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 55 records · Page 3

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↗

Beyond rules: The next generation of expert systems

The PARAGON Representation, Management, and Manipulation system is introduced. The concepts of knowledge representation, knowledge management, and knowledge manipulation are combined in a comprehensive system for solving real world problems requiring high levels of expertise in a real time environment. In most applications the complexity of the problem and the representation used to describe the domain knowledge tend to obscure the information from which solutions are derived. This inhibits the acquisition of domain knowledge verification/validation, places severe constraints on the ability to extend and maintain a knowledge base while making generic problem solving strategies difficult to develop. A unique hybrid system was developed to overcome these traditional limitations.

Ferguson, Jay C.↗

Deep Learning-Based Failure Prognostic Model for PV Inverter Using Field Measurements

Here, this study presents a novel approach for the precise monitoring and prognosis of photovoltaic (PV) inverter status, which is crucial for the proactive maintenance of PV systems. It addresses the gaps in traditional model-based methods, which tend to neglect the overall reliability of inverters, and the limitations of data-driven approaches that largely depend on simulated data. This research presents a robust solution applicable to real-world scenarios. The proposed data-driven model for PV inverter failure prognosis employs actual inverter measurements, integrating various operational and weather-related factors based on domain knowledge. This approach effectively represents inverter stressors and operational status. Utilizing an Enhanced Siamese Convolutional Neural Network (ESCNN), the model merges operational data with domain knowledge features, redefining the prognosis challenge as a classification task. Furthermore, the paper discusses an ESCNN-based real-time inverter failure monitoring method developed on the well-trained model. The proposed models are rigorously trained and tested with real inverter data and a novel filtering method is included to address accidental failures in practical scenarios. The results validate the model's efficacy, and the directions for future research are also outlined.

42 ENGINEERING↗

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 work loads for the pilot and tend to induce a higher probability of errors. Therefore, an architecture for a flight expert system (FLES) to assist pilots in monitoring, diagnosing and recovering from in-flight faults has been developed. A prototype of FLES has been implemented. A sensor simulation model was developed and employed to provide FLES with the airplane status information during the diagnostic process. The simulator is based partly on the Lockheed Advanced Concept System (ACS), a future generation airplane, and partly on the Boeing 737, an existing airplane. A distinction between two types of faults, maladjustments and malfunctions, has led us to take 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 assess abnormal situations during airplane flight. This paper describes the reasons for handling malfunctions and maladjustments separately and the use of domain knowledge in the diagnosis of each.

Ali, M.↗

On the prediction of critical heat flux using a physics-informed machine learning-aided framework

The critical heat flux (CHF) corresponding to the departure from nucleate boiling (DNB) crisis is essential to the design and safety of a two-phase flow boiling system. Despite the abundance of predictive tools available to the thermal engineering community, the path for an accurate, robust CHF model remains elusive due to lack of consensus on the DNB triggering mechanism. This work aims to apply a physics-informed machine learning (ML)-aided hybrid framework to achieve superior predictive capabilities. Such a hybrid approach takes advantage of existing understanding in the field of interest (i.e., domain knowledge) and uses ML to capture undiscovered information from the mismatch between the actual and domain knowledge-predicted target. A detailed case study is carried out with an extensive DNB-specific CHF database to demonstrate (1) the improved performance of the hybrid approach as compared to traditional domain knowledge-based models, and (2) the hybrid model’s superior generalization capabilities over standalone ML methods across a wide range of flow conditions. The hybrid framework could also readily extend its applicability domain and complexity on the fly, showing an elevated level of flexibility and robustness. Based on the case study conclusions, the window-type extrapolation mapping methodology is further proposed to better inform high-cost experimental work.

42 ENGINEERING↗

Spatio-Temporal Denoising Graph Autoencoders with Data Augmentation for Missing Photovoltaic Data Imputation

The integration of the global Photovoltaic (PV) market with real time data-loggers has enabled large scale PV data analytical pipelines for power forecasting and long-term reliability assessment of PV fleets. Nevertheless, the performance of PV data analysis heavily depends on the quality of PV timeseries data. This paper proposes a novel Spatio-Temporal Denoising Graph Autoencoder (STD-GAE) framework to impute missing PV Power Data. STDGAE exploits temporal correlation, spatial coherence, and value dependencies from domain knowledge to recover missing data. It is empowered by two modules. (1) To cope with sparse yet various scenarios of missing data, STD-GAE incorporates a domain-knowledge aware data augmentation module that creates plausible variations of missing data patterns. This generalizes STD-GAE to robust imputation over different seasons and environment. (2) STD-GAE nontrivially integrates spatiotemporal graph convolution layers (to recover local missing data by observed “neighboring” PV plants) and denoising autoencoder (to recover corrupted data from augmented counterpart) to improve the accuracy of imputation accuracy at PV fleet level. We have evaluated our proposed model on two realworld PV datasets. Experimental results show that STD-GAE can achieve a gain of 43.14% in imputation accuracy and remains less sensitive to missing rate, different seasons, and missing scenarios, compared with state-of-the-art data imputation methods such as MIDA and LRTC-TNN.

Fan, Yangxin↗

Analytic Steering: Inserting Context into the Information Dialog

An analyst’s intrinsic domain knowledge is a primary asset in almost any analysis task. Unstructured text analysis systems that apply un-supervised content analysis approaches can be more effective if they can leverage this domain knowledge in a manner that augments the information discovery process without obfuscating new or unexpected content. Current unsupervised approaches rely upon the prowess of the analyst to submit the right queries or observe generalized document and term relationships from ranked or visual results. We propose a new approach which allows the user to control or steer the analytic view within the unsupervised space. This process is controlled through the data characterization process via user supplied context in the form of a collection of key terms. We show that steering with an appropriate choice of key terms can provide better relevance to the analytic domain and still enable the analyst to uncover un-expected relationships; this paper discusses cases where various analytic steering approaches can provide enhanced analysis results and cases where analytic steering can have a negative impact on the analysis process.

Steering↗

IL-Net: Using Expert Knowledge to Guide the Design of Furcated Neural Networks

Deep neural networks (DNN) excel at extracting patterns. Through representation learning and automated feature engineering on large datasets, such models have been highly successful in computer vision and natural language applications. Designing optimal network architectures from a principled or rational approach however has been less than successful, with the best successful approaches utilizing an additional machine learning algorithm to tune the network hyperparameters. However, in many technical fields, there exist established domain knowledge and understanding about the subject matter. In this work, we develop a novel furcated neural network architecture that utilizes domain knowledge as high-level design principles of the network. We demonstrate proof-of-concept by developing IL-Net, a furcated network for predicting the properties of ionic liquids, which is a class of complex multi-chemicals entities. Compared to existing state-of-the-art approaches, we show that furcated networks can improve model accuracy by approximately 20-35%, without using additional labeled data. Lastly, we distill two key design principles for furcated networks that can be adapted to other domains.

Sakloth, Khushmeen↗

Interactive image quantification tools in nuclear material forensics

Morphological and microstructural features visible in microscopy images of nuclear materials can give information about the processing history of a nuclear material. Extraction of these attributes currently requires a subject matter expert in both microscopy and nuclear material production processes, and is a time consuming, and at least partially manual task, often involving multiple software applications. One of the primary goals of computer vision is to find ways to extract and encode domain knowledge associated with imagery so that parts of this process can be automated. In this paper we describe a user-in-the-loop approach to the problem which attempts to both improve the efficiency of domain experts during image quantification as well as capture their domain knowledge over time. This is accomplished through a sophisticated user-monitoring system that accumulates user-computer interactions as users exploit their imagery. We provide a detailed discussion of the interactive feature extraction and segmentation tools we have developed and describe our initial results in exploiting the recorded user-computer interactions to improve user productivity over time.

36 MATERIALS SCIENCE↗

9Cr steel visualization and predictive modeling

The goals of this work were to develop tools to visualize and characterize materials information, to explore new alloy compositions and/or processing, to better predict tensile strength and other mechanical properties. The 9–12 wt% Cr (9Cr) steel test data were compiled from several sources. About 3000 records with 30 predictor variables represent 82 unique steel alloy compositions, variations of thermo-mechanical processing steps and temperatures (homogenization, normalization and tempering cycles), test conditions and outcomes. Detailed data processing steps such as visualization and exploratory analysis, including univariate and bivariate analysis, different clustering techniques used to segment the data, statistical post-hoc analysis to verify significance of the findings, feature engineering to identify predictors of importance, and predictive modeling with cross-validation were performed. The outcome of analysis at each step was reviewed in the context of the domain knowledge of this class of steel, to see if there were underlying physical mechanisms that explain statistical relationships. Data analytics techniques and their parameters were fine-tuned to facilitate interpretation of the results as aligned with the insights from domain experts. Here, the ensemble predictive modeling using Random Forest regressor and post-hoc means comparison corroborated domain knowledge on the role of Co which is known to increase strength of steel alloys through solid-solution strengthening and to affect diffusion of alloying elements and precipitates. Alloys with Co content of 0.7–8 wt% had significantly higher mean strength than the ones without (while having Cr locked within 10–11 wt% range in either group and keeping other compositional element ratios fixed). End products of the computational techniques exploration are presented as the tools that can be used in iterative workflow of materials development and testing.

36 MATERIALS SCIENCE↗

Incorporating historical data and past analyses for improved tensile property prediction of 9% Cr steel

Data-driven analytical clustering and visualization techniques were applied to the dataset of 9% Cr experimental alloy data generated through the eXtremeMAT project. Techniques and results were compared with the resulting clusters obtained through similar analytical techniques on previous and reduced versions of the dataset. The principal components were generated in order to reduce the dimensionality of the complex dataset and to visualize the underlying trends in the data. Partitioning around medoids was performed on the resulting principal components to determine relevant clusters. Domain knowledge labels were further applied to the principal components to compare the labels with the trends identified through the clustering methods. The resulting clusters show the addition of the new data into the dataset, and show where alignment between the domain knowledge labels and partitioning around medoids (PAM) clusters show where the labels are appropriate. The clusters can be used to compare the tensile properties of the alloys, and to reduce the variation in the dataset.

Alloy design, Mechanical Properties, data analysis↗

Deep-freeze graph training for latent learning

Scientific and engineering advances are primarily driven by multi-tier conceptual constructs and conditional theoretical frameworks. The theories allow predictions of hypothetical system responses, given a set of approximate conditions (ranges of applicability) imposed on latent parameters that cannot be measured directly. Learning to estimate the latent variables (Latent Learning) helps to pinpoint the anticipated range-edge anomalies and improves the confidence in interpretation, interpolation and extrapolation of limited experimental data. Due to high dimensionality and extreme non-linearity of the materials science problems, very large datasets are typically required for conventional data-driven model development. The vital experimental data collection, particularly on microstructural phases, is very challenging, which makes it difficult to compile a high-quality database. Incorporation of the domain knowledge into the computational graph structure, initialization and optimization processes presents a viable mechanism for developing accurate models, with limited datasets. Furthermore, this study successfully utilized the approach to build the Deep Freeze Graph (DeepFreG) by mapping known causality relationships and by digitizing empirical domain knowledge for Latent Learning (LL), with specific applications in materials science.

36 MATERIALS SCIENCE↗

Real-time Event Detection Using Rank Signatures of Real-world PMU Data

Timely detection of power system events is a crucial task, which can facilitate the implementation of remedial actions to improve reliability, resiliency, and security of the system. Meanwhile, the widespread deployment of phasor measurement units (PMUs) makes it possible to develop data-driven event detection techniques. However, relying purely on data without incorporating domain knowledge for the event detection task in power systems poses substantial security and stability risks due to issues associated with data misinterpretation and model accuracy. In this regard, we propose a real-time event detection method using real-world PMU data by incorporating domain knowledge to adequately capture the event signatures. Specifically, we track the change in rank signatures of PMU data to accurately localize the events. To optimize the detection process, we incorporate an offline Bayesian optimization algorithm to tune the parameters by efficiently searching for the best values. The experiments using the real-world PMU dataset from a U.S. interconnection show that the proposed event detection approach can efficiently detect the events from PMU data streams with high accuracy.

Ghasemkhani, Amir↗

Machine learning–assisted prediction of heat fluxes through thermally anisotropic building envelopes

Thermally anisotropic building envelope (TABE) is a novel active building envelope that can save energy use to maintain thermal comfort in buildings by redirecting heat and coolness from building envelopes to thermal loops. Finite element models (FEMs) can be used to compute the heat fluxes through TABEs, but the high computational cost of finite element simulations has prevented parametric studies and design optimizations. This paper proposes a domain knowledge–informed, finite element–based machine learning framework to reduce the computation cost for the energy management of buildings installed with TABE that uses a ground thermal loop. First, the training heat flux data set was generated by FEM simulations with different thermal loop schedules. Then, both shallow learning models (i.e., multivariate linear regression and eXtreme Gradient Boost, or XGBoost) and a deep learning model (i.e., deep neural network, or DNN) were trained to predict the heat fluxes. Domain knowledge was used for data preprocessing and feature selection. Finally, the suitability of the selected machine learning model was tested under different thermal loop schedules. Herein, the case study results showed that: (1) XGBoost can be as accurate as DNN (coefficient of determination equal to 0.81) with much less training time; (2) the annual energy cost savings for different thermal loop schedules obtained by the XGBoost-predicted and FEM-calculated heat fluxes are consistent, having a difference of only 4%; and (3) XGBoost can reduce the computation time for the annual energy analysis of the case study building with a given thermal loop schedule from around 12 h by using FEM to less than 1 min.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

Descriptor Aided Bayesian Optimization for Many-Level Qualitative Variables With Materials Design Applications

Abstract Engineering design often involves qualitative and quantitative design variables, which requires systematic methods for the exploration of these mixed-variable design spaces. Expensive simulation techniques, such as those required to evaluate optimization objectives in materials design applications, constitute the main portion of the cost of the design process and underline the need for efficient search strategies—Bayesian optimization (BO) being one of the most widely adopted. Although recent developments in mixed-variable Bayesian optimization have shown promise, the effects of dimensionality of qualitative variables have not been well studied. High-dimensional qualitative variables, i.e., with many levels, impose a large design cost as they typically require a larger dataset to quantify the effect of each level on the optimization objective. We address this challenge by leveraging domain knowledge about underlying physical descriptors, which embody the physics of the underlying physical phenomena, to infer the effect of unobserved levels that have not been sampled yet. We show that physical descriptors can be intuitively embedded into the latent variable Gaussian process approach—a mixed-variable GP modeling technique—and used to selectively explore levels of qualitative variables in the Bayesian optimization framework. This physics-informed approach is particularly useful when one or more qualitative variables are high dimensional (many-level) and the modeling dataset is small, containing observations for only a subset of levels. Through a combination of mathematical test functions and materials design applications, our method is shown to be robust to certain types of incomplete domain knowledge and significantly reduces the design cost for problems with high-dimensional qualitative variables.

Engineering↗

Data Model Management for Space Information Systems

The Reference Architecture for Space Information Management (RASIM) suggests the separation of the data model from software components to promote the development of flexible information management systems. RASIM allows the data model to evolve independently from the software components and results in a robust implementation that remains viable as the domain changes. However, the development and management of data models within RASIM are difficult and time consuming tasks involving the choice of a notation, the capture of the model, its validation for consistency, and the export of the model for implementation. Current limitations to this approach include the lack of ability to capture comprehensive domain knowledge, the loss of significant modeling information during implementation, the lack of model visualization and documentation capabilities, and exports being limited to one or two schema types. The advent of the Semantic Web and its demand for sophisticated data models has addressed this situation by providing a new level of data model management in the form of ontology tools. In this paper we describe the use of a representative ontology tool to capture and manage a data model for a space information system. The resulting ontology is implementation independent. Novel on-line visualization and documentation capabilities are available automatically, and the ability to export to various schemas can be added through tool plug-ins. In addition, the ingestion of data instances into the ontology allows validation of the ontology and results in a domain knowledge base. Semantic browsers are easily configured for the knowledge base. For example the export of the knowledge base to RDF/XML and RDFS/XML and the use of open source metadata browsers provide ready-made user interfaces that support both text- and facet-based search. This paper will present the Planetary Data System (PDS) data model as a use case and describe the import of the data model into an ontology tool. We will also describe the current effort to provide interoperability with the European Space Agency (ESA)/Planetary Science Archive (PSA) which is critically dependent on a common data model.

RDF↗

Leveraging transfer learning and leaf spectroscopy for leaf trait prediction with broad spatial, species, and temporal applicability

Accurate and reliable prediction of leaf traits is crucial for understanding plant adaptations to environmental variation, monitoring terrestrial ecosystems, and enhancing comprehension of functional diversity and ecosystem functioning. Currently, various approaches (e.g., statistical, physical models) have been developed to estimate leaf traits through hyperspectral remote sensing and leaf spectroscopy. However, the absence of high-performing, transferable, and stable models across various domains of space, plant functional types (PFTs) and seasons hinder our ability to quantify and comprehend spatiotemporal variations in leaf traits. This study proposes robust and highly transferable models for better predicting leaf traits with hyperspectral reflectance. Initially, three datasets were assembled, pairing common leaf traits — chlorophyll (Chla+b), carotenoids (Ccar), leaf mass per area (LAM), equivalent water thickness (EWT) — with leaf spectra measurements collected across diverse geographic locations in the U.S. and Europe, PFTs, and seasons. Measurements were acquired using spectroradiometers (e.g., ASD FieldSpec 3/4/Pro and SVC HR-1024i) with integrating spheres, leaf clips, and contact probes. Here, we then developed transfer learning-based hybrid models that incorporated the domain knowledge of radiative transfer models (RTMs) through pretraining processes and were well-constrained by fine-tuning with field measurements. Through comparison with other state-of-the-art statistical models, including partial-least squares regression (PLSR) and Gaussian Process Regression (GPR), as well as pure physical models, we found that the proposed transfer learning models achieved better predictive performance and higher transferability. Specifically, compared to other statistical models and pure RTMs, the transfer learning model exhibited higher coefficient of determination (R 2 ) values with range of 0.01 to 0.79, lower normalized root mean square error (NRMSE) with range of 0.06 % to 33.25 % in model performance. Additionally, the models exhibited improved transferability, with higher R 2 values range from 0.04 to 0.32, lower NRMSE range from 0.08 % to 30.81 %. The findings underscore that transfer learning models through integrating domain knowledge from RTMs and limited observations, can harness the advantages of both RTMs and statistical models and serve as a promising approach for effectively predicting leaf traits.

59 BASIC BIOLOGICAL SCIENCES↗

Differentiable vertex fitting for jet flavor tagging

This work explores the use of differentiable programming to integrate domain knowledge, in the form of domain specific software, into neural networks to develop scientific machine learning systems. We propose a differentiable vertex fitting algorithm that estimates the crossing point of multiple curves. In the high energy physics setting, these curves are defined by particle equations of motion and the crossing point represents the origin of particle production. This differentiable vertex fitting algorithm can be seamlessly integrated into neural networks, and we show its utility and efficacy in the high energy physics application of the classification of jets, i.e., collimated streams of particles in particle detectors whose originating parent particle we aim to classify. We demonstrate how differentiable vertex fitting can be integrated into larger transformer-based models for jet flavor tagging and show improvements in heavy flavor jet classification when compared to baseline models. Published by the American Physical Society 2024

Smith, Rachel E. C. (ORCID:0000000335851262)↗