Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “feature importance analysis”

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
All sourcesNASA NTRSDOE OSTI
NASA NTRS1977

A study of key features of random atmospheric disturbance models for the approach flight phase

An analysis and brief simulator experiment were performed to identify and classify important features of random turbulence for the landing approach flight phase. The analysis of various wind models was carried out within the context of the longitudinal closed-loop pilot/vehicle system. The analysis demonstrated the relative importance of atmospheric disturbance scale lengths, horizontal versus vertical gust components, decreasing altitude, and spectral forms of disturbances versus the pilot/vehicle system. Among certain competing wind models, the analysis predicted no significant difference in pilot performance. This was confirmed by a moving base simulator experiment which evaluated the two most extreme models. A number of conclusions were reached: attitude constrained equations do provide a simple but effective approach to describing the closed-loop pilot/vehicle. At low altitudes the horizontal gust component dominates pilot/vehicle performance.

Heffley, R. K.↗
NASA NTRS2021

Automated Data Accountability for Missions in Mars Rover Data

As the Mars Curiosity Rover transmits data to the JPL Ground Data System (GDS), it frequently observes data loss and corruption, requiring re-transmits from the rover and Ground Data System Analysts (GDSA) to monitor the downlink process. As new missions are launched, the GDSA team redistributes analysts to these new missions, causing shortages in previous missions. The GDSA team can significantly benefit from the automation and optimization of the downlink process of telemetry data. In fact, there is a need for a better understanding of why the data is corrupted, so that the GDSA team can best determine the root cause of the issues in the GDS. This paper presents machine learning and deep learning based approaches to automate and optimize the detection of data loss. We first created a pipeline to automatically accumulate data from the telemetry databases (MAROS, Telemetry Data Storage, and GDS Elastic Search Database) in the downlink process. With our newly created datasets, we perform feature selection to supplement the GDSA understanding of the downlink process and provide supplemental analysis on the importance of different features. We implement various machine learning and deep learning based models, including support vector machines, ensemble methods, and deep neural networks and evaluate their accuracies in identifying whether a downlink process is complete or incomplete. We utilize fast hyperparameter optimization methods that allow our models to quickly be re-trained, allowing them to quickly be tuned and optimized on daily incoming data in real time. This hyperparameter optimization also allows our methods to be quickly integrated into other JPL missions. Our results show that our best-performing machine learning and deep learning based models outperform the existing GDSA detection software by 6 accuracy points and can aid analysts by providing insights into the data accountability problem. Since these various machine learning and deep learning approaches vary significantly in interpretability, we provide a discussion on the tradeoffs between their performance and trustworthiness in helping detect issues in data transmission.

Divsalar, Dariush↗
DOE OSTI2025

Interpretable machine learning-guided design of Fe-based soft magnetic alloys

Here, we present a machine learning (ML) guided approach to predict saturation magnetization (𝑀 S ) and coercivity (𝐻 C ) in Fe-rich soft magnetic alloys, particularly Fe-Si-B systems. ML models trained on experimental data reveal that increasing Si and B content reduces 𝑀 S from 1.81 T (DFT ≈ 2.04 T) to ≈1.54 T (DFT ≈ 1.56T) in Fe-Si-B, which is attributed to decreased magnetic density and structural modifications. Experimental validation of ML predicted magnetic saturation on Fe-1Si-1B (2.09 T), Fe-5Si-5B (2.01 T), and Fe-10Si-10B (1.54 T) alloy compositions further supports our findings. These trends are consistent with density functional theory predictions, which link increased electronic disorder and band broadening to lower 𝑀 S values. Experimental validation on selected alloys confirms the predictive accuracy of the ML model, with good agreement across compositions. Beyond predictive accuracy, detailed uncertainty quantification and model interpretability including through feature importance and partial dependence analysis reveal that 𝑀 S is governed by a nonlinear interplay between Fe content and early transition metal ratios, while 𝐻 C is more sensitive to processing conditions such as ribbon thickness and thermal treatment windows. The ML framework was further applied to Fe-Si-B/Cr/Cu/Zr/Nb alloys in a pseudoquaternary compositional space, which shows comparable magnetic properties to NANOMET (Fe 84.8 ⁢Si 0.5 ⁢B 9.4 ⁢Cu 0.8⁢ P 3.5 ⁢C 1 ), FINEMET (Fe 73.5 ⁢Si 13.5 ⁢B 9 Cu 1 ⁢Nb 3 ), NANOPERM (Fe 88 ⁢Zr 7⁢ B 4 ⁢Cu 1 ), and HITPERM (Fe 44 ⁢Co 44 ⁢Zr 7⁢ B 4 ⁢Cu 1 . Our findings demonstrate the potential of the ML framework for accelerated search of high-performance soft magnetic materials.

density functional theory↗
DOE OSTI2024

Platform Of Optimal Experiment Management

The platform of optimal experiment management, POEM, powered with automated machine learning to accelerate the discovery of optimal solutions, and automatically guide the design of experiments to be evaluated. POEM currently supports 1) random model explorations for experiment design, 2) sparse grid model explorations with Gaussian Polynomial Chaos surrogate model to accelerate experiment design ,3) time-dependent model sensitivity and uncertainty analysis to identify the importance features for experiment design, 4) model calibrations via Bayesian inference to integrate experiments to improve model performance, and 5) Bayesian optimization for optimal experimental design. In addition, POEM aims to simplify the process of experimental design for users, enabling them to analyze the data with minimal human intervention, and improving the technological output from research activities.

Wang, Congjian [Idaho National Laboratory (INL), I↗
NASA NTRS2003

Parametric Transformation Analysis

Because twisted coronal features are important proxies for predicting solar eruptive events, and, yet not clearly understood, we present new results to resolve the complex, non-potential magnetic field configurations of active regions. This research uses free-form deformation mathematics to generate the associated coronal magnetic field. We use a parametric representation of the magnetic field lines such that the field lines can be manipulated to match the structure of EUV and SXR coronal loops. The objective is to derive sigmoidal magnetic field solutions which allows the beta greater than 1 regions to be included, aligned and non-aligned electric currents to be calculated, and the Lorentz force to be determined. The advantage of our technique is that the solution is independent of the unknown upper and side boundary conditions, allows non-vanishing magnetic forces, and provides a global magnetic field solution, which contains high- and low-beta regimes and is consistent with all the coronal images of the region. We show that the mathematical description is unique and physical.

Gary, G. Allan↗
NASA NTRS2006

X-Windows PVT Widget Class

The X-Windows Process Validation Table (PVT) Widget Class ( Class is used here in the object-oriented-programming sense of the word) was devised to simplify the task of implementing network registration services for Information Sharing Protocol (ISP) graphical-user-interface (GUI) computer programs. Heretofore, ISP PVT programming tasks have required many method calls to identify, query, and interpret the connections and messages exchanged between a client and a PVT server. Normally, programmers have utilized direct access to UNIX socket libraries to implement the PVT protocol queries, necessitating the use of many lines of source code to perform frequent tasks. Now, the X-Windows PVT Widget Class encapsulates ISP client server network registration management tasks within the framework of an X Windows widget. Use of the widget framework enables an X Windows GUI program to interact with PVT services in an abstract way and in the same manner as that of other graphical widgets, making it easier to program PVT clients. Wrapping the PVT services inside the widget framework enables a programmer to treat a PVT server interface as though it were a GUI. Moreover, an alternate subclass could implement another service in a widget of the same type. This program was written by Matthew R. Barry of United Space Alliance for Johnson Space Center. For further information, contact the Johnson Technology Transfer Office at (281) 483-3809. MSC-23582 Shuttle Data Center File- Processing Tool in Java A Java-language computer program has been written to facilitate mining of data in files in the Shuttle Data Center (SDC) archives. This program can be executed on a variety of workstations or via Web-browser programs. This program is partly similar to prior C-language programs used for the same purpose, while differing from those programs in that it exploits the platform-neutrality of Java in implementing several features that are important for analysis of large sets of time-series data. The program supports regular expression queries of SDC archive files, reads the files, interleaves the time-stamped samples according to a chosen output, then transforms the results into that format. A user can choose among a variety of output file formats that are useful for diverse purposes, including plotting, Markov modeling, multivariate density estimation, and wavelet multiresolution analysis, as well as for playback of data in support of simulation and testing.

Barry, Matthew R.↗
NASA NTRS2006

Shuttle Data Center File-Processing Tool in Java

A Java-language computer program has been written to facilitate mining of data in files in the Shuttle Data Center (SDC) archives. This program can be executed on a variety of workstations or via Web-browser programs. This program is partly similar to prior C-language programs used for the same purpose, while differing from those programs in that it exploits the platform-neutrality of Java in implementing several features that are important for analysis of large sets of time-series data. The program supports regular expression queries of SDC archive files, reads the files, interleaves the time-stamped samples according to a chosen output, then transforms the results into that format. A user can choose among a variety of output file formats that are useful for diverse purposes, including plotting, Markov modeling, multivariate density estimation, and wavelet multiresolution analysis, as well as for playback of data in support of simulation and testing.

Barry, Matthew R.↗
DOE OSTI2025

On the minimum number of radiation field parameters to specify gas cooling and heating functions

Fast and accurate approximations of gas cooling and heating functions are needed for hydrodynamic galaxy simulations. We use machine learning to analyze atomic gas cooling and heating functions in the presence of a generalized incident local radiation field computed by Cloudy. We characterize the radiation field through binned radiation field intensities instead of the photoionization rates used in our previous work. We find a set of 6 energy bins whose intensities exhibit relatively low correlation. We use these bins as features to train machine learning models to predict Cloudy cooling and heating functions at fixed metallicity. We compare the relative SHapley Additive exPlanation (SHAP) value importance of the features. From the SHAP analysis, we identify a feature subset of 3 energy bins (0.5-1, 1-4, and 13-16Ry) with the largest importance and train additional models on this subset. We compare the mean squared errors and distribution of errors on both the entire training data table and a randomly selected 20% test set withheld from model training. The machine learning models trained with 3 and 6 bins, as well as 3 and 4 photoionization rates, have comparable accuracy everywhere, with errors ≳10 times smaller than for the interpolation table of Gnedin and Hollon (2012). We conclude that 3 energy bins (or 3 analogous photoionization rates: molecular hydrogen photodissociation, neutral hydrogen HI, and fully ionized carbon CVI) are sufficient to characterize the dependence of the gas cooling and heating functions on our assumed incident radiation field model.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2025

On the minimum number of radiation field parameters to specify gas cooling and heating functions

Fast and accurate approximations of gas cooling and heating functions are needed for hydrodynamic galaxy simulations. We use machine learning to analyze atomic gas cooling and heating functions computed by Cloudy in the presence of a generalized incident local radiation field. We characterize the radiation field through binned radiation field intensities instead of the photoionization rates used in our previous work. We find a set of 6 energy bins whose intensities exhibit relatively low correlation. We use these bins as features to train machine learning models to predict Cloudy cooling and heating functions at fixed metallicity. We compare the relative SHapley Additive exPlanation (SHAP) value importance of the features. From the SHAP analysis, we identify a feature subset of 3 energy bins ($0.5-1, 1-4$, and $13-16 \, \mathrm{Ry}$) with the largest importance and train additional models on this subset. We compare the mean squared errors and distribution of errors on both the entire training data table and a randomly selected 20% test set withheld from model training. The machine learning models trained with 3 and 6 bins, as well as 3 and 4 photoionization rates, have comparable accuracy everywhere, with errors $\gtrsim 10$ times smaller than for the interpolation table of Gnedin and Hollon (2012). We conclude that 3 energy bins (or 3 analogous photoionization rates: molecular hydrogen photodissociation, neutral hydrogen HI, and fully ionized carbon CVI) are sufficient to characterize the dependence of the gas cooling and heating functions on our assumed incident radiation field model.

79 ASTRONOMY AND ASTROPHYSICS↗
NASA NTRS1989

Propulsion over a wide Mach number range

Criteria is presented to assess the relative merits of different propulsion systems. Previous references focus mainly on subsonic or low supersonic flight speeds. The main focus here is on a higher range, from low supersonic to orbital velocities. Air breathing propulsion systems for hypersonic flight persent the engine designer with circumstances that differ in important fundamental ways from those encountered in engines designed for operation at subsonic or low supersonic speeds. This analysis highlights the importance of various features of hypersonic engine design. Since the performance of hypersonic engines are energy limited, unlike low speed engines which are stagnation pressure limited, the efficient use of the energy of the fuel used is critical to minimize the take-off fuel mass fraction of the vehicle. Furthermore, since the required energy increase of a vehicle per incremental speed change increases with speed, the engine must be designed to operate efficiently at high speed. An analysis of engine performance in terms of entropy changes of the flow passing through the engine allows comparison of various engine designs as well as a convenient method to determine the effect of individual engine component efficiencies on overall engine performance.

Resler, Edwin L., Jr.↗
DOE OSTI2026

Mitigating Algorithmic Bias in Cancer Site Classification Models

Purpose Integrating artificial intelligence in cancer diagnostics has improved tumor classification beyond rule-based systems. Despite these advancements, these models may still encode demographic biases. We conducted a large-scale, applied bias-probing study of a deep learning–based cancer site classifier to quantify race information encoded in document embeddings. We then evaluated how performance changes when race-correlated embedding dimensions are removed in a post-training sensitivity analysis. Methods The cancer site classifier was trained using 3.5 million electronic cancer pathology reports from six of the National Cancer Institute's SEER registries. We trained a hierarchical self-attention network to generate 400-dimensional document embeddings. These embeddings were used to train two downstream, gradient-boosted decision tree classifiers: one to classify the cancer sites and another to predict racial categories. We identified overlapping features by intersecting the top 50 feature-importance rankings from the site and race models and computed their cumulative feature importance in each model. As a post hoc sensitivity analysis, we progressively pruned these overlapping dimensions, retrained the site model, and compared overall macro-F1 and accuracy, race-stratified macro-F1, and group fairness metrics on the basis of demographic parity and equalized odds before and after pruning. Results The analysis revealed minimal feature overlap between the cancer site and race prediction models, and the cumulative importance scores indicated a negligible influence of racial information on clinical predictions. Post-training pruning of overlapping features did not compromise the models' diagnostic accuracy, with a 0.07% loss in accuracy. Conclusion Our findings demonstrate that HiSAN-generated embeddings from SEER data can be used effectively in cancer site classification without significant demographic bias influencing the outcomes. Post-training pruning therefore functions as a practical audit and sensitivity check.

Shivanna, Abhishek [ORNL] (ORCID:0009000665228593)↗
NASA NTRS1978

F100(3) parallel compressor computer code and user's manual

The Pratt & Whitney Aircraft multiple segment parallel compressor model has been modified to include the influence of variable compressor vane geometry on the sensitivity to circumferential flow distortion. Further, performance characteristics of the F100 (3) compression system have been incorporated into the model on a blade row basis. In this modified form, the distortion's circumferential location is referenced relative to the variable vane controlling sensors of the F100 (3) engine so that the proper solution can be obtained regardless of distortion orientation. This feature is particularly important for the analysis of inlet temperature distortion. Compatibility with fixed geometry compressor applications has been maintained in the model.

Mazzawy, R. S.↗
NASA NTRS2003

Use of EOS Data in AWIPS for Weather Forecasting

Operational weather forecasting relies heavily on real time data and modeling products for forecast preparation and dissemination of significant weather information to the public. The synthesis of this information (observations and model products) by the meteorologist is facilitated by a decision support system to display and integrate the information in a useful fashion. For the NWS this system is called Advanced Weather Interactive Processing System (AWIPS). Over the last few years NASA has launched a series of new Earth Observation Satellites (EOS) for climate monitoring that include several instruments that provide high-resolution measurements of atmospheric and surface features important for weather forecasting and analysis. The key to the utilization of these unique new measurements by the NWS is the real time integration of the EOS data into the AWIPS system. This is currently being done in the Huntsville and Birmingham NWS Forecast Offices under the NASA Short-term Prediction Research and Transition (SPORT) Program. This paper describes the use of near real time MODIS and AIRS data in AWIPS to improve the detection of clouds, moisture variations, atmospheric stability, and thermal signatures that can lead to significant weather development. The paper and the conference presentation will focus on several examples where MODIS and AIRS data have made a positive impact on forecast accuracy. The results of an assessment of the utility of these products for weather forecast improvement made at the Huntsville NWS Forecast Office will be presented.

Jedlovec, Gary J.↗
NASA NTRS1978

Digital image analysis of Landsat data

The paper defines some basic processing functions than can be performed digitally, discusses a variety of computer peripherals including hard copy output devices and television terminals, compares the characteristics of stand-alone image analysis systems, and briefly describes the important features of image processing and geographic information systems software. Several examples are discussed to demonstrate the efficient use of computers through implementation of hard copy image recorders and television display terminals. The commercially available stand-alone image analysis systems include a computer, appropriate peripherals, and sufficient software to perform analysis without other computational support.

Alford, W. L.↗
NASA NTRS1983

Crop identification of SAR data using digital textural analysis

After preprocessing SEASAT SAR data which included slant to ground range transformation, registration to LANDSAT MSS data and appropriate filtering of the raw SAR data to minimize coherent speckle, textural features were developed based upon the spatial gray level dependence method (SGLDM) to compute entropy and inertia as textural measures. It is indicated that the consideration of texture features are very important in SAR data analysis. The SEASAT SAR data are useful for the improvement of field boundary definitions and for an earlier season estimate of corn and soybean area location than is supported by LANDSAT alone.

Nuesch, D. R.↗
NASA NTRS2005

On Automating Failure Mode Analysis and Enforcing its Integrity

This paper reports our experience on the development of a design-for-safety (DFS) workbench called Risk Assessment and Management Environment (RAME) for microelectronic avionics systems. Our objective is to transform DFS practice from an ad-hoc, inefficient, error-prone approach to a stringent engineering process such that DFS can keep up with the rapidly growing complexity of avionics systems. In particular, RAME is built upon an information infrastructure that comprises a fault model, a knowledge base, and a failure reporting/tracking system. This infrastructure permits systematic learning from prior projects and enables the automation of failure modes, effects and criticality analysis (FMECA). Among other unique features, the most important advantage of RAME is its capability of directly accepting design source code in hardware description languages (HDLs) for automated failure mode analysis, which enables RAME to be compatible and to evolve with most electronic-computer-aided-design systems. Through an initial experimental evaluation of the RAME prototype, we show that our approach to FMECA automation improves failure mode analysis turn-around-time, completeness, and accuracy.

design for safety↗
NASA NTRS1983

AVE/VAS experiment: Synoptic summary and preliminary results

The AVE/VAS ground truth field experiment was conducted during the Spring of 1982 severe storms and weather research program. The experiment consisted of acquiring correlative ground truth measurements of rawinsonde data, corresponding to the time and space resolutions of VAS sounding data. The objectives of the AVE/VAS experiment are: (1) to acquire four dimensional data sets of the actual atmospheric structure down to the mesoscale; (2) to provide measurements for quantitative comparisons between ground based and VAS-derived atmospheric parameters; (3) to evaluate the impact of VAS data on diagnostic analysis of structural features and dynamical processes important to the development of mesoscale phenomena; (4) to evaluate the impact of VAS data on numerical model simulations, nowcasting, and other mesoscale forecasting systems.

Jedlovec, G. J.↗
NASA NTRS1984

Topological grid structure - A data structure for earth science modeling

The automated analysis of land surface features is increasingly important to earth scientists. User-friendly algorithms for studying these features can be integrated into geographic information systems through the use of topological grid structure, which maintains the simplicity and transportability of standard grid structure while providing the essential capability to treat groups of contiguous, identically-classified pixels (corresponding to lakes, forests, fields, etc.) as distinct spatial entities.

Goldberg, M.↗