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 901 records · Page 50

A toolkit for integrated deterministic and probabilistic assessment for hydrogen infrastructure.

There has been increasing interest in using Quantitative Risk Assessment [QRA] to help improve the safety of hydrogen infrastructure and applications. Hydrogen infrastructure for transportation (e.g. fueling fuel cell vehicles) or stationary (e.g. back-up power) applications is a relatively new area for application of QRA vs. traditional industrial production and use, and as a result there are few tools designed to enable QRA for this emerging sector. There are few existing QRA tools containing models that have been developed and validated for use in small-scale hydrogen applications. However, in the past several years, there has been significant progress in developing and validating deterministic physical and engineering models for hydrogen dispersion, ignition, and flame behavior. In parallel, there has been progress in developing defensible probabilistic models for the occurrence of events such as hydrogen release and ignition. While models and data are available, using this information is difficult due to a lack of readily available tools for integrating deterministic and probabilistic components into a single analysis framework. This paper discusses the first steps in building an integrated toolkit for performing QRA on hydrogen transportation technologies and suggests directions for extending the toolkit.

08 HYDROGEN↗

Optimization Routine for Generating Medical Kits for Spaceflight Using the Integrated Medical Model

The Integrated Medical Model (IMM) is a MATLAB model that provides probabilistic assessment of the medical risk associated with human spaceflight missions.Different simulations or profiles can be run in which input conditions regarding both mission characteristics and crew characteristics may vary. For each simulation, the IMM records the total medical events that occur and “treats” each event with resources drawn from import scripts. IMM outputs include Total Medical Events (TME), Crew Health Index (CHI), probability of Evacuation (pEVAC), and probability of Loss of Crew Life (pLOCL).The Crew Health Index is determined by the amount of quality time lost (QTL). Previously, an optimization code was implemented in order to efficiently generate medical kits. The kits were optimized to have the greatest benefit possible, given amass and/or volume constraint. A 6-crew, 14-day lunar mission was chosen for the simulation and run through the IMM for 100,000 trials. A built-in MATLAB solver, mixed-integer linear programming, was used for the optimization routine. Kits were generated in 10% increments ranging from 10%-100% of the benefit constraints. Conditions wheremass alone was minimized, volume alone was minimized, and where mass and volume were minimizedjointly were tested.

Medical Kit↗

The nature and orbit of the Ophiuchus stream

The Ophiuchus stream is a recently discovered stellar tidal stream in the Milky Way. We present high-quality spectroscopic data for 14 stream member stars obtained using the Keck and MMT telescopes. We confirm the stream as a fast moving (v{sub los} ∼ 290 km s{sup −1}), kinematically cold group (σ{sub v{sub l{sub o{sub s}}}}≲1 km s{sup −1}) of α-enhanced and metal-poor stars ([α/Fe] ∼ 0.4 dex, [Fe/H] ∼ −2.0 dex). Using a probabilistic technique, we model the stream simultaneously in line-of-sight velocity, color–magnitude, coordinate, and proper motion space, and so determine its distribution in 6D phase-space. We find that the stream extends in distance from 7.5 to 9 kpc from the Sun; it is 50 times longer than wide, merely appearing highly foreshortened in projection. The analysis of the stellar population contained in the stream suggests that it is ∼12 Gyr old, and that its initial stellar mass was ∼2 × 10{sup 4} M{sub ⊙} (or at least ≳7 × 10{sup 3} M{sub ⊙}). Assuming a fiducial Milky Way potential, we fit an orbit to the stream that matches the observed phase-space distribution, except for some tension in the proper motions: the stream has an orbital period of ∼350 Myr, and is on a fairly eccentric orbit (e ∼ 0.66) with a pericenter of ∼3.5 kpc and an apocenter of ∼17 kpc. The phase-space structure and stellar population of the stream show that its progenitor must have been a globular cluster that was disrupted only ∼240 Myr ago. We do not detect any significant overdensity of stars along the stream that would indicate the presence of a progenitor, and conclude that the stream is all that is left of the progenitor.

79 ASTRONOMY AND ASTROPHYSICS↗

Probabilistic Usage of the Multi-Factor Interaction Model

A Multi-Factor Interaction Model (MFIM) is used to predict the insulating foam mass expulsion during the ascending of a space vehicle. The exponents in the MFIM are evaluated by an available approach which consists of least squares and an optimization algorithm. These results were subsequently used to probabilistically evaluate the effects of the uncertainties in each participating factor in the mass expulsion. The probabilistic results show that the surface temperature dominates at high probabilities and the pressure which causes the mass expulsion at low probabil

Chamis, Christos C.↗

Performance Confirmation Strategies for the Waste Isolation Pilot Plant - A Historical Perspective from an Operating Disposal Facility - 12248

Performance confirmation is an important element of the Waste Isolation Pilot Plant (WIPP) program. Performance confirmation was first used during the early WIPP site characterization phase to focus experimental activities that address the development of probabilistic repository performance models and to address stakeholder assurance needs. The program is currently used to analyze the conditions of the repository and its surroundings to ensure that the basis for the repository's long-term radioactive waste containment predictions is valid. This basis is related to the parameters, assumptions, conceptual and numerical models that are used to predict or validate the potential radioactive waste containment performance of the system. The concept of performance confirmation for the WIPP is one that has evolved since the first repository work was initiated decades ago and plays an important role in assuring adequate repository performance both now and in the long-term. The WIPP mission has progressed from a pilot project to an operational disposal facility and will progress to eventual site closure when disposal operations are completed. Performance confirmation is an important part of each of these progressions. The concept of disposing radioactive waste in a geologic repository today involves a complete understanding of many technical, political, regulatory, societal and economic elements. Many of these elements overlap and solving all relevant issues necessary to site, operate and decommission a disposal facility should be done with knowledge of each element's requirements and impacts. Performance confirmation is one tool that can help to coordinate many of these elements into a program that actively investigates what is thought to be adequately understood about the system and what information is lacking. A performance confirmation program is used to determine ways to challenge and verify those areas that are thought to be understood and to find ways to understand those areas that are not well understood. Performance confirmation programs have been used twice at WIPP, first during site characterization and PA development and later in a Compliance Monitoring program. At first, only certain technical aspects of the system were deemed important because it was a scientifically-based, government project. Early site characterization work was design to gather information about the geology and hydrology of the area and the mechanical properties of the natural barrier. The information would be used in a PA to determine the long-term containment performance of the disposal system. A performance confirmation element was used to identify the sensitive elements of the system that were certain, well understood or justified and those that were not. It identified experimental and analytical programs that could be used to reduce uncertainty, confirm sensitive assumptions and provide useful data. This performance confirmation program provided data to justify the adequacy of the information used in PA to demonstrate compliance with EPA's containment requirements. Performance confirmation will continue to be used in the post-closure period for at least 30 years and likely up to the end of the 100-year institutional controls period. As the technical basis for the repository matures throughout the operational period, the currently planned post-closure monitoring program will need to be reassessed prior to implementation. However, the intent of the program will be the same as it was for the previous programs, to ensure the ultimate goal of the repository. This goal is to safely isolate waste from the accessible environment and ensure public and environmental safety. (authors)

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗

Data free inference with processed data products

Here, we consider the context of probabilistic inference of model parameters given error bars or confidence intervals on model output values, when the data is unavailable. We introduce a class of algorithms in a Bayesian framework, relying on maximum entropy arguments and approximate Bayesian computation methods, to generate consistent data with the given summary statistics. Once we obtain consistent data sets, we pool the respective posteriors, to arrive at a single, averaged density on the parameters. This approach allows us to perform accurate forward uncertainty propagation consistent with the reported statistics.

97 MATHEMATICS AND COMPUTING↗

Composite Multilinearity, Epistemic Uncertainty and Risk Achievement Worth

Risk Achievement Worth is one of the most widely utilized importance measures. RAW is defined as the ratio of the risk metric value attained when a component has failed over the base case value of the risk metric. Traditionally, both the numerator and denominator are point estimates. Relevant literature has shown that inclusion of epistemic uncertainty i) induces notable variability in the point estimate ranking and ii) causes the expected value of the risk metric to differ from its nominal value. We obtain the conditions under which the equality holds between the nominal and expected values of a reliability risk metric. Among these conditions, separability and state-of-knowledge independence emerge. We then study how the presence of epistemic uncertainty aspects RAW and the associated ranking. We propose an extension of RAW (called ERAW) which allows one to obtain a ranking robust to epistemic uncertainty. We discuss the properties of ERAW and the conditions under which it coincides with RAW. We apply our findings to a probabilistic risk assessment model developed for the safety analysis of NASA lunar space missions.

97 MATHEMATICS AND COMPUTING↗

Dominant balance-based adaptive mesh refinement for incompressible fluid flows

This work introduces a novel adaptive mesh refinement (AMR) method that utilizes dominant balance analysis (DBA) for efficient and accurate grid adaptation in computational fluid dynamics (CFD) simulations. The proposed method leverages a Gaussian mixture model (GMM) to classify grid cells into active and passive regions based on the dominant physical interactions within the equation space. By modeling truncation error probabilistically from discretized terms, the method identifies regions of high interaction where numerical accuracy is most sensitive to resolution. Unlike traditional AMR strategies, this approach does not rely on heuristic-based sensors or user-defined thresholds, providing a fully automated and problem-independent framework for AMR. Applied to the incompressible Navier-Stokes equations for steady and unsteady flow past a cylinder, the DBA-based AMR method achieves comparable accuracy to high-resolution grids while reducing computational costs by up to 70 %. The validation highlights the method’s effectiveness in capturing complex flow features while minimizing grid cells, directing computational resources toward regions with the most critical dynamics. This modular and scalable strategy is adaptable to a wide range of applications, presenting a promising tool for efficient high-fidelity simulations in CFD and other multiphysics domains.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Statistical analysis of S—N type environmental fatigue data of Ni-base alloy welds using weibull distribution

In this study, the probabilistic fatigue life model for Ni-base alloys was developed based on the Weibull distribution using statistical analysis of fatigue data reported in NUREG/CR-6909 and the new fatigue data of Alloy 52M/152 and 82/182. The developed Weibull model can consider right-censored data (i.e., non-failed data) and quantify the improved safety (or reliability) based on the level of failure probability. The overall margin in the current fatigue design limit model (ASME design curve + NUREG/CR-6909 F en model) is similar to that of the Weibull model with a cumulative failure probability of approximately 2.5%. The margin in the current fatigue design limit model demonstrated inconsistencies for the Ni-base alloy weld data, whereas the Weibull model showed a consistent margin. Therefore, the Weibull model can systematically mitigate the excessive safety margin.

36 MATERIALS SCIENCE↗

Intensity measure adequacy assessment for nonlinear site response using Information Theory

Nonlinear site response modeling is a crucial aspect of Probabilistic Seismic Hazard Analysis. Site amplification models routinely rely on a rock intensity measure to characterize the strength of the bedrock motion. However, the adequacy of such intensity measures towards predicting amplifications across the oscillator period range has not been investigated in the literature. This paper analyzes the adequacy of rock intensity measures using state of the art criterion established in Performance-Based Earthquake Engineering and techniques from Information Theory. The efficiency and the sufficiency of several rock intensity measure are assessed. It was found that spectral accelerations at low periods usually are adequate for predicting amplifications across the period range. This supports the current practice of using Peak Ground Acceleration in Ground Motion Models. However, for extremely soft sites, which demonstrate nonlinear effects well into the period range, it is better practice to ensure that amplification and spectral acceleration share the same oscillator period. Lastly, for predicting the peak shear strain (an important parameter that controls nonlinearity of site response), Peak Ground Velocity is generally adequate, and this conclusion is in line with the definition of proxy shear strain.

42 ENGINEERING↗

Big Data Sanitization and Cyber Situational Awareness: A Network Telescope Perspective

This paper addresses the problems of data sanitization and cyber situational awareness by analyzing 910 GB of real Internet-scale traffic, which has been passively collected by monitoring close to 16.5 million darknet IP addresses from a /8 and a /13 network telescopes. First, the paper offers a novel probabilistic darknet preprocessing model, which aims at sanitizing darknet data to prepare it for effective use in the task of cyber threat intelligence generation. Such model has been engineered using a distributed multithreaded approach, rendering it highly effective on darknet big data. Second, the paper further contributes by presenting an innovative approach to infer large-scale orchestrated probing campaigns by leveraging darknet data, for Internet cyber situational awareness. The approach uniquely reduces the dimensionality of such big data by utilizing its artifacts, instead of processing the actual raw data. This is accomplished by extracting and analyzing probing time series using formal methods rooted in Fourier transform and Kalman filtering. Thorough empirical evaluations indeed validate the accuracy and the performance of the proposed methods. We assert that such approaches are of significant value, given their highly applicable nature to the field of Internet measurements for cyber security in the era of big data.

Bou-Harb, Elias↗

Reliability-Informed Economic and Energy Evaluation for Bi-Level Design for Remanufacturing: A Case Study of Transmission and Hydraulic Manifold

Abstract Design for remanufacturing (DfRem) is one attractive strategy that encourages the reuse of a product and extends the product's life cycle. Traditional design processes often only consider product reliability at an early design stage. However, from the perspective of environmental sustainability, it is becoming increasingly important to evaluate the long-term economic and environmental impacts of design decisions during early-stage design. We propose a bi-level DfRem framework consisting of system-level reusability allocation and component-level design tradeoff analysis, considering reliability and product warranty policy. First, a system-level reusability allocation problem aims at a theoretical exploration of the design space where all the components comprising the system are allocated certain reuse rates to achieve target energy savings with minimum cost. Following the theoretical exploration at the system level, a component-level analysis looks at practical design options for each component and trades-off between the overall cost and energy consumption for multiple remanufacturing cycles. Both levels of the framework require modeling component reuse for multiple remanufacturing cycles, which we achieve by using a branched power-law model that provides probabilistic scenarios of reusing the component or replacing it with a new part. We demonstrate the utility of this framework with the case study of an infinitely variable transmission (IVT) used by some agricultural machines manufactured by John Deere and show snapshots of a prototype software tool that we developed for easy use by designers.

Engineering↗

Reliability-Informed Life-Cycle Warranty Cost Analysis: A Case Study on a Transmission in Agricultural Equipment

In agricultural and industrial equipment, both new and remanufactured systems are often available for warranty coverage. In such cases, it may be challenging for equipment manufacturers to properly trade-off between the system reliability and the cost associated with a replacement option (e.g., replace with a new or remanufactured system). To address this problem, we present a reliability-informed life-cycle warranty cost (LCWC) analysis framework that enables equipment manufacturers to evaluate different warranty policies. These warranty policies differ in whether a new or remanufactured system is used for replacement in the case of product failure. The novelty of this LCWC analysis framework lies in its ability to incorporate real-world field reliability data into warranty policy assessment using probabilistic warranty cost models that consider multiple life cycles. First, the reliability functions for the new and remanufactured systems are built as the time-to-failure distributions that provide the best-fit to the field reliability data. Then, these reliability functions and their corresponding warranty policies are used to build the LCWC models according to the specific warranty terms. Finally, Monte Carlo simulation is used to propagate the time-to-failure uncertainty of each system, modeled by its reliability function, through each LCWC model to produce a probability distribution of the LCWC. The effectiveness of the proposed reliability-informed LCWC analysis framework is demonstrated with a real-world case study on a transmission used in some agricultural equipment.

agricultural equipment↗

Reliability-Informed Economic and Energy Evaluation for Design for Remanufacturing: A Case Study on a Hydraulic Manifold

Design for Remanufacturing (DfRem) is an attractive approach for sustainable product development. Evaluation of DfRem strategies, from both economic and environmental perspectives, at an early design stage can allow the designers to make informed decisions when choosing the best design option. Studying the long-term implications of a particular design scenario requires quantifying the benefits of remanufacturing for multiple life cycles while considering the reliability of the product. In addition to comparing designs on a one-to-one basis, we find that including reliability provides a different insight into comparing design strategies. We present a reliability-informed cost and energy analysis framework that accounts for product reliability for multiple remanufacturing cycles within a certain warranty policy. The variation of reuse rate over successive remanufacturing cycles is formulated using a branched power-law model which provides probabilistic scenarios of reusing or replacing with new units. To demonstrate the utility of this framework, we use the case study of a hydraulic manifold, which is a component of a transmission used in some agricultural equipment, and use real-world field reliability data to quantify the transmission’s reliability. Three design improvement changes are proposed for the manifold and we quantify the costs and energy consumption associated with each of the design changes for multiple remanufacturing cycles.

design for remanufacturing↗

Computer-aided detection using non-convolutional neural network Gaussian processes

Deep convolutional neural networks (CNNs) have in recent years achieved record-breaking performance on many image classification tasks and are therefore well-suited for computer aided detection (CAD). The need for uncertainty quantification for CAD motivates the need for a probabilistic framework for deep learning. The most well-known probabilistic neural network model is the Bayesian neural network (BNN), but BNNs are notoriously difficult to sample for large complex network architectures, and as such their use is restricted to small problems. It is known that the limit of BNNs as their widths increase toward infinity is a Gaussian process (GP), and there has been considerable research interest in these infinitely wide BNNs. Recently, this classic result has been extended to deep architectures in what is termed the neural network Gaussian process (NNGP) model. In this work, we implement an NNGP model and apply it to the ChestXRay14 dataset at the full resolution of 1024x1024 pixels. Even without any convolutional aspects to the network architecture and without any data augmentation, our five layer deep NNGP model outperforms other non-convolutional models and therefore helps to narrow the performance gap between non-convolutional and convolutional models. Our NNGP model is fully Bayesian and therefore offers uncertainty information through its predictive variance that can be used to formulate a predictive confidence measure. We show that the performance of the NNGP model is significantly boosted after low-confidence predictions are rejected, suggesting that convolution is most beneficial only for these low-confidence examples. Finally, our results indicate that an extremely large fully-connected neural network with appropriate regularization could perform as well as the NNGP if not for the computational bottleneck resulting from the large number of model parameters.

Agrawal, Devanshu↗

Automata-based Debugging for Dissimilar Parallel Tasks

AutomaDeD is a tool that automatically diagnoses performance and correctness faults in MPI applications. It creates time and control-flow models of processes in an MPI program, and when a failure occus, these models are analyzed probabilistically to find the origin of the failure. The output of the tool is MPI processes ranked by the abnormality degree and regions of code where faults are first manifested. The granularity of code region is an MPI call or a computation region between two MPI calls.

Source record↗

Combining biomarker and virus phylogenetic models improves HIV-1 epidemiological source identification

To identify and stop active HIV transmission chains new epidemiological techniques are needed. Here, we describe the development of a multi-biomarker augmentation to phylogenetic inference of the underlying transmission history in a local population. HIV biomarkers are measurable biological quantities that have some relationship to the amount of time someone has been infected with HIV. To train our model, we used five biomarkers based on real data from serological assays, HIV sequence data, and target cell counts in longitudinally followed, untreated patients with known infection times. The biomarkers were modeled with a mixed effects framework to allow for patient specific variation and general trends, and fit to patient data using Markov Chain Monte Carlo (MCMC) methods. Subsequently, the density of the unobserved infection time conditional on observed biomarkers were obtained by integrating out the random effects from the model fit. This probabilistic information about infection times was incorporated into the likelihood function for the transmission history and phylogenetic tree reconstruction, informed by the HIV sequence data. To critically test our methodology, we developed a coalescent-based simulation framework that generates phylogenies and biomarkers given a specific or general transmission history. Testing on many epidemiological scenarios showed that biomarker augmented phylogenetics can reach 90% accuracy under idealized situations. Under realistic within-host HIV-1 evolution, involving substantial within-host diversification and frequent transmission of multiple lineages, the average accuracy was at about 50% in transmission clusters involving 5–50 hosts. Realistic biomarker data added on average 16 percentage points over using the phylogeny alone. Using more biomarkers improved the performance. Shorter temporal spacing between transmission events and increased transmission heterogeneity reduced reconstruction accuracy, but larger clusters were not harder to get right. More sequence data per infected host also improved accuracy. We show that the method is robust to incomplete sampling and that adding biomarkers improves reconstructions of real HIV-1 transmission histories. The technology presented here could allow for better prevention programs by providing data for locally informed and tailored strategies.

60 APPLIED LIFE SCIENCES↗