Search NASASearch

SEARCH · Search NASA

Results for “Interpretability”

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

AI for Interpreting Nuclear Power Plant Documents for Power Uprates

To reduce the cost and time needed for regulatory compliance, nuclear power plants (NPPs) can utilize artificial intelligence (AI) to assist in interpreting complex and voluminous documents that typically span thousands of pages. Usually, the process of interpreting a plant’s technical specifications (TSs) and associated documents is labor intensive. This study aims to understand what processes state-of-the-art large language models (LLMs) can automate and to identify the pitfalls associated with using LLMs to reduce human labor costs and time. This research uses a recent AI technology called retrieval augmented generation (RAG), which retrieves pages of information from TSs and associated documents to assist with NPP power uprates (cleared to produce more power). LLMs are integral to RAG because they create human-like responses based on the retrieved information, aiding in the interpretation and application processes. A baseline case demonstrates how LLMs can operate successfully for a power uprate application. Then five use cases show five types of potential failures: (1) RAG retrieving the incorrect information, (2) RAG misinterpreting the retrieved information, (3) RAG relying on knowledge not contained in the retrieved information, (4) RAG hallucinating, and (5) RAG refusing to answer. The results of the five use cases suggest that automating the human interpretation of TSs and associated documents with AI should be approached with caution. A subject-matter expert reviewed the AI outputs from the five use cases and concluded that an LLM can produce technical information that is needed to produce power uprate applications in certain instances.

21 - SPECIFIC NUCLEAR REACTORS AND ASSOCIATED PLAN

Interpretable Machine Learning for Characterizing Electric Vehicle Charging Behavior: Insights from Real-World Data

As electric vehicle (EV) adoption rises globally, concerns about the impact on aging electrical grids grow, particularly regarding the charging behavior of EV drivers. This study analyzes real-world driving and charging data from Ford battery electric vehicles (BEVs) collected between 2018 and 2019 to develop interpretable models that characterize charging behavior and quantify influencing factors. Prior research has relied on assumptions regarding driver behavior, often overlooking actual charging patterns. By employing generalized linear mixed models (GLMMs), this work offers insights into how various elements, such as next trip distance and state of charge (SOC), influence charging decisions. The dataset comprises over three million park-trip pairs from 1,997 vehicles, revealing that features related to driving behavior significantly dictate charging behavior, while infrastructure and regional factors have lesser impacts. The findings suggest that existing simulation models may oversimplify EV charging behavior assumptions. This work utilizes real-world EV driving and charging data to train interpretable models that describe charging behavior and quantify the factors most associated with how drivers use charging infrastructure. This research underscores the need for interpretable, data-driven methodologies to inform future EV infrastructure planning and grid management.

29 - ENERGY PLANNING, POLICY AND ECONOMY

Teaching Freight Mode Choice Models New Tricks Using Interpretable Machine Learning Methods

Understanding and forecasting the intricate freight mode choice behavior under various industry, policy, and technology contexts is essential in freight planning and policymaking. Numerous models have been developed in prior studies to provide insights into freight mode selection, the majority of which use discrete choice models such as multinomial logit (MNL) models. However, logit models often rely on linear specifications of independent variables, despite potential nonlinear relationships in the data. Moreover, there often lacks a heuristic and efficient approach to identify such complex relationships to define the logit model specifications. To fill this gap, we developed an MNL model for freight mode choice using the insights from state-of-the- art machine learning (ML) models. ML models can capture the nonlinear nature of the complex decision-making process, and recent advances in 'explainable AI' have greatly improved their interpretability. The interpretable ML methods help enhance the performance of MNL models and advance knowledge of freight mode choice. Specifically, the influential factors and their relationship with individual modes are identified using SHapley Additive exPlanations (SHAP) to improve the MNL's performance. The workflow is demonstrated in a case study of Austin, Texas, and the SHAP results reveal multiple nonlinear relationships predicted by ML models. Incorporating those relationships into MNL model specifications improves the interpretability and accuracy of the MNL model compared to a conventional MNL model. Findings from this study can be used to guide freight planning and inform policymakers and practitioners on how key factors affect freight decision-making.

ADVANCED PROPULSION SYSTEMS,MATHEMATICS AND COMPUT

L-VISP: LSTM Visualization for Interpretable Symptom Prediction in Patient Cohorts

Symptom modelling in head and neck cancer is challenged by the complexity of heterogeneous patient data, leading to an interest in deep learning approaches. Although Long Short-Term Memory Networks (LSTMs) have shown great results in patient risk prediction, their low interpretability requires data modellers to collaborate with clinical experts to validate the results. We present L-VISP, a human–machine solution that uses visual analytics for LSTM modelling in clinical research. L-VISP uses custom visual encodings to make multiple LSTM variants interpretable, supporting a full range of analysis, from understanding model operations and evaluating performance to interpreting results in a clinical context. We evaluate L-VISP with data modellers and a clinical oncologist and present the takeaways from this multidisciplinary collaboration.

LSTM modeling

Architectural design of an Algol interpreter

The design of a syntax-directed interpreter for a subset of Algol is described. It is a conceptual design with sufficient details and completeness but as much independence of implementation as possible. The design includes a detailed description of a scanner, an analyzer described in the Floyd-Evans productions, a hash-coded symbol table, and an executor. Interpretation of sample programs is also provided to show how the interpreter functions.

Jackson, C. K.

A technique for interpretation of multispectral remote sensor data

The author has identified the following significant results. The U.S. Army Engineer Waterways Experiment Station is engaged in a study to detect from ERTS-1 satellite data alterations to the absorption and scattering properties caused by movement of suspended particles and solutes in selected areas of the Chesapeake Bay and to correlate the data to determine the feasibility of delineating flow patterns, flushing action of the estuary, and sediment and pollutant dispersion. As a part of this study, ADP techniques have been developed that permit automatic interpretation of data from any multispectral remote sensor with computer systems which have limited memory capacity and computing speed. The multispectral remote sensor is considered as a reflectance spectrophotometer. The data which define the spectral reflectance characteristics of a scene are scanned pixel by pixel. Each pixel whose spectral reflectance matches a reference spectrum is identified, and the results are shown in a map that identifies the locations where spectrum matches were detected and spectrum that was matched. The interpretation technique is described and an example of interpreted data from ERTS-1 is presented.

Williamson, A. N.

Ground Operations Aerospace Language (GOAL). Volume 4: Interpretive code translator

This specification identifies and describes the principal functions and elements of the Interpretive Code Translator which has been developed for use with the GOAL Compiler. This translator enables the user to convert a compliled GOAL program to a highly general binary format which is designed to enable interpretive execution. The translator program provides user controls which are designed to enable the selection of various output types and formats. These controls provide a means for accommodating many of the implementation options which are discussed in the Interpretive Code Guideline document. The technical design approach is given. The relationship between the translator and the GOAL compiler is explained and the principal functions performed by the Translator are described. Specific constraints regarding the use of the Translator are discussed. The control options are described. These options enable the user to select outputs to be generated by the translator and to control vrious aspects of the translation processing.

Source record

Location of geologic structures from interpretation of ERTS-1 imagery, Carbon County, Wyoming

The author has identified the following significant results. Possible geologic structures in the basin sediments of Carbon County and vicinity were located by interpretation of ERTS-1 imagery. These same structures are not evident on existing conventional geologic maps of the area. Subsequent field checks confirmed much of the geologic interpretation, but revealed that two apparent closed structures identified on the ERTS-1 imagery were actually topographic pseudostructures in flat or homoclinal sediments. Stereoscopic coverage (where available) allows the interpreter to avoid such misinterpretations.

Marrs, R. W.

Interpretation of ERTS-1 imagery aided by photographic enhancement

Agfacontour film can be used to produce relatively economical enhancements of density differences recorded on film. Image interpreters often ask whether enhancement really increases the information content of the original images. Examples of ERTS-1 imagery enhancements show that it is possible to separate density levels which are not discernible by the human eye. It is stressed that these techniques do not replace the photo interpreter, but rather they aid him in the interpretation process. Subtle density variations are made to stand out and on objective classification of the densities forming the image is produced.

Nielsen, U.

Interdisciplinary application and interpretation of EREP data within the Susquehanna River Basin

The author has identified the following significant results. It has become that lineaments seen on Skylab and ERTS images are not equally well defined, and that the clarity of definition of a particular lineament is recorded somewhat differently by different interpreters. In an effort to determine the extent of these variations, a semi-quantitative classification scheme was devised. In the field, along the crest of Bald Eagle Mountain in central Pennsylvania, statistical techniques borrowed from sedimentary petrography (point counting) were used to determine the existence and location of intensely fractured float rock. Verification of Skylab and ERTS detected lineaments on aerial photography at different scales indicated that the brecciated zones appear to occur at one margin of the 1 km zone of brecciation defined as a lineament. In the Lock Haven area, comparison of the film types from the SL4 S190A sensor revealed the black and white Pan X photography to be superior in quality for general interpretation to the black and white IR film. Also, the color positive film is better for interpretation than the color IR film.

Mcmurtry, G. J.

Interpretation of an urban scene using multi-channel radar imagery

Four channel, SLAR imagery was studied by a group of individuals having no previous experience with either SLAR imagery or the urban area under scrutiny. This tactic was used because it was desired to define the nature of training needed when introducing people to radar imagery of urban scenes. Responses resulting from interpretations based on standard photointerpretation methods were subjected to a Chi-square analysis to determine the level of significance of the interpretations. For the urban scene studied, and for the two wavelengths (X /3.0 cm/ and L /23.0 cm/ Band) and polarizations (HH and HV) used, several types of urban land use were easily and accurately identified. It is shown that little formal training is required for obtaining quite high interpretation accuracies from multi-channel radar images of some urban scenes.

Bryan, M. L.

Rotational interpretation of recombination line widths in the galactic nucleus

The widths of the radio recombination lines observed in the direction of the galactic center are interpreted as due to pure rotation. This interpretation leads to a mass estimate for the inner 1.6-pc region of our Galaxy. It is noted that recent observations of M 31 suggest the existence of a distinct dynamic entity at its center - the nucleus. The rotational interpretation of the recombination line widths is shown to be consistent with the existence of a nucleus at the center of our own Galaxy which is similar in nature to the nucleus of M 31.

Kellman, S. A.

Basis for interpretation regarding the ages of the Serenitatis, Imbrium and Orientale events

One of the most important objectives of lunar study is to relate the lunar sample data to important lunar events. This paper utilizes as the basis of interpretation consideration of the following: (1) photogeologic data, (2) the choice of a cratering model, (3) estimates of temperature of impact ejecta and shock-induced heating, (4) petrologic data of lunar breccias and their thermal and shock history, and (5) meaningful age measurements. Both the author's interpretations and alternative views are discussed. The age of the Serenitatis event is not yet known. The interpreted age of the Imbrium event is between 3.90 and 3.84 Ga. The age of the Orientale event is 3.84 Ga.

Chao, E. C. T.

Method of interpretation of remotely sensed data and applications to land use

Instructional material describing a methodology of remote sensing data interpretation and examples of applicatons to land use survey are presented. The image interpretation elements are discussed for different types of sensor systems: aerial photographs, radar, and MSS/LANDSAT. Visual and automatic LANDSAT image interpretation is emphasized.

Parada, N. D. J.

Interpretation of remotely sensed data and its applications in oceanography

The methodology of interpretation of remote sensing data and its oceanographic applications are described. The elements of image interpretation for different types of sensors are discussed. The sensors utilized are the multispectral scanner of LANDSAT, and the thermal infrared of NOAA and geostationary satellites. Visual and automatic data interpretation in studies of pollution, the Brazil current system, and upwelling along the southeastern Brazilian coast are compared.

Parada, N. D. J.

Interpreting the solar wind ionization state

The ionization state of the solar coronal expansion is frozen within a few solar radii of the solar photosphere, and spacecraft measurements of the solar wind heavy ion charge state can therefore yield information about coronal conditions (e.g., electron temperature). Previous interpretations of the frozen-in ionization state have always assumed that in the coronal freezing-in region, (1) all heavy ions flow at the same bulk speed as protons, (2) the electron velocity distribution function is Maxwellian, and (3) conditions vary in space but not in time. The consequences of relaxing these assumptions for the interpretation of solar wind charge state measurements are examined. It is found that: (1) the temperature inferred by traditional interpretation of the interplanetary ionization state overestimates (underestimate) the actual coronal electron temperature if higher ion charge stages flow systematically faster (slower) than lower stages at the coronal freezing radius; (2) temperatures inferred from relative abundance measurements of ion-charge-stages with high ionization potentials moderately overestimate the actual coronal electron temperature if the high-energy tail of the coronal electron velocity distribution is enhanced relative to a Maxwellian distribution; (3) the propagation of a disturbance, e.g., a shock wave, through the corona can strongly affect the frozen-in charge state, but only over a time (a few times ten minutes) corresponding to the coronal transit time for the disturbance.

Owocki, S. P.

An improvement in SAR image interpretability provided by post-correlation signal processing

A presentation on the basis of subjective analysis of computer-generated SAR imagery depicts the improvement in interpretability obtained by post-correlation signal processing. A parametric study was conducted to determine the improvement in interpretability obtained by the application of signal weighting functions on the post-processed returns. The results suggest that a marked improvement in interpretability results from symmetrizing the exponential distribution of the fading signal. Preliminary analysis indicates that signal weighting improves the contrast ratio between the mean value of adjacent homogeneous regions in a SAR scene.

Matthews, N. D.

Interpretation of Landsat-4 Thematic Mapper and Multispectral Scanner data for forest surveys

Landsat-4 Thematic Mapper (TM) and Multispectral Scanner (MSS) data were evaluated by interpreting film and digital products and statistical data for selected forest cover types in California. Significant results were: (1) TM color image products should contain a spectral band in the visible (bands 1, 2, or 3), near infrared (band 4), and middle infrared (band 5) regions for maximizing the interpretability of vegetation types; (2) TM color composites should contain band 4 in all cases even at the expense of excluding band 5; and (3) MSS color composites were more interpretable than all TM color composites for certain cover types and for all cover types when band 4 was excluded from the TM composite.

Benson, A. S.