Search NASA⌕ Search

SEARCH · Search NASA

Results for “domain knowledge”

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 37 records · Page 2
All sourcesNASA NTRSDOE OSTI
NASA NTRS1993

Autonomously acquiring declarative and procedural knowledge for ICAT systems

The construction of Intelligent Computer Aided Training (ICAT) systems is critically dependent on the ability to define and encode knowledge. This knowledge engineering effort can be broadly divided into two categories: domain knowledge and expert or task knowledge. Domain knowledge refers to the physical environment or system with which the expert interacts. Expert knowledge consists of the set of procedures and heuristics employed by the expert in performing their task. Both these areas are a significant bottleneck in the acquisition of knowledge for ICAT systems. This paper presents a research project in the area of autonomous knowledge acquisition using a passive observation concept. The system observes an expert and then generalizes the observations into production rules representing the domain expert's knowledge.

Kovarik, Vincent J., Jr.↗
NASA NTRS1992

Tradeoffs in the utility of learned knowledge

Planning systems which make use of domain theories can produce more accurate plans and achieve more goals as the quality of their domain knowledge improves. MTR, a multi-strategy learning system, was designed to learn from system failures and improve domain knowledge used in planning. However, augmented domain knowledge can decrease planning efficiency. We describe how improved knowledge that becomes expensive to use can be approximated to yield calculated tradeoffs in accuracy and efficiency.

Kedar, Smadar↗
DOE OSTI2021

Domain knowledge-informed, process-mapping AI graph for designing Fe-based alloys

<span style="font-family: Calibri, sans-serif; font-size: 12pt;">Continuous improvement in efficiency of a power plant relies on designing materials for use at increasingly higher temperature and/or pressure, for 100,000s hours of operation. Due to complexity, non-linearity and high-dimensionality of the problem, traditional Machine Learning (ML) approaches require unreasonably large datasets for the data-driven model development. Science-based material and process engineering complements hard data with, sometimes soft and intuitive, empirical domain knowledge. Artificial Intelligence (AI) was used in this study to incorporate such knowledge into computational graph architecture (process-mimicking artificial neuron design, causal layer and graph structures, ensemble modeling of latent states) and learning procedures (variable transformation, fuzzy physics pre-training and freezing of deep layers, virtual microstructure representation, and adversarial multi-objective optimization). The first alloys design pathways suggested by the AI tool (pyroMind) passed a preliminary engineering review on soundness and transparency.</span>

Romanov, Vyacheslav↗
DOE OSTI2018

Integrated computational materials engineering of corrosion resistant alloys

Structure, composition and surface properties dictate corrosion resistance in any given environment. The degrees of freedom in alloy design are too numerous in emerging materials such as high entropy alloys and bulk metallic glasses for the use of high-throughput methods or trial and error. We review three domains of knowledge that can be applied towards the goal of corrosion resistant alloy (CRA) design: (a) the aggregation of knowledge gained through experience in developing CRAs empirically, (b) data-driven approaches that use descriptive metrics for alloy composition optimization, and (c) first-principles models of elementary processes that regulate corrosion informed by theory and inspired by phenomenological models in the literature. A path forward for integrated computational materials engineering (ICME) of CRAs that unites these three knowledge domains is introduced.

36 MATERIALS SCIENCE↗
NASA NTRS1992

Domain and Specification Models for Software Engineering

This paper discusses our approach to representing application domain knowledge for specific software engineering tasks. Application domain knowledge is embodied in a domain model. Domain models are used to assist in the creation of specification models. Although many different specification models can be created from any particular domain model, each specification model is consistent and correct with respect to the domain model. One aspect of the system-hierarchical organization is described in detail.

Iscoe, Neil↗
NASA NTRS2004

Realtime Knowledge Management (RKM): From an International Space Station (ISS) Point of View

We are developing automated methods to provide realtime access to spacecraft domain knowledge relevant a spacecraft's current operational state. The method is based upon analyzing state-transition signatures in the telemetry stream. A key insight is that documentation relevant to a specific failure mode or operational state is related to the structure and function of spacecraft systems. This means that diagnostic dependency and state models can provide a roadmap for effective documentation navigation and presentation. Diagnostic models consume the telemetry and derive a high-level state description of the spacecraft. Each potential spacecraft state description is matched against the predictions of models that were developed from information found in the pages and sections in the relevant International Space Station (ISS) documentation and reference materials. By annotating each model fragment with the domain knowledge sources from which it was derived we can develop a system that automatically selects those documents representing the domain knowledge encapsulated by the models that compute the current spacecraft state. In this manner, when the spacecraft state changes, the relevant documentation context and presentation will also change.

Robinson, Peter I.↗
NASA NTRS1992

Modeling software systems by domains

The Software Architectures Engineering (SAE) Project at the Software Engineering Institute (SEI) has developed engineering modeling techniques that both reduce the complexity of software for domain-specific computer systems and result in systems that are easier to build and maintain. These techniques allow maximum freedom for system developers to apply their domain expertise to software. We have applied these techniques to several types of applications, including training simulators operating in real time, engineering simulators operating in non-real time, and real-time embedded computer systems. Our modeling techniques result in software that mirrors both the complexity of the application and the domain knowledge requirements. We submit that the proper measure of software complexity reflects neither the number of software component units nor the code count, but the locus of and amount of domain knowledge. As a result of using these techniques, domain knowledge is isolated by fields of engineering expertise and removed from the concern of the software engineer. In this paper, we will describe kinds of domain expertise, describe engineering by domains, and provide relevant examples of software developed for simulator applications using the techniques.

Dippolito, Richard↗
NASA NTRS1990

Artificial intelligence and design: Opportunities, research problems and directions

The issues of industrial productivity and economic competitiveness are of major significance in the U.S. at present. By advancing the science of design, and by creating a broad computer-based methodology for automating the design of artifacts and of industrial processes, we can attain dramatic improvements in productivity. It is our thesis that developments in computer science, especially in Artificial Intelligence (AI) and in related areas of advanced computing, provide us with a unique opportunity to push beyond the present level of computer aided automation technology and to attain substantial advances in the understanding and mechanization of design processes. To attain these goals, we need to build on top of the present state of AI, and to accelerate research and development in areas that are especially relevant to design problems of realistic complexity. We propose an approach to the special challenges in this area, which combines 'core work' in AI with the development of systems for handling significant design tasks. We discuss the general nature of design problems, the scientific issues involved in studying them with the help of AI approaches, and the methodological/technical issues that one must face in developing AI systems for handling advanced design tasks. Looking at basic work in AI from the perspective of design automation, we identify a number of research problems that need special attention. These include finding solution methods for handling multiple interacting goals, formation problems, problem decompositions, and redesign problems; choosing representations for design problems with emphasis on the concept of a design record; and developing approaches for the acquisition and structuring of domain knowledge with emphasis on finding useful approximations to domain theories. Progress in handling these research problems will have major impact both on our understanding of design processes and their automation, and also on several fundamental questions that are of intrinsic concern to AI. We present examples of current AI work on specific design tasks, and discuss new directions of research, both as extensions of current work and in the context of new design tasks where domain knowledge is either intractable or incomplete. The domains discussed include Digital Circuit Design, Mechanical Design of Rotational Transmissions, Design of Computer Architectures, Marine Design, Aircraft Design, and Design of Chemical Processes and Materials. Work in these domains is significant on technical grounds, and it is also important for economic and policy reasons.

Amarel, Saul↗
NASA NTRS1985

Perspectives on knowledge in engineering design

Various perspectives are given of the knowledge currently used in engineering design, specifically dealing with knowledge-based expert systems (KBES). Constructing an expert system often reveals inconsistencies in domain knowledge while formalizing it. The types of domain knowledge (facts, procedures, judgments, and control) differ from the classes of that knowledge (creative, innovative, and routine). The feasible tasks for expert systems can be determined based on these types and classes of knowledge. Interpretive tasks require reasoning about a task in light of the knowledge available, where generative tasks create potential solutions to be tested against constraints. Only after classifying the domain by type and level can the engineer select a knowledge-engineering tool for the domain being considered. The critical features to be weighed after classification are knowledge representation techniques, control strategies, interface requirements, compatibility with traditional systems, and economic considerations.

Rasdorf, W. J.↗
NASA NTRS2010

Systems, methods and apparatus for verification of knowledge-based systems

Systems, methods and apparatus are provided through which in some embodiments, domain knowledge is translated into a knowledge-based system. In some embodiments, a formal specification is derived from rules of a knowledge-based system, the formal specification is analyzed, and flaws in the formal specification are used to identify and correct errors in the domain knowledge, from which a knowledge-based system is translated.

Hinchey, Michael G.↗
NASA NTRS1990

A flight expert system for on-board fault monitoring and diagnosis

An architecture for a flight expert system (FLES) to assist pilots in monitoring, diagnosing, and recovering from inflight faults is described. A prototype was implemented and an attempt was made to automate the knowledge acquisition process by employing a learning by being told methodology. The scope of acquired knowledge ranges from domain knowledge, including the information about objects and their relationships, to the procedural knowledge associated with the functionality of the mechanisms. AKAS (automatic knowledge acquisition system) is the constructed prototype for demonstration proof of concept, in which the expert directly interfaces with the knowledge acquisition system to ultimately construct the knowledge base for the particular application. The expert talks directly to the system using a natural language restricted only by the extent of the definitions in an analyzer dictionary, i.e., the interface understands a subset of concepts related to a given domain. In this case, the domain is the electrical system of the Boeing 737. Efforts were made to define and employ heuristics as well as algorithmic rules to conceptualize data produced by normal and faulty jet engine behavior examples. These rules were employed in developing the machine learning system (MLS). The input to MLS is examples which contain data of normal and faulty engine behavior and which are obtained from an engine simulation program. MLS first transforms the data into discrete selectors. Partial descriptions formed by those selectors are then generalized or specialized to generate concept descriptions about faults. The concepts are represented in the form of characteristic and discriminant descriptions, which are stored in the knowledge base and are employed to diagnose faults. MLS was successfully tested on jet engine examples.

Ali, Moonis↗
DOE OSTI2017

pnnl/safekit

Some of the most pressing machine learning applications such as cyber security and object recognition lack enough ground-truth training data to build a classifier. We solve this problem by using unsupervised deep learning techniques to determine when data were anomalous or deviated from the norm. Traditional methods require either specific context dependent expertise to construct models, or prior examples of events and large amounts of well-labeled data to perform machine learning. Some cutting-edge deep learning approaches have been used to characterize complex events, but these have not been applied to the streaming environment and require a great deal of data. We demonstrated the use of an autoencoder and relevant featurization techniques to both learn a feature space and then identify anomalous portions without the aid of labeled data or domain knowledge. In this way, we demonstrated the detection of anomalous features without domain knowledge or tagged examples, and delivered these features to users.

Central, PNNL Developer↗
NASA NTRS2011

Executable Architecture Research at Old Dominion University

Executable Architectures allow the evaluation of system architectures not only regarding their static, but also their dynamic behavior. However, the systems engineering community do not agree on a common formal specification of executable architectures. To close this gap and identify necessary elements of an executable architecture, a modeling language, and a modeling formalism is topic of ongoing PhD research. In addition, systems are generally defined and applied in an operational context to provide capabilities and enable missions. To maximize the benefits of executable architectures, a second PhD effort introduces the idea of creating an executable context in addition to the executable architecture. The results move the validation of architectures from the current information domain into the knowledge domain and improve the reliability of such validation efforts. The paper presents research and results of both doctoral research efforts and puts them into a common context of state-of-the-art of systems engineering methods supporting more agility.

Tolk, Andreas↗
NASA NTRS1992

A second generation expert system for checking and diagnosing AXAF's electric power system

AXAF - Advanced X-ray Astrophysics Facility - is a third NASA's great space observatory. Each of these observatories is intended to cover different parts of the electromagnetic spectrum (x-ray for AXAF) and to provide high resolution images of celestial sources in our universe. While the spacecraft is in orbit, the electric power system (EPS) performance is monitored via sensors measuring voltages, currents, pressures, and temperatures. The sensor data are sent from the spacecraft to the ground station as telemetry and analyzed on arrival. Monitoring, diagnosis and maintenance of such EPS is an arduous task which requires expertise and constant attention of the ground personnel. To help the ground crew in this task, much of it should be automated and delegated to expert systems, which draw engineer's attention to possible malfunctions and allows him to review the telemetry to determine the source of the trouble, diagnose the suspected fault and to propose a corrective action. Those systems are built on assumptions such as: (1) domain knowledge is available and can be represented as a set of rules; (2) domain knowledge is circumscribed, static, and monotonic; and (3) expert decision making can be emulated by a logical inference mechanism.

Bykat, Alex↗
NASA NTRS1987

A flight expert system (FLES) for on-board fault monitoring and diagnosis

The increasing complexity of modern aircraft creates a need for a larger number of caution and warning devices. But more alerts require more memorization and higher workloads for the pilot and tend to induce a higher probability of errors. Therefore, an architecture for a flight expert system (FLES) is developed to assist pilots in monitoring, diagnosing and recovering from in-flight faults. A prototype of FLES has been implemented. A sensor simulation model was developed and employed to provide FLES with airplane status information during the diagnostic process. The simulator is based on the Lockheed Advanced Concept System (ACS), a future generation airplane, and on the Boeing 737. A distinction between two types of faults, maladjustments and malfunctions, has led to two approaches to fault diagnosis. These approaches are evident in two FLES subsystems: the flight phase monitor and the sensor interrupt handler. The specific problem addressed in these subsystems has been that of integrating information received from multiple sensors with domain knowledge in order to access abnormal situations during airplane flight. Malfunctions and maladjustments are handled separately, diagnosed using domain knowledge.

Ali, Moonis↗
DOE OSTI2021

Performance Results for Sensor Assignment Problem as Solved on a Multi-Node Cluster

An earlier report described a procedure for optimal sensor set selection and its implementation on a computational cluster. This new and innovative capability was developed to facilitate a reduction in operations staffing levels to improve plant economics. By automating surveillance and maintenance tasks through early detection of degrading sensors and equipment, staff can be more efficiently deployed. The method uses automated reasoning and domain knowledge in the form of the conservation equations to infer from plant measurements the state of equipment health. Inclusion of domain knowledge addresses the problem that exists with pure data-driven methods that there are no rigorous guidelines for determining what constitutes an adequate sensor set. Formalizing the procedure for sensor set selection as we have done results in a more reliable and explainable diagnosis of plant equipment health. Importantly, from the standpoint of the plant owner, personnel are provided with an early and explicit diagnosis of an equipment problem. That in principle automates the process and eliminates having to send personnel into the plant to find the cause as typically occurs when a data-driven method detects an anomaly. In this report we describe first results obtained using a computational cluster to solve the sensor set selection problem as framed above. The case described addresses the problem of equipment health monitoring in the high-pressure (HP) feedwater system of a pressurized light water reactor as seen through the eyes of our collaborating utility partner. Maintenance of this system can amount to millions of dollars per year if equipment health issues go undiagnosed and lead to loss of function. On examining the potential that is inherent in the installed sensor set for diagnosing equipment health degradation, it was found that greater fault resolution capability can be achieved using a sensor set that is 20 percent fewer in number. The take-away is that compared to the installed sensor set there exists a more strategic assignment of sensors that will furnish better health monitoring capability and with fewer sensors. Where the problem defies solution by manual inspection, as is the case here, one can be found by an algorithm. The solution was obtained in four hours using 30 computational cores. The HP feedwater problem as posed above illustrates the added value of approaching the sensor selection problem as one amenable to algorithmic solution. This problem is of interest to advanced reactor designers and to utilities that are setting up remote monitoring and diagnostic centers.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗
DOE OSTI2024

Empirically Categorizing the Built Environment in Relation to Height

Buildings are a core component of the urban environment and affect human populations, energy usage, city development, city planning, and urban heat islands. Buildings span an enormous range of sizes, from a 2m tall shelter to the Burj Khalifa; and at the same time there are widely recognized categories of similar buildings, with homes, office buildings, or skyscrapers as some examples. Currently, there is no consistent method to quantitatively determine how a building should be categorized by its height, or how many categories there should be within the built environment. Additionally, these categories vary spatially, leading to multiple definitions at local scales of what it means to be a tall, medium, or short building. Here, we find across 17.59 million buildings in the United States, Germany, and Japan, that applying a K-nearest neighbor approach to quantitatively bin the built environment outperforms the current state-of-the-art, subjective domain knowledge. This was evidenced as our method of leveraging a K-nearest neighbor improved upon the existing approach of using domain knowledge by 10% with respect to precision, recall, F1-score and accuracy. Our results showcase the finding that it is possible to generate a global and consistent approach to categorizing the built environment in relation to height. This is significant in that there is now a quantitative way to categorize the built environment based on building height at a global scale, allowing researchers a consistent platform for comparison and collaboration across various applications.

Stipek, Clinton↗
DOE OSTI2024

Chemical reaction enhanced graph learning for molecule representation

Abstract Motivation Molecular representation learning (MRL) models molecules with low-dimensional vectors to support biological and chemical applications. Current methods primarily rely on intrinsic molecular information to learn molecular representations, but they often overlook effectively integrating domain knowledge into MRL. Results In this article, we develop a reaction-enhanced graph learning (RXGL) framework for MRL, utilizing chemical reactions as domain knowledge. RXGL introduces dual graph learning modules to model molecule representation. One module employs graph convolutions on molecular graphs to capture molecule structures. The other module constructs a reaction-aware graph from chemical reactions and designs a novel graph attention network on this graph to integrate reaction-level relations into molecular modeling. To refine molecule representations, we design a reaction-based relation learning task, which considers the relations between the reactant and product sides in reactions. In addition, we introduce a cross-view contrastive task to strengthen the cooperative associations between molecular and reaction-aware graph learning. Experiment results show that our RXGL achieves strong performance in various downstream tasks, including product prediction, reaction classification, and molecular property prediction. Availability and implementation The code is publicly available at https://github.com/coder-ACAC/RLM.

Biochemistry & Molecular Biology↗