SEARCH · Engineering Papers
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.
Viewgraphs on the system of experts for intelligent data management (SEIDAM) are presented. Topics covered include: multiple data sources integration; SEIDAM structure; machine learning; user interface; spatial knowledge expert system; digital terrain model (DTM) and segmentation expert systems; system validation; and tasks and plans.
A challenging opportunity in structural health monitoring of composite materials is using machine learning (ML) methods to classify acoustic emissions (AE) according to the damage mechanism that emitted the signal. Although a wide variety of ML frameworks have been developed, there is a distinct lack of ground truth datasets which has precluded any direct assessment of their accuracy. Here, we present a novel ground truth dataset gathered on simplified unidirectional SiC/SiC composite structures. Herein, AE is collected from minicomposites which are loaded to targeted percentages of the ultimate tensile stress. These minicomposites are then volumetrically imaged with XCT and individual damage events, along with the mechanism, are correlated to AE. We explore the signal features that allow for mechanism discrimination, along with the feasibility of both unsupervised and supervised frameworks for use in the online monitoring of composite structures.
Exposure to spaceflight poses risk to human health in complex ways. To help manage this risk, the Human Systems Risk Board (HSRB) at the National Aeronautics and Space Administration (NASA) maintains a set of causal diagrams that attempt to explain how spaceflight hazards generate health risks and lead to adverse outcomes both in-mission, immediately post-mission, and over the long term. These causal risk diagrams are formulated as directed acyclic graphs (DAGs) and can function as knowledge graphs of connected risks and outcomes. These DAGs have proven useful for communication, and, through network analysis, have allowed for the identification of structurally important factors in the risk network. However, the utility these DAGs provide is directly proportional to their verisimilitude, making assessment of this trait using empirical data – whether from actual human spaceflight or various spaceflight analogue exposures and model organisms – a high priority. In this research we explore the use of machine learning algorithms to learn DAG structure from empirical data as a means of evaluating human-elicited DAG structures. To do so, we test several different graph structure-learning algorithms on data concerning changes in the bones of rats and mice after exposure to either spaceflight or a spaceflight analogue. We explore potential methods for indexing the similarity between each algorithm’s output DAG with all the others and with that of the expert-elicited DAG. We discuss next steps in this ongoing line of research and open science initiatives underway to complete them.
The complexity of cause-and-effect relationships between spaceflight hazards and resulting health conditions clouds understanding of the totality of human system risk in space. In response, NASA has introduced Directed Acyclic Graphs (causal diagrams) into the human systems risk management process. These diagrams allow for a common understanding of the mechanisms that lead from unique hazards of spaceflight to the health outcomes important to agencies and astronauts. However, the paucity of available biomedical data from spaceflight creates a need for methods of validating causal models that can accommodate data from spaceflight model analogs. Here we outline one approach utilizing open-access rodent bone datasets from the Ames Life Sciences Data Archive. The properties of directed acyclic graphs themselves can provide an epistemological and statistical framework for validation of a priori causal representations of human system risk in space flight. The assumed causal connections on the graph creates sets of logical implications: variables that – if the causal diagram is correct – should be correlated, as well as sets that should be conditionally independent. By testing these implied correlations and conditional independencies both statistically and heuristically, we can provide evidence for or against specific causal pathways on the causal diagram. In addition to validation of expert-generated causal diagrams, machine learning techniques can learn the most likely structure of a causal diagram from a given dataset. Comparison with and reconciliation between machine-learned causal diagrams and expert-generated diagrams is another technique for challenging assumptions and improving our understanding of causal mechanisms. Accurately representing complex causation is essential to systemic understanding of human health risks in space travel. Having a robust system of validating causal diagrams helps us arrive at more accurate representations of causal systems. This process will be integral to developing the countermeasures necessary for extended exploration of the moon and Mars.
We describe empirical methods for selecting a neural network architecture to implement belief state inference on generic commercial transport aircraft. We highlight a case study on the planning, execution, and analysis of a set of experiments to determine the configurations of a conditional variational autoencoder (CVAE). Our main contribution is the application of a structured method that can be used for machine learning in many aerospace applications. This method optimizes the structure and training parameters of a neural network for belief state inference, using Design of Experiments (DOE) statistical methodologies. The motivation for this specific DOE analysis was to identify the appropriate hyperparameters for measuring the CVAE reconstruction probability and latent space, such that the measurements can be used to infer qualitative state changes for the aircraft. We demonstrate that this process yields information about a trained neural network’s utility for this specific application, along with a quantifiable range of certainty. We execute 84 experiments using loss-of-control flight maneuver data from the NASA T-2 aircraft, demonstrating that this empirical process allows us to construct cheap and simple models with specific attributes amenable to belief state inference in aerospace applications.
We describe empirical methods for selecting a neural network architecture to implement belief state inference on generic commercial transport aircraft. We highlight a case study on the planning, execution, and analysis of a set of experiments to determine the configurations of a conditional variational autoencoder (CVAE). Our main contribution is the application of a structured method that can be used for machine learning in many aerospace applications. This method optimizes the structure and training parameters of a neural network for belief state inference, using Design of Experiments (DOE) statistical methodologies. The motivation for this specific DOE analysis was to identify the appropriate hyperparameters for measuring the CVAE reconstruction probability and latent space, such that the measurements can be used to infer qualitative state changes for the aircraft. We demonstrate that this process yields information about a trained neural network’s utility for this specific application, along with a quantifiable range of certainty. We execute 84 experiments using loss-of-control flight maneuver data from the NASA T 2 aircraft, demonstrating that this empirical process allows us to construct cheap and simple models with specific attributes amenable to belief state inference in aerospace applications.
Abstract—We describe empirical methods for selecting a neural network architecture to implement belief state inference on generic commercial transport aircraft. We highlight a case study on the planning, execution, and analysis of a set of experiments to determine the configurations of a conditional variational autoencoder (CVAE). Our main contribution is the application of a structured method that can be used for machine learning in many aerospace applications. This method optimizes the structure and training parameters of a neural network for belief state inference, using Design of Experiments (DOE) statistical methodologies. The motivation for this specific DOE analysis was to identify the appropriate hyperparameters for measuring the CVAE reconstruction probability and latent space, such that the measurements can be used to infer qualitative state changes for the aircraft. We demonstrate that this process yields information about a trained neural network’s utility for this specific application, along with a quantifiable range of certainty. We execute 84 experiments using loss-of-control flight maneuver data from the NASA T-2 aircraft, demonstrating that this empirical process allows us to construct cheap and simple models with specific attributes amenable to belief state inference in aerospace applications.
The use of structural health monitoring systems on spacecraft structures can play a crucial role in ensuring the safety, reliability, and longevity of the structure by gathering and analyzing onboard sensor data. Of specific importance is monitoring for excessive loading at critical interfaces as any off-nominal structural excitations experienced by spacecraft structures can cause early unpredicted high structural life consumption or damage. The availability and cost of flight-certified sensors along with the size of spacecraft structures and allowable payload mass drives the need for a method to estimate loads using sparsely-located sensors. Numerous approaches such as physics-based, statistical learning, and physics-enhanced statistical learning algorithms have gained popularity among structural prognostics applications. However, developing noise-robust prediction models to assess loads and structural life predictions from a sparse multi-sensor data acquisition system can be a challenging task. This paper discusses the evaluation of physics-based versus machine-learning algorithms for predicting loads and structural life at mission critical locations on the spacecraft structure using a finite element loads analysis with the application of simulated noise and noise reduction techniques. To estimate the loads from accelerations, the physics-based algorithm leverages a loads transformation matrix from a Craig-Bampton reduced finite element model. A System Equivalent Reduction Expansion Process (SEREP) and a pseudo-inverse approach are considered to expand from the onboard sensor degrees of freedom to the Craig-Bampton model degrees of freedom. The machine learning algorithm provides a data driven solution/mapping of the sensor accelerations to the loads at the mission critical locations using a high dimensionality analysis. Although these strategies produce comparable loads prediction without noise, the limitations of these strategies with incorporating simulated noise and noise reduction techniques with low signal to noise ratio signals are evaluated. The study demonstrates the immense potential of statistical learning algorithms for sparse structural prognostic models and enhancing signal denoising techniques. These findings also highlight the need for noise-resilient prognostic models and low-noise data acquisition systems onboard spacecraft structures.
Machine learning is a powerful tool for data exploration and model building with large data sets. This project aimed to use machine learning techniques to explore the inherent structure of data from rotorcraft gear tests, relationships between features and damage states, and to build a system for predicting gear health for future rotorcraft transmission applications. Classical machine learning techniques are difficult, if not irresponsible to apply to time series data because many make the assumption of independence between samples. To overcome this, Hidden Markov Models were used to create a binary classifier for identifying scuffing transitions and Recurrent Neural Networks were used to leverage long distance relationships in predicting discrete damage states. When combined in a workflow, where the binary classifier acted as a filter for the fatigue monitor, the system was able to demonstrate accuracy in damage state prediction and scuffing identification. The time dependent nature of the data restricted data exploration to collecting and analyzing data from the model selection process. The limited amount of available data was unable to give useful information, and the division of training and testing sets tended to heavily influence the scores of the models across combinations of features and hyper-parameters. This work built a framework for tracking scuffing and fatigue on streaming data and demonstrates that machine learning has much to offer rotorcraft health monitoring by using Bayesian learning and deep learning methods to capture the time dependent nature of the data. Suggested future work is to implement the framework developed in this project using a larger variety of data sets to test the generalization capabilities of the models and allow for data exploration.
The synergistic use of active and passive remote sensing (i.e., data fusion) demonstrates the ability of spaceborne light detection and ranging (LiDAR), synthetic aperture radar (SAR) and multispectral imagery for achieving the accuracy requirements of a global forest biomass mapping mission. This data fusion approach also provides a means to extend 3D information from discrete spaceborne LiDAR measurements of forest structure across scales much larger than that of the LiDAR footprint. For estimating biomass, these measurements mix a number of errors including those associated with LiDAR footprint sampling over regional - global extents. A general framework for mapping above ground live forest biomass (AGB) with a data fusion approach is presented and verified using data from NASA field campaigns near Howland, ME, USA, to assess AGB and LiDAR sampling errors across a regionally representative landscape. We combined SAR and Landsat-derived optical (passive optical) image data to identify forest patches, and used image and simulated spaceborne LiDAR data to compute AGB and estimate LiDAR sampling error for forest patches and 100m, 250m, 500m, and 1km grid cells. Forest patches were delineated with Landsat-derived data and airborne SAR imagery, and simulated spaceborne LiDAR (SSL) data were derived from orbit and cloud cover simulations and airborne data from NASA's Laser Vegetation Imaging Sensor (L VIS). At both the patch and grid scales, we evaluated differences in AGB estimation and sampling error from the combined use of LiDAR with both SAR and passive optical and with either SAR or passive optical alone. This data fusion approach demonstrates that incorporating forest patches into the AGB mapping framework can provide sub-grid forest information for coarser grid-level AGB reporting, and that combining simulated spaceborne LiDAR with SAR and passive optical data are most useful for estimating AGB when measurements from LiDAR are limited because they minimized forest AGB sampling errors by 15 - 38%. Furthermore, spaceborne global scale accuracy requirements were achieved. At least 80% of the grid cells at 100m, 250m, 500m, and 1km grid levels met AGB density accuracy requirements using a combination of passive optical and SAR along with machine learning methods to predict vegetation structure metrics for forested areas without LiDAR samples. Finally, using either passive optical or SAR, accuracy requirements were met at the 500m and 250m grid level, respectively.
The synergistic use of active and passive remote sensing (i.e., data fusion) demonstrates the ability of spaceborne light detection and ranging (LiDAR), synthetic aperture radar (SAR) and multispectral imagery for achieving the accuracy requirements of a global forest biomass mapping mission (+/-20 Mg/ha or 20%, the greater of the two, for at least 80% of grid cells). A data fusion approach also provides a means to extend 3D information from discrete spaceborne LiDAR measurements of forest structure across scales much larger than that of the LiDAR footprint. For estimating biomass, these measurements mix a number of errors including those associated with LiDAR footprint sampling over regional-global extents. A general framework for mapping above ground live forest biomass density (AGB) with a data fusion approach is presented and verified using data from NASA field campaigns near Howland, ME, USA, to assess AGB and LiDAR sampling errors across a regionally representative landscape. We combined SAR and Landsat-derived optical (passive optical) image data to identify contiguous areas (>0.5 ha) that are relatively homogenous in remote sensing metrics (forest patches). We used this image-derived data with simulated spaceborne LiDAR derived from orbit and cloud cover simulations and airborne data from NASA's Laser Vegetation Imaging Sensor (LVIS) to compute AGB and estimate LiDAR sampling error for forest patches and 100 m, 250 m, 500 m, and 1 km grid cells. At both the patch and grid scales, we evaluated differences in AGB estimation and sampling error from the combined use of LiDAR with both SAR and passive optical and with either SAR or passive optical alone. First, this data fusion approach demonstrates that incorporating forest patches into the AGB mapping framework can provide sub-grid forest information for coarser grid-level AGB reporting. Second, a data fusion approach for estimating AGB using simulated spaceborne LiDAR with SAR and passive optical image combinations reduced forest AGB sampling errors 12%-38% from those where LiDAR is used with SAR or passive optical alone. In absolute terms, sampling errors were reduced from 14-40 Mg/ha to 11-28 Mg/ha across all grid scales and prediction methods, where minimum sampling errors were 11, 15, 18, and 22 Mg/ha for 1 km, 500 m, 250 m, and 100 m grid scales, respectively. Third, spaceborne global scale accuracy requirements were achieved whereby at least 80% of the grid cells at 100 m, 250 m, 500 m, and 1 km grid levels met AGB accuracy requirements using a combination of passive optical and SAR along with machine learning methods to predict vegetation structure metrics for forested areas without LiDAR samples. Finally, using either passive optical or SAR, accuracy requirements were met at the 500 m and 250 m grid level, respectively..
Composite materials are increasingly being used in aerospace applications due to their superior strength-to-weight ratio compared to commonly used metals. A current limitation to widespread adoption is the certification of adhesively bonded joints. One approach to improving adhesive bonding in composites is accurately measuring the thickness of adhesive bondlines in composite laminates. Precise bondline thickness control is essential for aerospace applications where adhesive layer thickness directly affects joint fracture properties and structural performance. This study focused on implementing machine learning techniques to determine the ultrasonic time of flight (directly correlated to thickness) in adhesive bondlines throughout autoclave cure cycles. A high-temperature (use up to 180°C) ultrasonic scanning system was deployed in an autoclave to provide time of flight data through composite panels. Three experiments were conducted on the curing of 305 mm × 305 mm unidirectional composite panels. In the first experiment, a piecewise function was fit for the temperature correction factor to account for changing autoclave temperatures. Due to deficiencies in the first calibration experiment, a second experiment was run, and the results were used to train a machine learning model. The revised experiment, in combination with the machine learning model, significantly increased the accuracy of the bondline time of flight predictions (~14% error reduced to <1%). Data was processed using the Regression Learner Application in MATLAB®, with a Support Vector Machine selected for the model. The result was a machine learning algorithm capable of reliably quantifying ultrasonic time of flight through adhesive bondlines. The third experiment provided independent test data for the machine learning model, demonstrating that the model produces accurate predictions from data beyond its training set.
A lot has been learned from past experience with structural and machine element failures. The understanding of failure modes and the application of an appropriate design analysis method can lead to improved structural and machine element safety as well as serviceability. To apply Probabilistic Design Methodology (PDM), all uncertainties are modeled as random variables with selected distribution types, means, and standard deviations. It is quite difficult to achieve a robust design without considering the randomness of the design parameters which is the case in the use of the Deterministic Design Approach. The US Navy has a fleet of submarine-launched ballistic missiles. An umbilical plug joins the missile to the submarine in order to provide electrical and cooling water connections. As the missile leaves the submarine, an umbilical retract mechanism retracts the umbilical plug clear of the advancing missile after disengagement during launch and retrains the plug in the retracted position. The design of the current retract mechanism in use was based on the deterministic approach which puts emphasis on factor of safety. A new umbilical retract mechanism that is simpler in design, lighter in weight, more reliable, easier to adjust, and more cost effective has become desirable since this will increase the performance and efficiency of the system. This paper reports on a recent project performed at Tennessee State University for the US Navy that involved the application of PDM to the design of an umbilical retract mechanism. This paper demonstrates how the use of PDM lead to the minimization of weight and cost, and the maximization of reliability and performance.
Kinases are a family of proteins that function as molecular switches, regulating several essential cellular activities such as cell proliferation. Dysfunctional kinases are implicated in several types of cancers and hence they are actively pursued as drug targets. Given the vast number of complex kinase structures that are available in the protein data bank (PDB), there is a necessity to develop methodologies that can identify structurally important moieties of the kinases in an automated fashion, for such techniques can be instrumental in identifying novel drug targets. In this work, we develop a graph neural network (GNN) based deep learning framework for classifying the functionally active and inactive states of a large set of eukaryotic protein kinases, making use of their 3D structure from the PDB. We show that GNN based machine learning models can classify protein states with an accuracy greater than 97%. We further use the GNN models to automatically identify regions of the kinases that are important for its function. For this purpose, Gradient-weighted Class Activation Mapping (Grad-CAM) was implemented on the protein graphs. Remarkably, Grad-CAM consistently identifies the highly conserved DFG motif as the most important part of the protein across the entire kinome, without any prior input. Other regions of the hydrophobic core such as the HRD motif were also identified by the interpretable GNN framework, consistent with the literature. We discuss the significance of each of these regions in detail.
Lidar provides precise measurements of the three-dimensional structure of the clouds, aerosols, ocean/land/snow/ice surfaces, as well as ocean subsurface. Lidar also provides unique information about physical propertiesof particulates in the atmosphere for both radiative transfer and air quality applications.In this talk, I will present an overview of our recent studies of aerosols, clouds, ocean and snow using space-based lidar measurements (e.g., LITE, CALIPSO and ICESat-2), such as classifications of aerosols and thermodynamics phase of clouds, cloud microphysical properties, snow depths and phytoplankton biomass. I will also introduce a new concept of 3D Earth system observations with data fusion though combined active/passive remote sensing and machine learning. The new concept aims to reveal vertical structure of the aerosols/clouds/surfaces/subsurface from passive sensors by taking advantage of lidar measurements to effectively resolve the vertical structureby unscrambling the highly convoluted multi-angle, spectral and polarization information from passive sensors and apply the knowledge to a large swath where lidar measurements are not available.
There are multiple different algorithms to perform modal test sensor placement optimization: effective independence, residual kinetic energy, iterative Guyan reduction, genetic algorithms, or a brute-force methodology. However, any of these methods may be computationally expensive, especially for structural models with a large number of degrees of freedom. Given the high-cost and the need to optimize the solution, modal sensor placement is a great application for machine learning (ML) algorithms. In this paper, we will apply ML algorithms to determine the optimal sensor locations for simple and complex structures. We will also discuss the benefits and drawbacks of using machine learning over other sensor placement algorithms.
Automated Fiber Placement (AFP) has become a standard manufacturing technique in the creation of large scale composite structures due to its high production rates. However, the associated rapid layup that accompanies AFP manufacturing has a tendency to induce defects. We forward an inspection system that utilizes machine learning (ML) algorithms to locate and characterize defects from profilometry scans coupled with a data storage system and a user interface (UI) that allows for informed manufacturing. A Keyence LJ-7080 blue light profilometer is used for fast 2D height profiling. After scans are collected, they are process by ML algorithms, displayed to an operator through the UI, and stored in a database. The overall goal of the inspection system is to add an additional tool for AFP manufacturing. Traditional AFP inspection is done manually adding to manufacturing time and being subject to inspector errors or fatigue. For large parts, the inspection process can be cumbersome. The proposed inspection system has the capability of accelerating this process while still keeping a human inspector integrated and in control. This allows for the rapid capability of the automated inspection software and the robustness of a human checking for defects that the system either missed or misclassified.
Woven, ablative thermal protection system (TPS) materials provide a robust option for aggressive (re)entries and thus have been baselined for the upcoming Mars Sample Return (MSR) mission's Earth Entry System (EES). The reliability requirements for MSR-EES necessitate understanding of material property variability, which could be significant given the complex structure and anisotropic nature of properties in TPS weaves, as well as the response to potential impact with micrometeoroids or orbital debris during the EES re-entry. The TPS Certification by Analysis effort within the Entry Systems Modeling project seeks to provide computational models and analyses that support the certification against such material-based risks. For the present talk, focus will be given to the characterization of baseline woven TPS material properties and mechanical failure limits, which entails (1) use of computational techniques (e.g., machine learning) to interpret computed tomography images of the weave to generate representative structural models and (2) application of multiscale material modeling approaches to characterize thermomechanical and failure properties.