Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “machine reasoning”

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 289 records · Page 16
All sourcesNASA NTRSDOE OSTI
DOE OSTI2022

Instantiation of HCML Demonstrating Bayesian Predictive Modeling for Attentional Control

The research team developed models of Attentional Control (AC) that are unique to existing modeling approaches in the literature. The goal was to enable the research team to (1) make predictions about AC and human performance in real-world scenarios and (2) to make predictions about individual characteristics based on human data. First, the team developed a proof-of-concept approach for representing an experimental design and human subjects data in a Bayesian model, then demonstrated an ability to draw inferences about conditions of interest relevant to real-world scenarios. Ultimately, this effort was successful, and we were able to make reasonable (meaning supported by behavioral data) inferences about conditions of interest to develop a risk model for AC (where risk is defined as a mismatch between AC and attentional demand). The team additionally defined a path forward for a human-constrained machine learning (HCML) approach to make predictions about an individual's state based on performance data. The effort represents a successful first step in both modeling efforts and serves as a basis for future work activities. Numerous opportunities for future work have been defined.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2020

An Applied Strategy for Using Empirical and Hybrid Models in Online Monitoring

The monitoring of plant equipment for failure prediction is one of the key contributors to operation and maintenance (O&M) costs for a nuclear power plant (NPP) because O&M monitoring depends on labor-intensive activities that are required to meet high equipment reliability standards. These activities rely primarily on humans for information gathering, condition diagnosis, and predictive analysis. Online monitoring aims to automate these activities by relying on sensors to replace human information gathering and machine learning to replace human analysis and decision making. To facilitate automated monitoring, a systematic strategy for anomaly detection is needed to optimally use the available sensor data, empirical models, and physics-supported models. This strategy is essential to provide credible reasoning on why and when an empirical (i.e., purely data-driven) versus hybrid (i.e., physics-supported) approach should be used and to determine the ideal mix of these two approaches for a defined anomaly detection scope. The extant methods usually adopt an ad hoc trial-and-error approach that, in addition to being time-consuming and costly, is also highly subjective; it is impacted by the background and the skill set of the personnel making the decisions. Thus, such an approach cannot guarantee an optimum outcome. This represents the motivation of the current research effort, which is focused on devising a scientifically supported strategy for the optimum selection of anomaly detection methods. This report presents a detailed assessment of the main anomaly detection techniques within the empirical or hybrid method streams. Empirical methods include pattern, statistical, and causal inference. Hybrid methods include the use of physics models to train and test data methods, reduce data dimensionality, reduce data-model complexity, augment data, and reduce empirical uncertainty; hybrid methods also include the use of data to tune physics models. The listed techniques within these two streams represent the vast majority of techniques performed for anomaly detection. Using the techniques as outcomes, a strategy was developed to enable a systematic decision-making process to lead to one of these techniques. The strategy is driven by key decision points related to data relevance, simple modeling feasibility, data inference, physics-modeling value, data dimensionality, physics knowledge, method of validation, performance, data availability and suitability for training and testing, cause-effect, entropy inference, and model fitting. Each of these decision points in the strategy is explained in detail in this report with examples, along with the scientific basis behind the decisions and outcomes in common and simplified terminology. The strategy is developed for use by any NPP staff with basic engineering or science knowledge. A user-friendly graphical state flow diagram was also developed as a visual presentation of the strategy. The strategy was tested and demonstrated through two pilot projects for the application of anomaly detection at an NPP. Each pilot had two use cases: an initial case in which certain decisions were made that resulted in one or more empirical techniques and a revised use case where one or more key decisions were modified resulting in using a set of hybrid methods.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗
DOE OSTI2022

Impact of a track trigger at a multi-TeV muon collider

In the panorama of future particle accelerators, the Muon Collider seems to be the best choice for several reasons that will be explored in section 1.2. The main challenge is represented by the instability of the muons. At the desired instantaneous luminosities, an intense flow of secondary and tertiary particles from the interactions of the beam decay products with the machine elements reaches the detector, compromising its performance. Several studies are currently underway to mitigate the background and to have an efficient trigger. My work represents a first step towards understanding the basic characteristics of a trigger selection, which will subsequently guide the design of the trigger system. With this aim I produced and analyzed Higgs and di-Higgs signals with MadGraph and CERN-ROOT. In first section there is an introduction to Muon Collider, a brief history, a discussion of its benefits, a description of the detector and of the effects of the Beam Induced Background. Several operating center-of-mass energies are possible for this machine, in this report a center-of-mass of 3 TeV is assumed. Then I describe the event generation process. Finally I describe the entire analysis, in particular I analyzed the efficiency and the number of events for each process.

43 PARTICLE ACCELERATORS↗
NASA NTRSDate not supplied

Predicting Adverse Events and their Precursors in Aviation Using Multi-Class Multiple-Instance Learning

In recent years, there has been a rapid growth in the application of machine learning techniques that leverage aviation data collected from commercial airline operations to improve safety. Anomaly detection and predictive maintenance have been the main targets for machine learning applications. However, this paper focuses on the identification of precursors, which is a relatively newer application. Precursors are events correlated with adverse events that happen prior to the adverse event itself. Therefore, precursor mining provides many benefits including understanding the reasons behind a safety incident and the ability to identify signatures, which can be tracked throughout a flight to alert the operators of an potential upcoming adverse event. This work proposes using the multiple-instance learning (MIL) framework, a weakly supervised learning task, combined with a carefully designed Multi-Head Convolutional Neural Networks-Recurrent Neural Networks (MHCNN-RNN) architecture to predict different type of adverse events for any given flights and identify their precursors with little to no post-processing.Results obtained show that the MHCNN-RNN is able to accurately forecast high speed and high path angle events during the approach, and that it is also capable of determining the aircraft’s parameters that are correlated to these events. These parameters can be considered precursors to the events.

multiple instance learning↗
DOE OSTI2024

SCA Tools - SCRM Value Add or Lossy Noise Machines

Software supply chain risk management (SCRM) depends upon accurate information regarding the software components that comprise any given software system. The collection of components included in a software package can be organized within a software bill of materials, or SBOM. SBOMs are ideally generated when the software components are put together, such as at compile time, but for many reasons that has not and is not always possible. For example, legacy or proprietary software packages often do not have SBOMs available to downstream consumers of that software. It’s not just end users that are affected, manufacturers themselves also must deal with this problem. To answer these questions, the market has seen the rise of several commercial software composition analysis (SCA) tools. These tools aim to peer into completed software systems, automatically identifying hidden software dependencies and looking up known vulnerabilities associated with those dependencies to enable end-users to enhance their cyber supply chain risk management processes. These tools are potentially a huge boon to end users of legacy and proprietary software – and a potential bane, depending on how accurate they are. This research asks that question – how accurate are currently available binary SCA tools – and provides answers to several other questions: What does it mean to be “accurate”? What limitations do the tools have in identifying common edge cases that take place in modern software development? Can they help you avoid a devastating supply chain attack, or is it all just noise? After researching SCA tools on the market, we identified three vendors that fit our use case and would provide analysis on compiled binaries. Using these tools, we submitted firmware for critical infrastructure devices for analysis and SBOM generation. The SBOM outputs were then cross referenced with SBOMs generated through manual analysis for comparison. In addition to the firmware samples, we also submitted edge case samples based off a popular open-source library that were specifically crafted to evaluate each tools’ ability to accurately identify components. These samples were customized to be consistent with modifications we have seen in modern software development as well as a couple that are representative of supply chain attacks.

97 MATHEMATICS AND COMPUTING↗
NASA NTRS1997

Team play with a powerful and independent agent: operational experiences and automation surprises on the Airbus A-320

Research and operational experience have shown that one of the major problems with pilot-automation interaction is a lack of mode awareness (i.e., the current and future status and behavior of the automation). As a result, pilots sometimes experience so-called automation surprises when the automation takes an unexpected action or fails to behave as anticipated. A lack of mode awareness and automation surprises can he viewed as symptoms of a mismatch between human and machine properties and capabilities. Changes in automation design can therefore he expected to affect the likelihood and nature of problems encountered by pilots. Previous studies have focused exclusively on early generation "glass cockpit" aircraft that were designed based on a similar automation philosophy. To find out whether similar difficulties with maintaining mode awareness are encountered on more advanced aircraft, a corpus of automation surprises was gathered from pilots of the Airbus A-320, an aircraft characterized by high levels of autonomy, authority, and complexity. To understand the underlying reasons for reported breakdowns in human-automation coordination, we also asked pilots about their monitoring strategies and their experiences with and attitude toward the unique design of flight controls on this aircraft.

NASA Discipline Space Human Factors↗
NASA NTRS2022

Using an Explainable Machine Learning Approach to Characterize Earth System Model Errors: Application of SHAP Analysis to Modeling Lightning Flash Occurrence

Computational models of the Earth System are critical tools for modern scientific inquiry. Effortstoward evaluating and improving errors in representations of physical and chemical processes inthese large computational systems are commonly stymied by highly nonlinear and complexerror behavior. Recent work has shown that these errors can be effectively predicted usingmodern Artificial Intelligence (A.I.) techniques. In this work, we go beyond these previousstudies to apply an interpretable A.I. technique to not only predict model errors but also movetoward understanding the underlying reasons for successful error prediction. We use XGBoostclassification trees and SHapley Additive exPlanations (SHAP) analysis to explore the errors inthe prediction of lightning occurrence in the NASA GEOS model, a widely used Earth SystemModel. This explainable error prediction system can effectively predict the model error andindicates that the errors are strongly related to convective processes and the characteristics ofthe land surface.

Artificial intelligence↗
NASA NTRS1987

Distributed intelligence for supervisory control

Supervisory control systems must deal with various types of intelligence distributed throughout the layers of control. Typical layers are real-time servo control, off-line planning and reasoning subsystems and finally, the human operator. Design methodologies must account for the fact that the majority of the intelligence will reside with the human operator. Hierarchical decompositions and feedback loops as conceptual building blocks that provide a common ground for man-machine interaction are discussed. Examples of types of parallelism and parallel implementation on several classes of computer architecture are also discussed.

Wolfe, W. J.↗
NASA NTRS2014

Analyzing Array Manipulating Programs by Program Transformation

We explore a transformational approach to the problem of verifying simple array-manipulating programs. Traditionally, verification of such programs requires intricate analysis machinery to reason with universally quantified statements about symbolic array segments, such as "every data item stored in the segment A[i] to A[j] is equal to the corresponding item stored in the segment B[i] to B[j]." We define a simple abstract machine which allows for set-valued variables and we show how to translate programs with array operations to array-free code for this machine. For the purpose of program analysis, the translated program remains faithful to the semantics of array manipulation. Based on our implementation in LLVM, we evaluate the approach with respect to its ability to extract useful invariants and the cost in terms of code size.

Program Transformation↗
NASA NTRSDate not supplied

SPHINX: An SEP Model Validation Infrastructure developed through Community Challenges and the SEP Scoreboards

Solar Energetic Particle (SEP) events are interesting from a scientific perspective as they are the product of a broad set of physical processes from the corona out through the extent of the heliosphere, and provide insight into processes of particle acceleration and transport that are widely applicable in astrophysics. From the operations perspective, SEP events pose a radiation hazard for aviation, electronics in space, and human space exploration, in particular for missions outside of the Earth’s protective magnetosphere including to the Moon and Mars (Whitman et al 2022). For these reasons, SEP modelers have developed a rich and diverse set of models with a wide variety of aims. Some models probe the basic physics at the heart of particle acceleration and transport. Others produce fast statistical forecasts or employ disruptive new techniques like Machine Learning with the goal to assist end users in making operational decisions. To enable a consistent and quantitative understanding of SEP model performance, a generalized, automated validation infrastructure, called SPHINX, is being developed at NASA SRAG in close collaboration with NASA CCMC, NASA M2M, NOAA SWPC, and BIRA-IASB. This infrastructure has been built up through a multi-year community challenge. Starting in 2018 at the SHINE workshop, an effort was launched through SHINE, ISWAT, and ESWW to encourage quantitative, comprehensive, and consistent validation of SEP models. This effort has defined a set of challenge SEP events with the aim of generating quantitative comparisons between forecasts and observations and a set of challenge “non-events” to assess false alarms. In 2023, these challenge lists have been extended to statistically significant numbers with a prescribed set of rules for producing forecasts and supported through the dedicated SEPVAL working meetings. The participation of the research community has allowed the infrastructure to validate all the types of outputs being produced by SEP models. In parallel, the SPHINX code is being applied to real time forecasts submitted to the SEP Scoreboards, ensuring that the validation infrastructure can interpret forecasts produced in an operational scenario and provide metrics meaningful for operations. Upon completion, SPHINX and its interactive user interface, SPHINX-Web, will be made available for public use.

space weather↗
DOE OSTI2022

Accelerating Materials Discovery: Artificial Intelligence for Sustainable, High-Performance Polymers

PolyID enables the discovery of polymers with advanced performance and greater sustainability while reducing material development timelines. The material design space is immense and cannot be reasonably probed using an Edisionian approach. High-throughput property prediction, enabled by artificial intelligence provides a hypothesis driven approach for down selection of candidate polymers to pursue experimentally. To aid experimentalists in the down selection of material targets this high-throughput, machine learning-based tool is capable of predicting polymer properties simply from molecular structures. Currently, transport, thermal, and mechanical properties across 7 polymer class (polyamides, polyesters, polycarbonates, polyimides, polyolefins, polyacrylates, and polyurethanes) can be predicted, and the PolyID platform has been flexibly designed so new materials and properties can be added.

artificial intelligence↗
DOE OSTI2024

Yet Another Discriminant Analysis (YADA): A Probabilistic Model for Machine Learning Applications

This paper presents a probabilistic model for various machine learning (ML) applications. While deep learning (DL) has produced state-of-the-art results in many domains, DL models are complex and over-parameterized, which leads to high uncertainty about what the model has learned, as well as its decision process. Further, DL models are not probabilistic, making reasoning about their output challenging. In contrast, the proposed model, referred to as Yet Another Discriminate Analysis(YADA), is less complex than other methods, is based on a mathematically rigorous foundation, and can be utilized for a wide variety of ML tasks including classification, explainability, and uncertainty quantification. YADA is thus competitive in most cases with many state-of-the-art DL models. Ideally, a probabilistic model would represent the full joint probability distribution of its features, but doing so is often computationally expensive and intractable. Hence, many probabilistic models assume that the features are either normally distributed, mutually independent, or both, which can severely limit their performance. YADA is an intermediate model that (1) captures the marginal distributions of each variable and the pairwise correlations between variables and (2) explicitly maps features to the space of multivariate Gaussian variables. Numerous mathematical properties of the YADA model can be derived, thereby improving the theoretic underpinnings of ML. Validation of the model can be statistically verified on new or held-out data using native properties of YADA. However, there are some engineering and practical challenges that we enumerate to make YADA more useful.

97 MATHEMATICS AND COMPUTING↗
NASA NTRS1988

Knowledge elicitation for an operator assistant system in process control tasks

A knowledge based system (KBS) methodology designed to study human machine interactions and levels of autonomy in allocation of process control tasks is presented. Users are provided with operation manuals to assist them in normal and abnormal situations. Unfortunately, operation manuals usually represent only the functioning logic of the system to be controlled. The user logic is often totally different. A method is focused on which illicits user logic to refine a KBS shell called an Operator Assistant (OA). If the OA is to help the user, it is necessary to know what level of autonomy gives the optimal performance of the overall man-machine system. For example, for diagnoses that must be carried out carefully by both the user and the OA, interactions are frequent, and processing is mostly sequential. Other diagnoses can be automated, in which the case the OA must be able to explain its reasoning in an appropriate level of detail. OA structure was used to design a working KBS called HORSES (Human Orbital Refueling System Expert System). Protocol analysis of pilots interacting with this system reveals that the a-priori analytical knowledge becomes more structured with training and the situation patterns more complex and dynamic. This approach can improve the a-priori understanding of human and automatic reasoning.

Boy, Guy A.↗
NASA NTRS1994

Program to develop a performance and heat load prediction system for multistage turbines

Flows in low-aspect ratio turbines, such as the SSME fuel turbine, are three dimensional and highly unsteady due to the relative motion of adjacent airfoil rows and the circumferential and spanwise gradients in total pressure and temperature, The systems used to design these machines, however, are based on the assumption that the flow is steady. The codes utilized in these design systems are calibrated against turbine rig and engine data through the use of empirical correlations and experience factors. For high aspect ratio turbines, these codes yield reasonably accurate estimates of flow and temperature distributions. However, future design trends will see lower aspect ratio (reduced number of parts) and higher inlet temperature which will result in increased three dimensionality and flow unsteadiness in turbines. Analysis of recently acquired data indicate that temperature streaks and secondary flows generated in combustors and up-stream airfoils can have a large impact on the time-averaged temperature and angle distributions in downstream airfoil rows.

Sharma, OM↗
NASA NTRS2017

Wind Tunnel Strain-Gage Balance Calibration Data Analysis Using a Weighted Least Squares Approach

A new approach is presented that uses a weighted least squares fit to analyze wind tunnel strain-gage balance calibration data. The weighted least squares fit is specifically designed to increase the influence of single-component loadings during the regression analysis. The weighted least squares fit also reduces the impact of calibration load schedule asymmetries on the predicted primary sensitivities of the balance gages. A weighting factor between zero and one is assigned to each calibration data point that depends on a simple count of its intentionally loaded load components or gages. The greater the number of a data point's intentionally loaded load components or gages is, the smaller its weighting factor becomes. The proposed approach is applicable to both the Iterative and Non-Iterative Methods that are used for the analysis of strain-gage balance calibration data in the aerospace testing community. The Iterative Method uses a reasonable estimate of the tare corrected load set as input for the determination of the weighting factors. The Non-Iterative Method, on the other hand, uses gage output differences relative to the natural zeros as input for the determination of the weighting factors. Machine calibration data of a six-component force balance is used to illustrate benefits of the proposed weighted least squares fit. In addition, a detailed derivation of the PRESS residuals associated with a weighted least squares fit is given in the appendices of the paper as this information could not be found in the literature. These PRESS residuals may be needed to evaluate the predictive capabilities of the final regression models that result from a weighted least squares fit of the balance calibration data.

calibration analysis↗
DOE OSTI2023

Embedding ceramic components in metal structures with hybrid directed energy deposition

The combined benefit of both additive and subtractive manufacturing within the same gantry system enables hybrid directed energy deposition to create complex geometries with smooth surface finish and superior dimensional accuracy. Moreover, with layer-by-layer access to the structure during both the addition and subtraction of material, the insertion of components is now possible, assuming the components can survive the high temperatures associated with the subsequent metal deposition. Ceramic inserts are of interest for a variety of reasons including (1) to create complex interwoven ductile/brittle composites for ballistics or high-temperature applications or (2) to integrate high-temperature strain or temperature sensors protected within ceramic substrate subsumed into a larger metal structure. Here, in this work, stainless steel substrates were machined to create an internal cavity for the insertion of a ceramic component. During the investigation of several different over-the-ceramic deposition strategies, components were inserted, and different process sequences were allowed to continue to envelop the inserted ceramic with varying success. Unmelted powder was used to serve both as a thermal buffer and to provide a flush surface upon which the laser cladding could continue. Subsequent depositions were attempted with both dry and wet powder (addition of machining coolant to wet). The wet powder has previously been demonstrated to not significantly impact the mechanical properties of a final structure and provided a thermal barrier to protect the ceramic piece from the extreme temperatures of the final metal deposition. The wetting of the powder provided stability and minimized displacement caused by the powder flow from the laser cladding head. Finally, the use of an oblique angle for laser cladding allowed for the redirection of some fraction of the introduced thermal energy away from the ceramic component and, consequently, improved the survival of the ceramic inserts. With this combination of techniques, ceramic inserts survived full embedding within a 3D-printed stainless steel structure.

36 MATERIALS SCIENCE↗
DOE OSTI2026

Deconstruction by C. thermocellum —from microbe mediated to dynamic redistribution of cellulosomes

Clostridium thermocellum is one of the most efficient microorganisms for the deconstruction of cellulosic biomass. To achieve this high level of cellulolytic activity, C. thermocellum uses large multienzyme complexes known as cellulosomes to break down complex polysaccharides, notably cellulose, found in plant cell walls. The attachment of bacterial cells to the nearby substrate via the cellulosome has been hypothesized to be the reason for this high efficiency. The region lying between the cell and the substrate has shown great variation and dynamics that are affected by the growth stage of cells and the substrate used for growth. Here, we used both super-resolution imaging and machine-learning approaches to study the distribution of C. thermocellum cellulosomes at different stages of growth. We show that C. thermocellum initially retains its cellulosomes primarily on the cell surface but then relocates large cellulosome clusters to the interface with biomass, therefore depleting its cell surface of cellulosomes. These results indicate dynamic redistribution of cellulosomes during growth, with a functional shift toward substrate-associated degradation later during growth on biomass.

09 BIOMASS FUELS↗
NASA NTRS1986

Limits to size for terrestrial telescopes - Two-stage optics

A wave front corrector inserted at the image of a surface producing an error can remove the wave front error over a reasonable field of view. The relationship of magnitude of error, constants of a given configuration, and the residual wave front error is explored. Results are presented of the production of a sample wave front error surface by diamond turning with the Large Optical Diamond Turning Machine (LODTM). The cost consequences of this new ability to upgrade a submillimeter-quality telescope to a high-acuity optical telescope are shown.

Meinel, Marjorie P.↗