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.
This whitepaper addresses the following focal area: (1) Data acquisition and assimilation enabled by machine learning, AI, and advanced methods including experimental/network design/optimization, unsupervised learning (including deep learning), and hardware-related efforts involving AI (e.g., edge computing)
The JIRIAF project aims to combine geographically diverse computing facilities into an integrated science infrastructure. This project starts by dynamically evaluating temporarily unallocated or idled compute resources from multiple providers. These resources are integrated to handle additional workloads without affecting local running jobs. This paper describes our approach to launch best-effort batch tasks which exploit these underutilized resources. Our system measures the real-time behavior of jobs running on a machine and learns to distinguish typical performance from outliers. Unsupervised ML techniques are used to analyze hardware-level performance measures, followed by a real-time cross-correlation analysis to determine which applications cause performance degradation. We then ameliorate bad behavior by throttling these processes. We demonstrate that problematic performance interference can be detected and acted on, which makes it possible to continue to share resources between applications and simultaneously maintain high utilization levels in a computing cluster. We relocate the CLAS12 data processing workflow to a remote data center for a case study, preventing file migration and temporal data persistency.
Industrial Control Systems (ICS) are used to control physical processes in critical infrastructure. These systems are used in a wide variety of operations such as water treatment, power generation and distribution, and manufacturing. While the safety and security of these systems are of serious concern, recent reports have shown an increase in targeted attacks aimed at manipulating physical processes to cause catastrophic consequences. This trend emphasizes the need for algorithms and tools that provide resilient and smart attack detection mechanisms to protect ICS. In this paper, we propose an anomaly detection framework for ICS based on a deep neural network. The proposed methodology uses dilated convolution and long short-term memory (LSTM) layers to learn temporal as well as long term dependencies within sensor and actuator data in an ICS. The sensor/actuator data are passed through a unique feature engineering pipeline where wavelet transformation is applied to the sensor signals to extract features that are fed into the model. Additionally, this paper explores four variations of supervised deep learning models, as well as an unsupervised support vector machine (SVM) model for this problem. The proposed framework is validated on Secure Water Treatment testbed results. This framework detects more attacks in a shorter period of time than previously published methods.
Cryogenic electron tomography (cryoET) is a powerful tool in structural biology, enabling detailed 3D imaging of biological specimens at a resolution of nanometers. Despite its potential, cryoET faces challenges such as the missing wedge problem, which limits reconstruction quality due to incomplete data collection angles. Recently, supervised deep learning methods leveraging convolutional neural networks (CNNs) have considerably addressed this issue; however, their pretraining requirements render them susceptible to inaccuracies and artifacts, particularly when representative training data is scarce. To overcome these limitations, we introduce a proof-of-concept unsupervised learning approach using coordinate networks (CNs) that optimizes network weights directly against input projections. This eliminates the need for pretraining, reducing reconstruction runtime by 3–20× compared to supervised methods. Our in silico results show improved shape completion and reduction of missing wedge artifacts, assessed through several voxel-based image quality metrics in real space and a novel directional Fourier Shell Correlation (FSC) metric. Our study illuminates benefits and considerations of both supervised and unsupervised approaches, guiding the development of improved reconstruction strategies.
Focal Area(s): (1) We put forward the concepts of data assimilation enabled by machine learning, AI, and advanced methods including experimental/network design/optimization and unsupervised learning applied to downscale information within Earth System Models (ESMs). (2) We discuss predictive modeling through the use of AI techniques and other tools to design a prediction system comprising of a hierarchy of models (e.g., AI-driven model/component/parameterization selection) to improve the characterization of extreme hydroclimate events in ESMs. The AI-based models will run five-six order of magnitude times faster, yet will provide similar accuracy, allowing us to provide range bounds on uncertainty faster and thus enabling faster extreme event identification. (3) Further, we consider the insight gleaned from complex data (observed/simulated) using AI, big data analytics, and other advanced methods, including explainable AI and physics- or knowledge-guided AI to improve the characterization of extreme hydroclimate events in ESMs.
Abstract We present a machine learning workflow to discover signatures in acoustic measurements that can be utilized to create a low-dimensional model to accurately predict the location of keyhole pores formed during additive manufacturing processes. Acoustic measurements were sampled at 100 kHz during single-layer laser powder bed fusion (LPBF) experiments, and spatio-temporal registration of pore locations was obtained from post-build radiography. Power spectral density (PSD) estimates of the acoustic data were then decomposed using non-negative matrix factorization with custom $$\varvec{k}$$ k -means clustering (NMF $$\varvec{k}$$ k ) to learn the underlying spectral patterns associated with pore formation. NMF $$\varvec{k}$$ k returned a library of basis signals and matching coefficients to blindly construct a feature space based on the PSD estimates in an optimized fashion. Moreover, the NMF $$\varvec{k}$$ k decomposition led to the development of computationally inexpensive machine learning models which are capable of quickly and accurately identifying pore formation with classification accuracy of supervised and unsupervised label learning greater than 95% and 90%, respectively. The intrinsic data compression of NMF k , the relatively light computational cost of the machine learning workflow, and the high classification accuracy makes the proposed workflow an attractive candidate for edge computing toward in-situ keyhole pore prediction in LPBF.
Methods and systems for predicting irradiance include learning a classification model using unsupervised learning based on historical irradiance data. The classification model is updated using supervised learning based on an association between known cloudiness states and historical weather data. A cloudiness state is predicted based on forecasted weather data. An irradiance is predicted using a regression model associated with the cloudiness state.
This submission contains an ESRI map package (.mpk) with an embedded geodatabase for GIS resources used or derived in the Nevada Machine Learning project, meant to accompany the final report. The package includes layer descriptions, layer grouping, and symbology. Layer groups include: new/revised datasets (paleo-geothermal features, geochemistry, geophysics, heat flow, slip and dilation, potential structures, geothermal power plants, positive and negative test sites), machine learning model input grids, machine learning models (Artificial Neural Network (ANN), Extreme Learning Machine (ELM), Bayesian Neural Network (BNN), Principal Component Analysis (PCA/PCAk), Non-negative Matrix Factorization (NMF/NMFk) - supervised and unsupervised), original NV Play Fairway data and models, and NV cultural/reference data. See layer descriptions for additional metadata. Smaller GIS resource packages (by category) can be found in the related datasets section of this submission. A submission linking the full codebase for generating machine learning output models is available through the "Related Datasets" link on this page, and contains results beyond the top picks present in this compilation.
The adatom arrays on surfaces offer an ideal playground to explore the mechanisms of chemical bonding via changes in the local electronic tunneling spectra. While this information is readily available in hyperspectral scanning tunneling spectroscopy data, its analysis has been considerably impeded by a lack of suitable analytical tools. Here we develop a machine learning based workflow combining supervised feature identification in the spatial domain and unsupervised clustering in the energy domain to reveal the details of structure-dependent changes of the electronic structure in adatom arrays on the Co 3 Sn 2 S 2 cleaved surface. This approach, in combination with first-principles calculations, provides insight for using artificial neural networks to detect adatoms and classifies each based on their local neighborhood comprised of other adatoms. These structurally classified adatoms are further spectrally deconvolved. The unexpected inhomogeneity of electronic structures among adatoms in similar configurations is unveiled using this method, suggesting there is not a single atomic species of adatoms, but rather multiple types of adatoms on the Co 3 Sn 2 S 2 surface. This is further supported by a slight contrast difference in the images (or slight size variation) of the topography of the adatoms.
Focal Area: Data acquisition and assimilation enabled by machine learning (ML), artificial intelligence (AI), and advanced methods including experimental/network design/optimization, unsupervised learning (deep learning), leveraging advanced hardware (e.g., edge computing). This development is to enable solving the science questions regarding the human and climatic factors that interact with and drive global water scarcity.
Using supervised learning to train a machine learning model to predict an on-coming edge localized mode (ELM) requires a large number of labeled samples. Creating an appropriate data set from the very large database of discharges at a long-running tokamak, such as DIII-D, would be a very time-consuming process for a human. Considering this need and difficulty, we use coincidence anomaly detection, an unsupervised learning technique, to train an ELM-identifier to identify and label ELMs in the DIII-D discharge database. This ELM-identifier shows, simultaneously, a precision of 0.68 and a recall of 0.63 (AUC is 0.73) on identifying ELMs in example time series pulled from thousands of discharges spanning five years. In a test set of 50 discharges, the algorithm finds over 26 thousand ELM candidates, more than 5 times the existing catalog of ELMs labeled by humans.
Detecting quenches in superconducting (SC) magnets during training is a challenging process that involves capturing physical events that occur at different frequencies and appear as various signal features. These events may be correlated across instrumentation type, thermal cycle, and ramp. These events together build a more complete picture of continuous processes occurring in the magnet, and may allow us to flag potential precursors for quench detection. We present our work on building an automatic machine learning (ML) based quench detection system. We build upon our existing work on unsupervised auto-encoders for acoustic sensors and quench antenna (QA) by first establishing a supervised ML training pipeline. We show the results of an event tagging, analysis, and simulation framework on our QA and acoustic data which are used concurrently to build a training dataset for a supervised implementation. We then show how this supervised training can be used as a prior in a semi-supervised framework and compare this to the unsupervised neural network auto-encoder performance.This allows us to have a more concrete understanding of the performance of our algorithms relative to physical events occurring in the magnet, and also provides a baseline software tool to generically evaluate our quench prediction autoencoders under completely unsupervised, supervised, and semi-supervised training conditions.
Diagnosing battery states such as health, state-of-charge, or temperature is crucial for ensuring the safety and reliability of electrochemical energy storage systems. While some states, such as temperature, may be measured using cheap sensors, accurate diagnosis of battery health metrics usually requires time-consuming performance measurements, making them infeasible for use in real-world operation. These health metrics can be measured during lab-testing and then estimated on-line using predictive life models or via state observer algorithms such as Kalman filters, but these predictive methods should be supplemented by actual measurement of battery health whenever possible to ensure reliability. Rapid measurement of battery health may be done by various types of fast diagnostic techniques such as electrochemical impedance spectroscopy (EIS), which can be performed in only a few minutes and require only a fraction of the energy and power needed for a full charge and discharge measurement. But there is a substantial challenge for estimating battery health using EIS data, as EIS is sensitive to cell temperature, state-of-charge, current, and resting time in addition to health. Thus, utilizing EIS data to predict battery capacity requires correcting for all these additional variables, a task that is extremely difficult to handle analytically. This talk utilizes machine-learning methods to estimate the effectiveness of battery capacity prediction from EIS data, leveraging a data set of hundreds of EIS measurements recorded at varying temperature and state-of-charge throughout a 500-day aging study of 32 commercial, large-format NMC-Graphite lithium-ion batteries. Using EIS as input to machine-learning models is complicated by the nonlinear response of impedance to battery health, temperature, and state-of-charge, as well as the collinearity between the impedance response at neighboring frequencies, which can easily lead to overfit models. To train robust models, features from EIS data need to be extracted from the data or some subset of critical frequencies selected. Many approaches for extracting and selecting features from EIS data from electrochemical analysis and machine-learning fields were identified for analysis: using the entire raw spectra; selection of one, two, or many frequencies from the entire spectra; selecting interesting points from the EIS measurement using domain knowledge; fitting EIS with an equivalent-circuit model; calculating statistics on the raw impedance values; and reducing the dimensionality of the data using unsupervised linear (principal component analysis) and non-linear (uniform manifold approximation and projection) methods. These approaches were rigorously compared using a machine-learning pipeline approach, training linear, Gaussian process, and random forest regression models and quantifying performance using cross-validation as well as a held-out test set. An artificial neural network model trained on the raw spectra was also tested. Promising pipelines were fine-tuned via Bayesian hyperparameter optimization using cross-validation loss and training with class-specific weights to counter data set imbalance. The most reliable method for utilizing impedance in this work was the selection of two optimal frequencies through an exhaustive search, resulting in about 2% mean absolute error on test data for both Gaussian process and random forest model architectures. Interrogation of a variety of models reveals critical frequencies of 100 Hz and 103 Hz for this data set, though the optimal set of frequencies is not necessarily intuitive, i.e., the best performing models are not simply those that use impedance at frequencies that have the highest correlation to the relative discharge capacity. The best performing model is an ensemble model, which is able to predict battery capacity with 1.9% mean absolute error for unseen cells using impedance recorded at a variety of temperatures and states-of-charge.
Education researchers are deeply interested in understanding the way students organize their knowledge. Card sort tasks, which require students to group concepts, are one mechanism to infer a student’s organizational strategy. However, the limited resolution of card sort tasks means they necessarily miss some of the nuance in a student’s strategy. Here in this work, we propose new machine learning strategies that leverage a potentially richer source of student thinking: free-form written language justifications associated with student sorts. Using data from a university chemistry card sort task, we use vectorized representations of language and unsupervised learning techniques to generate qualitatively interpretable clusters, which can provide unique insight in how students organize their knowledge. We compared these to machine learning analysis of the students’ sorts themselves. Machine learning-generated clusters revealed different organizational strategies than those built into the task; for example, sorts by difficulty or even discipline. There were also many more categories generated by machine learning for what we would identify as more novice-like sorts and justifications than originally built into the task, suggesting students’ organizational strategies converge when they become more expert-like. Finally, we learned that categories generated by machine learning for students’ justifications did not always match the categories for their sorts, and these cases highlight the need for future research on students’ organizational strategies, both manually and aided by machine learning. In sum, the use of machine learning to analyze results from a card sort task has helped us gain a more nuanced understanding of students’ expertise, and demonstrates a promising tool to add to existing analytic methods for card sorts.
Recently, robust PCA has seen its wide application in various industries for its ability to perform the task of anomaly detection. The essence of robust PCA approach is to break down the signal into a low rank component and sparse component. In many applications, a simple breakdown of the signal without accounting for the signs of low rank components and sparse components would violate the physical constraints of the decomposed signal. In addition, often times, the signals in the real world collected for a long duration has smooth changes within a day and between days. As an example, the power signals collected in a photovoltaic (PV) system are cyclostationary, exhibiting these characteristics. Neglecting the smoothness of signals would result in miss detection of anomalous signals which are smooth within a day but non-smooth between days and vice versa. In this paper, we developed a signal decomposition approach for the purpose of anomaly detection based on the idea of low rank and sparse decomposition taking into consideration the signs of the decomposed low rank and sparse components and the within-day and between-day smooth changes in the original signals. The proposed unsupervised approach for fault detection eliminates the need for faulty samples required by other machine learning methods. It does not require the full I-V characteristics to work. Furthermore, there is no need for complex modelling of PV systems as in the case of power loss analysis. Using Monte Carlo simulations, we demonstrate the ability of our proposed approach for detecting anomalies of different duration and severity in PV systems.
International safeguards currently rely on material accountancy to verify that declared nuclear material is present and unmodified. Although effective, material accountancy for large bulk facilities can be expensive to implement due to the high precision instrumentation required to meet regulatory targets. Process monitoring has long been considered to improve material accountancy. However, effective integration of process monitoring has been met with mixed results. Given the large successes in other domains, machine learning may present a solution for process monitoring integration. Past work has shown that unsupervised approaches struggle due to measurement error. Although not studied in depth for a safeguards context, supervised approaches often have poor generalization for unseen classes of data (e.g., unseen material loss patterns). This work shows that engineered datasets, when used for training, can improve the generalization of supervised approaches. Further, the underlying models needed to generate these datasets need only accurately model certain high importance features.
SAND2025-11465O VoroClust, also known as Voronoi Clustering, is a fast, density-based unsupervised clustering algorithm applicable to high-resolution and high-dimensional data. It operates as quickly as distance-based clustering methods while effectively capturing complex regional geometries, matching the performance of current density-based methods. VoroClust employs a data-centered sphere cover to reduce computational demands while preserving data topology. It propagates clusters outward from local density peaks. Although supervised machine learning is powerful for applications like image classification and segmentation, it requires comprehensive, consistent datasets, which many applications lack. Unsupervised clustering algorithms analyze the structure of each dataset rather than relying on similarities with other examples, making them well-suited for practical applications with insufficient or inappropriate data for supervised learning. Sandia National Laboratories is a multimission laboratory managed and operated by National Technology & Engineering Solutions of Sandia, LLC, a wholly owned subsidiary of Honeywell International Inc., for the U.S. Department of Energy’s National Nuclear Security Administration under contract DE-NA0003525.
Human error (HE) is an important concern in safety-critical systems such as nuclear power plants (NPPs). HE has played a role in many accidents and outage incidents in NPPs. Despite the increased automation in NPPs, HE remains unavoidable. Hence, the need for HE detection is as important as HE prevention efforts. In NPPs, HE is rather rare. Hence, anomaly detection, a widely used machine learning technique for detecting rare anomalous instances, can be repurposed to detect potential HE. In this study, we develop an unsupervised anomaly detection technique based on generative adversarial networks (GANs) to detect anomalies in manually collected surveillance data in NPPs. More specifically, our GAN is trained to detect mismatches between automatically recorded sensor data and manually collected surveillance data, and hence, identify anomalous instances that can be attributed to HE. We test our GAN on both a real-world dataset and an external dataset obtained from a testbed, and we benchmark our results against state-of-the-art unsupervised anomaly detection algorithms, including one-class support vector machine and isolation forest. Our results show that the proposed GAN provides improved anomaly detection performance. Our study is promising for the future development of artificial intelligence based HE detection systems.