Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “feature selection”

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 199 records · Page 11
All sourcesNASA NTRSDOE OSTI
DOE OSTI2025

EOSPAC User's Manual: Version 6.5 Second Edition (Rev. 3)

The EOSPAC utility package is a collection of interface routines, which can be used to access the SESAME data library and perform various data adjustments and interpolations on the SESAME data. The SESAME data library contains both thermodynamic (e.g., equation of state) and transport coefficients (e.g., opacity and conductivity). Note, for simplicity, the term EOS (equation of state) used herein includes both thermodynamic variables and transport coefficients. The EOSPAC utility package is designed to be used by physics codes (henceforth ”host codes”) written in multiple languages and on multiple platforms. The remainder of this manual is organized into several sections. Chapter 2 discusses conventions such as data organization and routine names. Chapter 3 provides a general overview of basic theory and models implemented within EOSPAC. Chapter 4 provides a general overview of how to use the EOSPAC interface library. Chapters 5 to 7 describe the public interfaces of EOSPAC in detail. Chapter 8 provides a brief introduction to some related tools, which may be of use to the user. Chapter 9 provides details related to some selected numerical features of EOSPAC. Chapter 10 gives examples for using the interface routines described in chapters 5 to 7. Chapter 11 provides technical support contact information. Chapter 12 contains a brief set of acknowledgments. Chapter 13 contains a list of referenced documents. Finally, chapter 14 lists the “table types: mnemonic conventions”, “table types: grouped by category, sorted by name”, “table types: eospac version 5 cross reference”, “options: setup phase”, “data information parameters”, “meta-data information parameters”, “options: interpolation phase”, and the “error codes”.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗
DOE OSTI2021

Next Generation Co-Molded One-Piece Automotive Parts

The purpose of this project was to determine the feasibility of co-molding Class A Sheet Molding Compound (SMC), structural SMC, and continuous fiber prepreg materials to produce a single piece co-molded automotive part, such as a hood. The combination of the three molding materials and the incorporation of selective design features (ribs, flanges, corrugations) was expected to eliminate the need for inner reinforcement panels, significantly reducing tooling costs and simplifying the manufacturing process. A multi-material solution would also result in significant weight savings. The scope of this project included the material characterization of the three materials, development of cure and flow simulation models, and validation of the models against parts molded with different combinations of the materials on an 11”x11” plaque tool with rib features. The intent of this project was to apply the learnings obtained from co-molding a part with simplified geometry to a Phase 2 project that would produce a single piece hood, co-molded with the same materials. Resins from INEOS Composites were provided to IDI to produce SMC and continuous fiber prepregs. Purdue University and INEOS Composites characterized the rheological and curing behavior of the different materials as well as the mechanical properties of the co-molded parts. This data was used by Purdue University to create simulation models. Models were created to predict both flow patterns and predict mechanical properties of different laminate constructions in and around the ribs. Michigan State University - Corktown validated Purdue’s models by co-molding the three materials in different combinations on a tool containing rib features provided by Century Tool. The co-molded parts were evaluated against the simulation models by Purdue, Corktown and INEOS. The project team demonstrated the following: • The ability to obtain a cohesive co-molded structure made up of Class A SMC, Structural SMC, and continuous fiber prepreg. • The ability to obtain a co-molded part with a Class A surface • Modeling of flow and fiber orientation • Modeling to predict mechanical properties of multi-material co-molded structures The predicted flow behavior and fiber orientation from simulations were then compared with the molded samples. Exact local orientation state was difficult to compare between microscopy and flow simulation, but captured general trends. Multi-material flow behavior was generally modeled well with SMC materials, but the introduction of woven material sheets requires further model development. Purdue compared the predicted versus actual mechanical properties, and INEOS Composites evaluated the surface appearance of unpainted and painted co-molded parts. The models developed by Purdue demonstrated that the mechanical properties can be predicted for parts with varying material configurations. The resulting models can be applied to future co-molded part design, tooling design, and molding conditions.

36 MATERIALS SCIENCE↗
DOE OSTI2024

Impact of Color Space and Color Resolution on Vehicle Recognition Models

In this study, we analyze both linear and nonlinear color mappings by training on versions of a curated dataset collected in a controlled campus environment. We experiment with color space and color resolution to assess model performance in vehicle recognition tasks. Color encodings can be designed in principle to highlight certain vehicle characteristics or compensate for lighting differences when assessing potential matches to previously encountered objects. The dataset used in this work includes imagery gathered under diverse environmental conditions, including daytime and nighttime lighting. Experimental results inform expectations for possible improvements with automatic color space selection through feature learning. Moreover, we find there is only a gradual decrease in model performance with degraded color resolution, which suggests the need for simplified data collection and processing. By focusing on the most critical features, we could see improved model generalization and robustness, as the model becomes less prone to overfitting to noise or irrelevant details in the data. Such a reduction in resolution will lower computational complexity, leading to quicker training and inference times.

47 OTHER INSTRUMENTATION↗
DOE OSTI2024

A Generative Model for Realistic Galaxy Cluster X-Ray Morphologies

Abstract The X-ray morphologies of clusters of galaxies display significant variations, reflecting their dynamical histories and the nonlinear dependence of X-ray emissivity on the density of the intracluster gas. Qualitative and quantitative assessments of X-ray morphology have long been considered a proxy for determining whether clusters are dynamically active or “relaxed.” Conversely, the use of circularly or elliptically symmetric models for cluster emission can be complicated by the variety of complex features realized in nature, spanning scales from megaparsecs down to the resolution limit of current X-ray observatories. In this work, we use mock X-ray images from simulated clusters from The Three Hundred project to define a basis set of cluster image features. We take advantage of the clusters’ approximate self-similarity to minimize the differences between images before encoding the remaining diversity through a distribution of high-order polynomial coefficients. Principal component analysis then provides an orthogonal basis for this distribution, corresponding to natural perturbations from an average model. This representation allows novel, realistically complex X-ray cluster images to be easily generated, and we provide code to do so. The approach provides a simple way to generate training data for cluster image analysis algorithms and could be straightforwardly adapted to generate clusters displaying specific types of features or selected by physical characteristics available in the original simulations.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2022

Methods for rule-based genome design

Methods and systems for designing, testing, and validating genome designs based on rules or constraints or conditions or parameters or features and scoring are described herein. A computer-implemented method includes receiving data for a known genome and a list of alleles, identifying and removing occurrences of each allele in the known genome, determining a plurality of allele choices with which to replace occurrences in the known genome, generating a plurality of alternative gene sequences for a genome design based on the known genome, wherein each alternative gene sequence comprises a different allele choice, applying a plurality of rules or constraints or conditions or parameters or features to each alternative gene sequence by assigning a score for each rule or constraint or condition or parameter or feature in each alternative gene sequence, resulting in scores for the applied plurality of rules or constraints or conditions or parameters or features, scoring each alternative gene sequence based on a weighted combination of the scores for the plurality of rules or constraints or conditions or parameters or features, and selecting at least one alternative gene sequence as the genome design based on the scoring.

Kuznetsov, Gleb↗
DOE OSTI2021

Characterizing Signatures of Geothermal Exploration Data with Machine Learning Techniques: An Application to the Nevada Play Fairway Analysis

We are introducing machine learning methods to the play fairway analysis to generate geothermal potential maps to support the evaluation of geothermal resource potential and the exploration for undiscovered blind geothermal systems in the Nevada Great Basin region. Our project aims to identify new ways to combine the play fairway data and empirically organize relationships between feature weights and labels in an improved workflow. As a means of doing this, we introduce machine learning methods to evaluate the influence of certain geological and geophysical features/feature sets in predicting geothermal favorability. This report highlights promising approaches based on supervised and unsupervised learning methods. First, we demonstrate a filter method applied to supervised classification modeling. The supervised filter method is based on permutation analysis to evaluate every possible feature combination/drop out scenario and rank feature influence based on the performance variance of supervised classification models. Additionally, we present an unsupervised factor analysis based on principal component analysis coupled with a semi-supervised kmeans clustering algorithm. This analysis allows us to identify the optimal number of groups/clusters for training sites and structural settings to identify feature patterns including correlation, variance, and latent and dominant feature relationships. The results from these methods offer a promising avenue for identifying favorable sources of predictive information to identify the locations of blind geothermal systems and furthering our understanding of complex geothermal feature and label relationships in the Great Basin region and beyond.

15 GEOTHERMAL ENERGY↗
DOE OSTI2023

Deciphering the evolution of flavin-dependent monooxygenase stereoselectivity using ancestral sequence reconstruction

Controlling the selectivity of a reaction is critical for target-oriented synthesis. Accessing complementary selectivity profiles enables divergent synthetic strategies, but is challenging to achieve in biocatalytic reactions given enzymes’ innate preferences of a single selectivity. Thus, it is critical to understand the structural features that control selectivity in biocatalytic reactions to achieve tunable selectivity. Here, we investigate the structural features that control the stereoselectivity in an oxidative dearomatization reaction that is key to making azaphilone natural products. Crystal structures of enantiocomplementary biocatalysts guided the development of multiple hypotheses centered on the structural features that control the stereochemical outcome of the reaction; however, in many cases, direct substitutions of active site residues in natural proteins led to inactive enzymes. Ancestral sequence reconstruction (ASR) and resurrection were employed as an alternative strategy to probe the impact of each residue on the stereochemical outcome of the dearomatization reaction. These studies suggest that two mechanisms are active in controlling the stereochemical outcome of the oxidative dearomatization reaction: one involving multiple active site residues in AzaH and the other dominated by a single Phe to Tyr switch in TropB and AfoD. Moreover, this study suggests that the flavin-dependent monooxygenases (FDMOs) adopt simple and flexible strategies to control stereoselectivity, which has led to stereocomplementary azaphilone natural products produced by fungi. This paradigm of combining ASR and resurrection with mutational and computational studies showcases sets of tools for understanding enzyme mechanisms and provides a solid foundation for future protein engineering efforts.

Science & Technology - Other Topics↗
DOE OSTI2022

Carbon Capture in Novel Porous Liquids

Direct air capture (DAC) of CO 2 is one of the negative emission technologies under development to limit the impacts of climate change. The dilute concentration of CO 2 in the atmosphere (~400 ppm) requires new materials for carbon capture with increased CO 2 selectivity that is not met with current materials. Porous liquids (PLs) are an emerging material that consist of a combination of solvents and porous hosts creating a liquid with permanent porosity. PLs have demonstrated excellent CO 2 selectivity, but the features that control how and why PLs selectively capture CO 2 is unknown. To elucidate these mechanisms, density functional theory (DFT) simulations were used to investigate two different PLs. The first is a ZIF-8 porous host in a water/glycol/2-methylimidazole solvent. The second is the CC13 porous organic cage with multiple bulky solvents. DFT simulations identified that in both systems, CO 2 preferentially bound in the pore window rather than in the internal pore space, identifying that the solvent-porous host interface controls the CO 2 selectivity. Additionally, SNL synthesized ZIF-8 based PL compositions. Evaluation of the long-term stability of the PL identified no change in the ZIF-8 crystallinity after multiple agitation cycles, identifying its potential for use in carbon capture systems. Through this project, SNL has developed a fundamental understanding of solvent-host interactions, as well as how and where CO 2 binds in PLs. Through these results, future efforts will focus not on how CO 2 behaves inside the pore, but on the porous host-solvent interface as the driving force for PL stability and CO 2 selectivity.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2022

Predicting battery capacity from impedance at varying temperature and state of charge using machine learning

Prediction of battery health from electrochemical impedance spectroscopy (EIS) data can enable rapid measurement of battery state in real-world applications without using additional sensors or time-consuming performance measurements. However, deconvoluting the effect of capacity, state of charge, and temperature on EIS response is complicated analytically. Here, various machine-learning models, such as linear, Gaussian process, random forest, and artificial neural network regression, are utilized to predict capacity from EIS using hundreds of capacity, direct current (DC) resistance, and EIS measurements recorded under varying conditions of health, temperature, and state of charge (SOC). Several feature extraction and selection methods from traditional electrochemical analysis and statistical modeling are explored using machine-learning pipelines. EIS data from just two frequencies can accurately predict capacity, and interrogation shows that the optimal set of frequencies is not usually intuitive. Best results are achieved with an ensemble model, which predicts battery capacity with a mean absolute error of 1.9% on data from unobserved cells.

25 ENERGY STORAGE↗
DOE OSTI2023

Supersonically expanded sodium metal-dilute halogen gas interactions. The importance of reaction populated and energy storing reservoir states and population inversion created amplification in Na2

The reactions of Cl, Br, and I with Nan=2,3 produced in a supersonic expansion form Na2* and Na* excited states extending across the visible and ultraviolet regions. Emission in the region extending from 410 to 600 nm indicates selectively formed excited state Na2 emission features. Experimental evidence suggests that this emission is associated with Na3 + X reactions. Broadband (0.5 cm−1) laser measurements demonstrate gain (population inversion) for select features at∼524–528(1%), ∼492(0.3%), and ∼458.7–461(0.8%) nm. Single mode (0.007 cm−1) measurements extending from 528.03 to 527.63 nm demonstrate amplification involving five to six individual rovibronic levels with a maximum gain close to 3% recorded at 527.9 nm. The observed gain is associated with select transitions from levels of the Na2 11Πu state populated, via identified curve crossings, through collision induced transfer from long-lived Na2 21Σg+ and 11Πg reservoir states. Collision induced population buildup in the lowest vibrational levels of these reservoir states and collision induced transfer to the Na2 11Πu state create a population inversion in transitions to the X 1Σg+ state of Na2. The observed amplification is aided by rapid vibrational and rotational relaxation in both the Na2 ground and excited reservoir states producing amplifiers in the visible region like the HF amplifier in the infrared. This study suggests the importance of reaction populated and energy storing long-lived reservoir states in small sodium molecule combustion processes and indicates the potential for providing new short wavelength visible and ultraviolet amplifiers for future laser-based chemical propulsion concepts.

Chemistry↗
DOE OSTI2022

Background-Aware 3-D Point Cloud Segmentation With Dynamic Point Feature Aggregation

With the proliferation of LiDAR sensors and 3-D vision cameras, 3-D point cloud analysis has attracted significant attention in recent years. In this article, we propose a novel 3-D point cloud learning network, referred to as dynamic point feature aggregation network (DPFA-Net), by selectively performing the neighborhood feature aggregation (FA) with dynamic pooling and an attention mechanism. DPFA-Net has two variants for semantic segmentation and classification of 3-D point clouds. As the core module of the DPFA-Net, we propose an FA layer, in which features of the dynamic neighborhood of each point are aggregated via a self-attention mechanism. In contrast to other segmentation models, which aggregate features from fixed neighborhoods, our approach can aggregate features from different neighbors in different layers providing a more selective and broader view to the query points and focusing more on the relevant features in a local neighborhood. In addition, to further improve the performance of semantic segmentation, we exploit the background–foreground (BF) information and present two novel approaches, namely, two-stage BF-Net and BF regularization. Experimental results show that the proposed DPFA-Net achieves the state-of-the-art overall accuracy score of 89.22% for semantic segmentation on the Stanford large-scale 3-D Indoor Spaces (S3DIS) dataset and provides consistently satisfactory performance across different tasks of semantic segmentation, part segmentation, and 3-D object classification. Furthermore, our model achieves 93.1% accuracy on the ModelNet40 dataset and provides a mean shape intersection-over-union (IoU) value of 85.5% for part segmentation on the ShapeNet-Part dataset. It is a also computationally more efficient compared to other methods.

3-D↗
DOE OSTI2021

A review of machine learning in building load prediction

The surge of machine learning in recent years has been empowering engineer modeling in various fields. The decreasing hardware cost, increasing data accessibility, and advances of building automation system (BAS) allow the collection and storage of a significant amount of building operation data. The two facts provide great opportunities of applying machine learning to building energy systems modeling and analysis. There are a great number of research papers on this topic but there lacks a comprehensive and general review to summarize the current development, limitations, gaps and future trend. In this review paper series, machine learning techniques in building energy system modeling and analysis are reviewed under the organization and logic of the machine learning definition by Tom M. Mitchell: a computer program is said to learn from experience E with respect to some class of tasks T and performance measure P if its performance at tasks in T, as measured by P, improves with experience E. This paper is the first part of the review paper series, which focuses on building load prediction. First, the applications of building load prediction model (task T) are reviewed. Then, the modeling algorithms improving machine learning performance and accuracy (performance P) are reviewed. At the same time, the literature on the data perspective for modeling (experience E), including data engineering from sensors level to data level, pre-processing, feature extraction and selection, is reviewed. Finally, what is well-studied and what is lacking but with great potential are concluded; the gaps between present and future utilization of machine learning techniques are identified; the future trend and development are also predicted. The target readers of this paper are not only researchers from the building side who can get exposed to cutting edge machine learning tools, but also those from machine learning side who can understand the potential and challenge to apply machine learning in buildings.

Liang, Zhang↗
DOE OSTI2025

Balancing Trade-offs: Adaptive Differential Privacy in Interpretable Machine Learning Models

In the advancing field of machine learning, balancing accuracy, interpretability, and privacy represents a significant challenge. The problem is exacerbated by the widespread deployment of pre-trained models locally in diverse applications, which could lead to various amounts of privacy leakage. Conventional Differential Privacy strategies, in which uniform noises are applied to model gradients, guarantee data privacy at the expense of accuracy and interpretability. This paper introduces a Feature-Sensitive Adaptive Differential Privacy (FADP) framework with a unique noise-adding strategy. Noises are adaptively added based on feature importance clustering, where important features are considered for interpretability. By employing a unique masking technique, FADP selectively preserves crucial features with minimal noise interference, maintaining accuracy while enhancing interpretability. The FADP framework addresses the limitations of traditional DP methods by preserving critical channels and improving interpretability — a vital requirement in machine learning applications that demand transparency in model decisions. Through comprehensive testing, FADP is shown to balance the trade-offs among accuracy, privacy, and interpretability, marking a substantial advancement in the field of privacy-preserving machine learning.

Farhad Riya, Farhin [University of Tennessee, Knox↗
DOE OSTI2023

Kinetic Control via Binding Sites within the Confined Space of Metal Metalloporphyrin-Frameworks for Enhanced Shape-Selectivity Catalysis

One striking feature of enzyme is its controllable ability to trap substrates via synergistic or cooperative binding in the enzymatic pocket, which renders the shape-selectivity of product by the confined spatial environment. The success of shape-selective catalysis relies on the ability of enzyme to tune the thermodynamics and kinetics for chemical reactions. In emulation of enzyme's ability, we showcase herein a targeting strategy with the substrate being anchored on the internal pore wall of metal-organic frameworks (MOFs), taking full advantage of the sterically kinetic control to achieve shape-selectivity for the reactions. For this purpose, a series of binding site-accessible metal metalloporphyrin-frameworks (MMPFs) have been investigated to shed light on the nature of enzyme-mimic catalysis. They exhibit a different density of binding sites that are well arranged into the nanospace with corresponding distances of opposite binding sites. Furthermore, such a structural specificity results in a facile switch in selectivity from an exclusive formation of the thermodynamically stable product to the kinetic product. Thus, the proposed targeting strategy, based on the combination of porous materials and binding events, paves a new way to develop highly efficient heterogeneous catalysts for shifting selectivity.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗
DOE OSTI2022

Post-hazard condition assessment of nuclear piping-equipment systems: Novel approach to feature extraction and deep learning

Over the past decade, the use of artificial intelligence techniques in the field of health-monitoring has gained significant interest, especially for structures such as building and bridges. However, applications to industrial systems such as equipment-piping systems in nuclear plants have not been explored. Here, in this paper, it is shown that the existing techniques developed for buildings and bridges cannot be extended directly to equipment-piping systems as the response of such systems is governed by multiple localized modes unlike that in buildings and bridges. This paper proposes a new approach that consists of three key aspects: (i) a novel vector of degradation-sensitive features extracted from measured data, (ii) using a deep Artificial Neural Network (ANN) for diagnosis of degradation location and degradation severity, and (iii) consideration of uncertainty in degradation severity when training the ANN. Degradation in piping-equipment systems can occur due to flow-accelerated erosion and corrosion. These locations can potentially exhibit damage such as localized yielding or initiation of cracking due to an external event such as an earthquake. Moreover, such locations can at times go undetected by current inspection techniques. Therefore, a robust framework is needed for detection of degradation after a seismic event. This manuscript proposes a proof-of-concept framework, which utilizes data collected from sensors to generate a deep ANN database for predicting degraded locations and severity in a piping-equipment system. Degradation severity is classified as minor, moderate, and severe. In the suggested methodology, a novel vector of degradation-sensitive features is extracted from the sensor data to train the ANN. A simple piping-equipment system is selected to demonstrate feature extraction as a means to simplify pattern recognition, explore the design and parameters of an ANN, and develop a sensor placement strategy. The effectiveness of the proposed framework is demonstrated on a realistic primary safety system of a two-loop nuclear reactor. It is shown that the proposed post-hazard condition assessment framework is able to detect degraded locations along with the severity levels, including minor degradation, with considerably higher accuracy.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗
DOE OSTI2024

Detecting Important Drivers of Gridded Population Modeling With Machine Learning

High-resolution population datasets have been lever-aged across a broad swath of domains, such as climate change, public policy, humanitarian aid, and rescue operations, among others. Machine learning methods were adopted to generate high-resolution or gridded population estimates by using various geospatial input features such as buildings, roads, and nighttime lights. In this study, we evaluate the importance of population features using Random Forest models across three levels of analysis, utilizing permutation measures. Our research aims to address key questions to enhance our understanding of high-resolution population modeling, such as: Are certain features globally (10 countries collectively) more important than others? Do optimal features vary by country? Within each country, do feature importance differ across administrative units? What similarities exist in feature importance at the global, country, and administrative unit levels? To answer these questions, we leverage the Kneedle algorithm to automate the selection of optimum features. We find that there are patterns displayed by features across spatial boundaries, evidenced by the same feature being the most important indicator of population across 7 of the 10 countries modeled. Our findings indicate that while important features may vary across geographies, certain features consistently hold greater importance than others agnostic of geography.

Lebakula, Viswadeep [ORNL] (ORCID:0000000152935914↗
DOE OSTI2025

Improving galaxy cluster selection with the outskirt stellar mass of galaxies

The number density and redshift evolution of optically selected galaxy clusters offer an independent measurement of the amplitude of matter fluctuations, 𝑆 8 . However, recent results have shown that clusters chosen by the redMaPPer algorithm show richness-dependent biases that affect the weak lensing signals and number densities of clusters, increasing uncertainty in the cluster mass calibration and reducing their constraining power. Here, in this work, we evaluate an alternative cluster proxy, outskirt stellar mass, 𝑀 out , defined as the total stellar mass within a [50, 100] kpc envelope centered on a massive galaxy. This proxy exhibits scatter comparable to redMaPPer richness, 𝜆, but is less likely to be subject to projection effects. We compare the Dark Energy Survey Year 3 redMaPPer cluster catalog with a 𝑀 out selected cluster sample from the Hyper-Suprime Camera survey. We use weak lensing measurements to quantify and compare the scatter of 𝑀 out and 𝜆 with halo mass. Our results show 𝑀 out has a scatter consistent with 𝜆, with a similar halo mass dependence, and that both proxies contain unique information about the underlying halo mass. We find 𝜆-selected samples introduce features into the measured Δ⁢Σ signal that are not well fit by a log-normal scatter only model, absent in 𝑀 out selected samples. Our findings suggest that 𝑀 out offers an alternative for cluster selection with more easily calibrated selection biases, at least at the generally lower richnesses probed here. Combining both proxies may yield a mass proxy with a lower scatter and more tractable selection biases, enabling the use of lower mass clusters in cosmology. Finally, we find the scatter and slope in the 𝜆 −𝑀 out scaling relation to be 0.49 ±0.02 and 0.38 ±0.09.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2024

Event Detection and Classification Using Machine Learning Applied to PMU Data for the Western US Power System

Smart grid technology enhances our comprehension and reliability of the power grid, leveraging Phasor Measurement Unit (PMU) data—time-synchronized, high-frequency measurements gathered across the US power grid. This paper employs machine learning techniques to effectively analyze the vast PMU data in Wide Area Monitoring Systems (WAMS) for power grid event detection and classification. Analyzing several months of real-world PMU data, the paper focuses on machine learning for fast, precise event detection and classification, corroborated by utility event logs. Practical challenges like feature extraction, dimensionality reduction, and model selection are addressed. A novel feature yielding improved results is discovered, and a supplementary algorithm for detecting small power grid faults is developed. The final algorithm is validated using a month-long real PMU data set, demonstrating its capability in accurately identifying power grid events in near real-time.

machine learning, event detection, PMU↗