporating the model-assisted probability of detection, and the probability density function (PDF) of the actual flaw size is derived. Two general scenarios, namely the ultrasonic inspection with an identified flaw indication and the ultrasonic inspection without flaw indication, are considered in the derivation. To perform estimations for fatigue reliability and remaining useful life, uncertainties from ultrasonic flaw sizing and fatigue model parameters are systematically included and quantified. The model parameter PDF is estimated using Bayesian parameter estimation and actual fatigue testing data. The overall method is demonstrated using a realistic application of steam turbine rotor, and the risk analysis under given safety criteria is provided to support maintenance planning.

42 ENGINEERING↗

The relationship between anatomic noise and volumetric breast density for digital mammography

Purpose: The appearance of parenchymal/stromal patterns in mammography have been characterized as having a Wiener power spectrum with an inverse power-law shape described by the exponential parameter, {beta}. The amount of fibroglandular tissue, which can be quantified in terms of volumetric breast density (VBD), influences the texture and appearance of the patterns formed in a mammogram. Here, a large study is performed to investigate the variations in {beta} in a clinical population and to indicate the relationship between {beta} and breast density. Methods: From a set of 2686 cranio-caudal normal screening mammograms, the parameter {beta} was extracted from log-log fits to the Wiener spectrum over the range 0.15-1 mm{sup -1}. The Wiener spectrum was calculated from regions of interest in the compression paddle contact region of the breast. An in-house computer program, Cumulus V, was used to extract the volumetric breast density and identify the compression paddle contact regions of the breast. The Wiener spectra were calculated with and without modulation transfer function (MTF) correction to determine the impact of VBD on the intrinsic anatomic noise. Results: The mean volumetric breast density was 25.5% ({+-}12.6%) over all images. The mean {beta} following a MTF correction which decreased the {beta} slightly ( Almost-Equal-To -0.08) was found to be 2.87. Varying the maximum of the spatial frequency range of the fits from 0.7 to 1.0, 1.25 or 1.5 mm{sup -1} showing small decreases in the result, although the effect of the quantum noise power component on reducing {beta} was clearly observed at 1.5 mm{sup -1}. Conclusions: The texture parameter, {beta}, was found to increase with VBD at low volumetric breast densities with an apparent leveling off at higher densities. The relationship between {beta} and VBD measured here can be used to create probabilistic models for computer simulations of detectability. As breast density is a known risk predictor for breast cancer, the correlation between {beta} and VBD suggests that {beta} may provide predictive information and this will be investigated in the future.

62 RADIOLOGY AND NUCLEAR MEDICINE↗

Grid Frequency Extreme Event Analysis and Modeling: Preprint

Sudden losses of generation or load can lead to instantaneous changes in electric grid frequency and voltage. Extreme frequency events pose a major threat to grid stability. As renewable energy sources supply power to grids in increasing proportions, it becomes increasingly important to examine when and why extreme events occur to prevent destabilization of the grid. To better understand frequency events, including extrema, historic data were analyzed to fit probability distribution functions to various frequency metrics. Results showed that a standard Cauchy distribution fit the difference between the frequency nadir and prefault frequency (f_(C-A)) metric well, a standard Cauchy distribution fit the settling frequency (f_B) metric well, and a standard normal distribution fit the difference between the settling frequency and frequency nadir (f_(B-C)) metric very well. Results were inconclusive for the frequency nadir (f_C) metric, meaning it likely has a more complex distribution than those tested. This probabilistic modeling should facilitate more realistic modeling of grid faults.

24 POWER TRANSMISSION AND DISTRIBUTION↗

Probabilistic material degradation model for aerospace materials subjected to high temperature, mechanical and thermal fatigue, and creep

A probabilistic general material strength degradation model has been developed for structural components of aerospace propulsion systems subjected to diverse random effects. The model has been implemented in two FORTRAN programs, PROMISS (Probabilistic Material Strength Simulator) and PROMISC (Probabilistic Material Strength Calibrator). PROMISS calculates the random lifetime strength of an aerospace propulsion component due to as many as eighteen diverse random effects. Results are presented in the form of probability density functions and cumulative distribution functions of lifetime strength. PROMISC calibrates the model by calculating the values of empirical material constants.

Boyce, L.↗

Interactive statistical decision tools (GiSdT): An open source software integration of decision analysis and stakeholder judgment for sustainable waste management, disposal and remediation - 15650

The vast majority of remediation and waste management decisions are made without quantitative consideration of economic and socio-political factors. They are made instead based on quantitative metrics of human health risk. Sustainable decisions need to be made based on all three 'pillars of sustainability' (economics, environment and social), and require understanding and characterization of the costs and values associated with each pillar. In addition, such decisions need to conform to regulatory or other legal requirements, which often constrains the decision space of interest. Although efforts are often made to include factors across all three pillars (for example, for decisions made under CERCLA), these efforts are usually qualitative, and hence difficult to defend. They lack technical defensibility, transparency and traceability. The purpose of GiSdT (Guided Interactive Statistical Decision Tools) is to provide a quantitative framework whereby all aspects of a decision problem, such as remediation and waste management decisions, can be addressed quantitatively, and hence, defensibly, transparently and traceably. Most decision-makers do not currently have access to useful or usable methods and approaches when are presented with choices that have significant impacts across all three pillars of sustainability. The goal of GiSdT is to provide that access by identifying or developing effective and user-friendly decision methods and approaches that empower decision-makers to explicitly and routinely incorporate all aspects of sustainability into their decision-making. To identify and develop these methods and approaches, GiSdT provides the tools needed for decision-makers and stakeholders to understand and characterize their knowledge of their current decision-making processes, and what methods and approaches they need to pro-actively and quantitatively address all aspects of sustainability in their decisions. GiSdT is an open source interactive framework program that provides the tools needed to address the relevant components of a sustainability-based decision problem, including describing the decision landscape, translating that landscape into goals and objectives, valuing the measurable attributes that describe each objective, identifying decision options and addressing the uncertainty in the attributes (e.g., human health risk, institutional controls) through appropriate probabilistic models. In addition to addressing these quantitative technical components, the GiSdT program is aimed at facilitating stakeholder involvement so that the perspectives, values and objectives are all relevant stakeholders are addressed explicitly in the decision making process. The technical underpinnings of our software program, called Guided Interactive Statistical Decision Tools (GiSdT), are described. (authors)

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗

A Hybrid AI/ML and Computational Mechanics Based Approach for Time-Series State and Fatigue Life Estimation of Nuclear Reactor Components

Environmental fatigue modeling is a complex problem due to multiple failure modes and their intermixing. The failure modes are function of various underlying causes in addition to the corrosive effect of reactor coolant environment. Some of the major causes are time-dependence of material associated with cyclic loading, load sequence effect associated with random/variable amplitude loading, effect of strain amplitude and rates, effect of varying temperature (along both temporal and spatial directions) and the effect of mean strain and stress. The nonlinear intermixing of failure modes associated with above mentioned causing parameters makes the environmental fatigue modeling is a challenging task. Because of this challenge, fatigue is traditionally being modeled based on experimental data. However, test based empirical approach often requires hundreds of fatigue tests to model the above-mentioned intermixing failure causes even for a single material system. The problem is further exaggerated for reactor component made from multi-material systems such as made from both carbon and stainless-steel base metals and their similar and dissimilar metal welds. With the difficulty of conducting hundreds of fatigue tests to capture the above-mentioned intermixing failure causes, fatigue modeling approaches often depends on empirical models based on limited available test data such as available through ASME code and NUREG 6909. However, these limited test-data-based models may not be enough to accurately predict the life of reactor components. Accurate prediction of life of reactor component would become a necessity, particularly when the license of the reactors to be extended for long-term-operation (LTO) that is for well beyond its original design life of 40 years. The requirement of extending the license of reactor under LTO requires hundreds of fatigue tests to be conducted to understand the mechanism associated with the above-mentioned interdependent failure causes. However, conducting large number of fatigue tests is not a feasibility due to the cost involved. To address this issues Argonne National Laboratory (ANL) with the sponsorship of DOE Light Water Reactor Sustainability (LWRS) program trying to develop a hybrid predictive modeling approach. This is based on limited experiment-data, Artificial-intelligence (AI) – Machine-Learning (ML) - Deep-Learning (DL) based techniques and Multiphysics-computational-mechanics based modeling tools. The hybrid approach not-only can improve the accuracy of the existing stress analysis and fatigue modeling approach but also can reduce the over-dependency on test-based approach. Towards this goal following are some of the major contributions based on ANL’s FY-20 environmental fatigue modeling activities: 1) A cyclic plasticity material model database for 82/182 dissimilar metal weld, which can be readily shared with US nuclear industry and regulatory agency on request. 2) A well validated analytical modeling methodology to perform cycle-by-cycle stress prediction under both constant amplitude fatigue loading and variable amplitude fatigue loading (with load-sequence effect). 3) An AI/ML/DL based methodology to predict unmeasurable cyclic strain based on other available sensor signals. This type of approach can be used for estimating strain in real reactor components from other sensor readings. 4) An AI/ML based approach to improve the US capability on environmental fatigue testing. This is by improving ANL’s existing environmental fatigue testing capacity to conduct ASME required strain-controlled tests (by controlling strain amplitudes and its rate), while not measuring the strain (due to the difficulty of placing an extensometer in a narrow autoclave in a PWR-water-test system). 5) A simulation and experiment based probabilistic modeling methodology for time-series fatigue state and life estimation of reactor metal such as dissimilar metal weld.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗

CAHS: Context-Aware Homology Search

Protein homology search is foundational to bioinformatics: it supports annotation transfer, structure/function inference, and evolutionary analysis over rapidly expanding sequence repositories (e.g., UniProtKB). Profile hidden Markov models (pHMMs), as implemented in HMMER, remain the most widely trusted approach because they provide statistically calibrated E-values; however, their gap behavior is fixed once a profile is trained, despite biological evidence that insertion/deletion tolerance varies across flexible loops and intrinsically disordered regions. We present CAHS (Context-Aware Homology Search), a lightweight query-time adapter for pHMM search that incorporates learned and biologically motivated signals without changing HMMER's downstream search pipeline or its calibrated E-value reporting. Given a query sequence, CAHS computes per-residue representations from a protein language model and a disorder predictor, maps these to profile coordinates, and modulates only match-state transition rows (gap-open and gap-extension probabilities) while preserving Plan7 constraints. We comprehensively evaluate CAHS across six structurally diverse protein families and multi-domain architectures against a 570k-sequence target corpus. CAHS expands detection capability, retrieving thousands of additional remote homologs at relaxed thresholds by maintaining alignment quality through flexible regions. For multi-domain proteins, context-aware modulation resolves 94% of fragmented alignments. Crucially, CAHS preserves hit-set invariance at stringent operating points (E<10-10), demonstrating increased statistical confidence without inflating false positives. Furthermore, sharper statistical distinction between homologs and background noise during early filter stages yields up to a 3.87× acceleration in end-to-end wall-clock time on high-performance computing clusters. Overall, CAHS illustrates a practical AI-for-science design pattern: augmenting a trusted probabilistic model with query-specific learned signals to improve interpretable, reproducible inference in data-rich biology.

Bhattaram, Swethasree [Georgia Institute of Techno↗

Enhancing Unknown Waveform Detection by Learning Intra and Inter-domain Dependencies with Advanced Attention Fusion Mechanisms

Detection of unknown waveforms in mission-critical communications is a crucial area of interest for the Department of Energy (DoE). Traditional methods and recent deep learning-based approaches often assume that the training set includes all possible classes, which is impractical for detecting new waveforms. This limitation gives rise to the problem of open-set recognition (OSR), which involves correctly identifying known classes while detecting and rejecting unknown or unseen classes. To address this limitation, we propose a novel dual-domain complex-valued neural architecture that jointly processes time-domain and frequency-domain signal representations using transformer mechanisms. A transformer model is a deep learning architecture that uses self-attention mechanisms to process and learn relationships in sequential data. Our model employs a cosine similarity loss to extract domain-specific features and incorporates a transformer architecture in the latent space to weigh the importance of different features from the time and frequency domains. The transformer layer includes stacked self-attention and cross-attention modules to learn intra-domain and inter-domain dependencies, creating a more holistic signal representation. An attention-based fusion module intelligently combines the time and frequency-domain features using multi-head attention, enabling the network to learn the optimal feature for each domain in each input signal. Quantitative results demonstrate the impact of these architectural choices on overall performance, showing significant improvement after incorporating self and cross-attention modules and using complex attention fusion over simple weighted fusion. Our ongoing work will focus on addressing the limitations of threshold-based OSR methods by developing a novel generative framework that integrates a conditional diffusion probabilistic model (DPM). DPM is a generative framework that learns to synthesize complex data by reversing a gradual noising process using a neural network trained to denoise step-by-step. Our goal is to leverage the inherent strengths of DPMs for identifying unknown signals more robustly. One primary advantage of using a DPM is its ability to provide a more reliable anomaly score based on the model's reconstruction error, rather than relying solely on classifier confidence. Additionally, the iterative denoising process of DPMs makes this approach naturally resilient to low Signal-to-Noise Ratio (SNR) conditions, where traditional methods often fail. By implementing this generative framework, we aim to enhance the model's capability to accurately detect unknown waveforms and maintain performance in challenging environments.

99 - GENERAL AND MISCELLANEOUS↗