Search NASA⌕ Search

SEARCH · Search NASA

Results for “data driven model”

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 1,783 records · Page 99

Unsupervised Parsing of Gaze Data with a Beta-process Vector Auto-regressive Hidden Markov Model

The first stage of analyzing eye-tracking data is commonly to code the data into sequences of fixations and saccades. This process is usually automated using simple, predetermined rules for classifying ranges of the time series into events, such as “if the dispersion of gaze samples is lower than the threshold, then code as a fixation; otherwise code as a saccade.” More recent approaches incorporate additional eye-movement categories in automated parsing algorithms, particularly glissades, by using time-varying, data-driven thresholds. We describe an alternative approach using the beta-process auto-regressive hidden Markov model (BP-AR-HMM). The BP-AR-HMM offers two main advantages over existing frameworks. First, it provides a statistical model for eye movement classification rather than a single estimate. Second, the BP-AR-HMM uses a latent process to model the number and nature of the types of eye-movements and hence is not constrained to predetermined categories. We applied the BP-AR-HMM both to high-sampling rate gaze data from Andersson, Larsson, Holmqvist, Stridh, and Nyström (2016) and to low-sampling rate data from the DIEM project (Mital, Smith, Hill, & Henderson, 2011). Driven by the data properties, the BP-AR-HMM identified clusters of samples resembling saccades, fixations and smooth pursuit, and a category that approximately corresponds to glissades (as well as blinks, errors, etc.). The BP-AR-HMM serves as an effective algorithm for data-driven event parsing alone or as an initial step in exploring the characteristics of gaze data sets.

eye tracking, fixation, fixation, saccade, smooth ↗

Off-Angle Iris Correction Methods

In many real world iris recognition systems obtaining consistent frontal images is problematic do to inexperienced or uncooperative users, untrained operators, or distracting environments. As a result many collected images are unusable by modern iris matchers. In this chapter we present four methods for correcting off-angle iris images to appear frontal which makes them compatible with existing iris matchers. The methods include an affine correction, a retraced model of the human eye, measured displacements, and a genetic algorithm optimized correction. The affine correction represents a simple way to create an iris image that appears frontal but it does not account for refractive distortions of the cornea. The other method account for refraction. The retraced model simulates the optical properties of the cornea. The other two methods are data driven. The first uses optical flow to measure the displacements of the iris texture when compared to frontal images of the same subject. The second uses a genetic algorithm to learn a mapping that optimizes the Hamming Distance scores between off-angle and frontal images. In this paper we hypothesize that the biological model presented in our earlier work does not adequately account for all variations in eye anatomy and therefore the two data-driven approaches should yield better performance. Results are presented using the commercial VeriEye matcher that show that the genetic algorithm method clearly improves over prior work and makes iris recognition possible up to 50 degrees off-angle.

iris recognition↗

Quantifying model-structure- and parameter-driven uncertainties in spring wheat phenology prediction with Bayesian analysis

Recent international efforts have brought renewed emphasis on the comparison of different agricultural systems models. Thus far, analysis of model-ensemble simulated results has not clearly differentiated between ensemble prediction uncertainties due to model structural differences per se and those due to parameter value uncertainties. Additionally, despite increasing use of Bayesian parameter estimation approaches with field-scale crop models, inadequate attention has been given to the full posterior distributions for estimated parameters. The objectives of this study were to quantify the impact of parameter value uncertainty on prediction uncertainty for modeling spring wheat phenology using Bayesian analysis and to assess the relative contributions of model-structure-driven and parameter-value-driven uncertainty to overall prediction uncertainty. This study used a random walk Metropolis algorithm to estimate parameters for 30 spring wheat genotypes using nine phenology models based on multi-location trial data for days to heading and days to maturity. Across all cases, parameter-driven uncertainty accounted for between 19 and 52% of predictive uncertainty, while model-structure-driven uncertainty accounted for between 12 and 64%. Here, this study demonstrated the importance of quantifying both model-structure- and parameter-value-driven uncertainty when assessing overall prediction uncertainty in modeling spring wheat phenology. More generally, Bayesian parameter estimation provided a useful framework for quantifying and analyzing sources of prediction uncertainty.

60 APPLIED LIFE SCIENCES↗

Dynamic Model Agnostic Reliability Evaluation of Machine-Learning Models Integrated in Instrumentation & Control Systems

In recent years, the field of data-driven neural network-based machine learning (ML) algorithms has grown significantly and spurred research in its applicability to instrumentation and control systems. While they are promising in operational contexts, the trustworthiness of such algorithms is not adequately assessed. Failures of ML-integrated systems are poorly understood; the lack of comprehensive risk modeling can degrade the trustworthiness of these systems. In recent reports by the National Institute for Standards and Technology, trustworthiness in ML is a critical barrier to adoption and will play a vital role in intelligent systems' safe and accountable operation. Thus, in this work, we demonstrate a real-time model-agnostic method to evaluate the relative reliability of ML predictions by incorporating out-of-distribution detection on the training dataset. It is well documented that ML algorithms excel at interpolation (or near-interpolation) tasks but significantly degrade at extrapolation. This occurs when new samples are "far" from training samples. The method, referred to as the Laplacian distributed decay for reliability (LADDR), determines the difference between the operational and training datasets, which is used to calculate a prediction's relative reliability. LADDR is demonstrated on a feedforward neural network-based model used to predict safety significant factors during different loss-of-flow transients. LADDR is intended as a "data supervisor" and determines the appropriateness of well-trained ML models in the context of operational conditions. Ultimately, LADDR illustrates how training data can be used as evidence to support the trustworthiness of ML predictions when utilized for conventional interpolation tasks.

97 MATHEMATICS AND COMPUTING↗

An adaptive, data-driven multiscale approach for dense granular flows

The accuracy of coarse-grained continuum models of dense granular flows is limited by the lack of high-fidelity closure models for granular rheology. One approach to addressing this issue, referred to as the hierarchical multiscale method, is to use a high-fidelity fine-grained model to compute the closure terms needed by the coarse-grained model. The difficulty with this approach is that the overall model can become computationally intractable due to the high computational cost of the high-fidelity model. In this work, we describe a multiscale modeling approach for dense granular flows that utilizes neural networks trained using high-fidelity discrete element method (DEM) simulations to approximate the constitutive granular rheology for a continuum incompressible flow model. Our approach leverages an ensemble of neural networks to estimate predictive uncertainty that allows us to determine whether the rheology at a given point is accurately represented by the neural network model. Additional DEM simulations are only performed when needed, minimizing the number of additional DEM simulations required when updating the rheology. This adaptive coupling significantly reduces the overall computational cost of the approach while controlling the error. In addition, the neural networks are customized to learn regularized rheological behavior to ensure well-posedness of the continuum solution. We first validate the approach using two-dimensional steady-state and decelerating inclined flows. We then demonstrate the efficiency of our approach by modeling three-dimensional sub-aerial granular column collapse for varying initial column aspect ratios, where our multiscale method compares well with the computationally expensive computational fluid dynamics (CFD)-DEM simulation.

Dense granular flows↗

Projection pursuit adaptation on polynomial chaos expansions

Here, the present work addresses the issue of accurate stochastic approximations in high-dimensional parametric space using tools from uncertainty quantification (UQ). The basis adaptation method and its accelerated algorithm in polynomial chaos expansions (PCE) were recently proposed to construct low-dimensional approximations adapted to specific quantities of interest (QoI). The present paper addresses one difficulty with these adaptations, namely their reliance on quadrature point sampling, which limits the reusability of potentially expensive samples. Projection pursuit (PP) is a statistical tool to find the “interesting” projections in high-dimensional data and thus bypass the curse-of-dimensionality. In the present work, we combine the fundamental ideas of basis adaptation and projection pursuit regression (PPR) to propose a novel method to simultaneously learn the optimal low-dimensional spaces and PCE representation from given data. While this projection pursuit adaptation (PPA) can be entirely data-driven, the constructed approximation exhibits mean-square convergence to the solution of an underlying governing equation and thus captures the supports and probability distributions associated with the physics constraints. The proposed approach is demonstrated on a borehole problem and a structural dynamics problem, demonstrating the versatility of the method and its ability to discover low-dimensional manifolds with high accuracy with limited data. In addition, the method can learn surrogate models for different quantities of interest while reusing the same data set.

97 MATHEMATICS AND COMPUTING↗

Modeling of the metal–insulator transition temperature in alio-valently doped VO 2 through symbolic regression

The correlated semiconductor vanadium dioxide (VO 2 ) exhibits an insulator–metal transition (IMT) near room temperature, which is of interest in various device applications. Precise IMT temperature control is crucial to determine the use cases across technologies such as thermochromic windows, actuators for robots or neuronal oscillators. Doping the cation or anion sites can modulate the IMT by several tens of degrees and control hysteresis. However, modeling the effects of control parameters (e.g., doping concentration, type of dopants) is challenging due to complex experimental procedures and limited data, hindering the use of traditional data-driven machine learning approaches. Symbolic regression (SR) can bridge this gap by identifying nonlinear expressions connecting key input parameters to target properties, even with small data sets. In this work, we develop SR models to capture the IMT trends in VO 2 influenced by different dopant parameters. Using experimental data from the literature, our study reveals a dual nature of the IMT temperature with varying tungsten (W) doping concentrations. The symbolic model captures data trends and accounts for experimental variability, providing a complementary approach to first-principles calculations. Our feature-driven analysis across a broader class of dopants informs selectivity and provides qualitative insights into tuning phase transition properties valuable for neuromorphic computing and thermochromic windows.

36 MATERIALS SCIENCE↗

Machine learning models inaccurately predict current and future high-latitude C balances

The high-latitude carbon (C) cycle is a key feedback to the global climate system, yet because of system complexity and data limitations, there is currently disagreement over whether the region is a source or sink of C. Recent advances in big data analytics and computing power have popularized the use of machine learning (ML) algorithms to upscale site measurements of ecosystem processes, and in some cases forecast the response of these processes to climate change. Due to data limitations, however, ML model predictions of these processes are almost never validated with independent datasets. To better understand and characterize the limitations of these methods, we develop an approach to independently evaluate ML upscaling and forecasting. We mimic data-driven upscaling and forecasting efforts by applying ML algorithms to different subsets of regional process-model simulation gridcells, and then test ML performance using the remaining gridcells. In this study, we simulate C fluxes and environmental data across Alaska using ecosys, a process-rich terrestrial ecosystem model, and then apply boosted regression tree ML algorithms to training data configurations that mirror and expand upon existing AmeriFLUX eddy-covariance data availability. We first show that a ML model trained using ecosys outputs from currently-available Alaska AmeriFLUX sites incorrectly predicts that Alaska is presently a modeled net C source. Increased spatial coverage of the training dataset improves ML predictions, halving the bias when 240 modeled sites are used instead of 15. However, even this more accurate ML model incorrectly predicts Alaska C fluxes under 21st century climate change because of changes in atmospheric CO 2 , litter inputs, and vegetation composition that have impacts on C fluxes which cannot be inferred from the training data. Our results provide key insights to future C flux upscaling efforts and expose the potential for inaccurate ML upscaling and forecasting of high-latitude C cycle dynamics.

54 ENVIRONMENTAL SCIENCES↗

Statistical Treatment of Convolutional Neural Network Superresolution of Inland Surface Wind for Subgrid-Scale Variability Quantification

Abstract Machine learning models have been employed to perform either physics-free data-driven or hybrid dynamical downscaling of climate data. Most of these implementations operate over relatively small downscaling factors because of the challenge of recovering fine-scale information from coarse data. This limits their compatibility with many global climate model outputs, often available between ∼50- and 100-km resolution, to scales of interest such as cloud resolving or urban scales. This study systematically examines the capability of a type of superresolving convolutional neural network (SR-CNNs) to downscale surface wind speed data over land from different coarse resolutions (25-, 48-, and 100-km resolution) to 3 km. For each downscaling factor, we consider three convolutional neural network (CNN) configurations that generate superresolved predictions of fine-scale wind speed, which take between one and three input fields: coarse wind speed, fine-scale topography, and diurnal cycle. In addition to fine-scale wind speeds, probability density function parameters are generated through which sample wind speeds can be generated, accounting for the intrinsic stochasticity of wind speed. For assessing generalization to new data, CNN models are tested on regions with different topography and climate that are unseen during training. The evaluation of superresolved predictions focuses on subgrid-scale variability and the recovery of extremes. Models with coarse wind and fine topography as inputs exhibit the best performance when compared with other model configurations, operating across the same downscaling factor. Our diurnal cycle encoding results in lower out-of-sample generalizability when compared with other input configurations.

17 WIND ENERGY↗

Sensitivity of Global Modeling Initiative Model Predictions of Antarctic Ozone Recovery to Input Meteorological Fields

We use the Global Modeling Initiative chemistry and transport model to simulate the evolution of stratospheric ozone between 1995 and 2030, using boundary conditions consistent with the recent World Meteorological Organization ozone assessment. We compare the Antarctic ozone recovery predictions of two simulations, one driven by an annually repeated year of meteorological data from a general circulation model (GCM), the other using a year of output from a data assimilation system (DAS), to examine the sensitivity of Antarctic ozone recovery predictions to the characteristic dynamical differences between GCM- and DAS-generated meteorological data. Although the age of air in the Antarctic lower stratosphere differs by a factor of 2 between the simulations, we find little sensitivity of the 1995-2030 Antarctic ozone recovery between 350 and 650 K to the differing meteorological fields, particularly when the recovery is specified in mixing ratio units. Percent changes are smaller in the DAS-driven simulation compared to the GCM-driven simulation because of a surplus of Antarctic ozone in the DAS-driven simulation which is not consistent with observations. The peak ozone change between 1995 and 2030 in both simulations is approx.20% lower than photochemical expectations, indicating that changes in ozone transport due to changing ozone gradients at 450 K between 1995 and 2030 constitute a small negative feedback. Total winter/spring ozone loss during the base year (1995) of both simulations and the rate of ozone loss during August and September is somewhat weaker than observed. This appears to be due to underestimates of Antarctic Cl(sub y) at the 450 K potential temperature level.

Considine, David B.↗

Search for phenomena beyond the Standard Model in events with large $b$-jet multiplicity using the ATLAS detector at the LHC

A search is presented for new phenomena in events characterised by high jet multiplicity, no leptons (electrons or muons), and four or more jets originating from the fragmentation of b -quarks ( b -jets). The search uses 139 fb -1 of $\sqrt{s}$ = 13 TeV proton–proton collision data collected by the ATLAS experiment at the Large Hadron Collider during Run 2. The dominant Standard Model background originates from multijet production and is estimated using a data-driven technique based on an extrapolation from events with low b -jet multiplicity to the high b -jet multiplicities used in the search. No significant excess over the Standard Model expectation is observed and 95% confidence-level limits that constrain simplified models of R-parity-violating supersymmetry are determined. The exclusion limits reach 950 GeV in top-squark mass in the models considered.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

Computation and turbulence modeling for three-dimensional boundary layers including turbomachinery rotor flows

A method is developed for predicting the behavior of three-dimensional, turbulent boundary layers occurring in internal flows, including those on turbomachinery rotor blades. These boundary layers are complex, turbulent, and subject to Coriolis and centrifugal forces. The major thrust of this paper is the development and use of an algebraic Reynolds stress model (ARSM) that captures the changes in turbulent flow structure arising from curvature, rotation, and three dimensionality. The prediction of pressure-driven secondary flow agrees well with the measured data, and all three turbulence models (k-epsilon, algebraic eddy viscosity, and ARSM) show the same level of agreement. The prediction of boundary-layer development on rotor blades shows much better agreement with measurements with the ARSM. It is essential to employ higher-order turbulence models to capture the effects of rotation, curvature, and three dimensionality on boundary layers in turbomachinery.

Zhang, J.↗

Modelling Temporal Variability in the Carbon Balance of a Spruce/Moss Boreal Forest

A model of the daily carbon balance of a black spruce/feathermoss boreal forest ecosystem was developed and results compared to preliminary data from the 1994 BOREAS field campaign in northern Manitoba, Canada. The model, driven by daily weather conditions, simulated daily soil climate status (temperature and moisture profiles), spruce photosynthesis and respiration, moss photosynthesis and respiration, and litter decomposition. Model agreement with preliminary field data was good for net ecosystem exchange (NEE), capturing both the asymmetrical seasonality and short-term variability. During the growing season simulated daily NEE ranged from -4 g C m(exp -2) d(exp -1) (carbon uptake by ecosystem) to + 2 g C m(exp -2) d(exp -1) (carbon flux to atmosphere), with fluctuations from day to day. In the early winter simulated NEE values were + 0.5 g C m(exp -2) d(exp -1), dropping to + 0.2 g C m(exp -2) d(exp -1) in mid-winter. Simulated soil respiration during the growing season (+ 1 to + 5 g C m(exp -2) d(exp -1)) was dominated by metabolic respiration of the live moss, with litter decomposition usually contributing less than 30% and live spruce root respiration less than 10% of the total. Both spruce and moss net primary productivity (NPP) rates were higher in early summer than late summer. Simulated annual NEE for 1994 was -51 g C m(exp -2) y(exp -1), with 83% going into tree growth and 17% into the soil carbon accumulation. Moss NPP (58 g C m(exp -2) d(exp -1)) was considered to be litter (i.e. soil carbon input; no net increase in live moss biomass). Ecosystem respiration during the snow-covered season (84 g Cm(exp -2)) was 58% of the growing season net carbon uptake. A simulation of the same site for 1968-1989 showed about 10-20% year-to-year variability in heterotrophic respiration (mean of + 113 g C m-2 y@1). Moss NPP ranged from 19 to 114 g C m(exp -2) y(exp -1); spruce NPP from 81 to 150 g C nt-2 y,@l; spruce growth (NPP minus litterfall) from 34 to 103 g C m(exp -2) y(exp -1); NEE ranged from +37 to -142 g C m(exp -2) y(exp -1). Values for these carbon balance terms in 1994 were slightly smaller than the 1969 - 89 means. Higher ecosystem productivity years (more negative NEE) generally had early springs and relatively wet summers; lower productivity years had late springs and relatively dry summers.

Frolking, S.↗

General Purpose Data-Driven Online System Health Monitoring with Applications to Space Operations

Modern space transportation and ground support system designs are becoming increasingly sophisticated and complex. Determining the health state of these systems using traditional parameter limit checking, or model-based or rule-based methods is becoming more difficult as the number of sensors and component interactions grows. Data-driven monitoring techniques have been developed to address these issues by analyzing system operations data to automatically characterize normal system behavior. System health can be monitored by comparing real-time operating data with these nominal characterizations, providing detection of anomalous data signatures indicative of system faults, failures, or precursors of significant failures. The Inductive Monitoring System (IMS) is a general purpose, data-driven system health monitoring software tool that has been successfully applied to several aerospace applications and is under evaluation for anomaly detection in vehicle and ground equipment for next generation launch systems. After an introduction to IMS application development, we discuss these NASA online monitoring applications, including the integration of IMS with complementary model-based and rule-based methods. Although the examples presented in this paper are from space operations applications, IMS is a general-purpose health-monitoring tool that is also applicable to power generation and transmission system monitoring.

Iverson, David L.↗

Probing interfacial momentum closures in two-phase bubbly flow with machine learning-aided methods

Computational fluid dynamics (CFD) approach has already reached a high level of maturity for single-phase flows, however the development of closure models for two-phase flow requires additional attention. Multiphase CFD (M-CFD) methods resolve the conservation equations for mass, momentum and energy while differing in the approaches and strategies adopted in the physical closure models. The most widely adopted framework for M-CFD is the Eulerian-Eulerian two-fluid approach which assumes that all phases are co-existing inside each computational cell. For each fluid, the full set of conservation equations is solved; therefore, each fluid has a different velocity field. For adiabatic two-phase flow, the mechanisms of the interfacial momentum transfer are modeled by the interfacial forces representing different physical mechanisms. One of the crucial issues in the development and application of two-fluid model is the understanding of the interfacial momentum closures which determines the bubble distribution and migration behaviors. Dedicated experiments are performed to support the physical understanding and drive the closures’ development. However, limitations exist due to the uncertainties in the experimental measurement and the simplified analytical assumptions which have difficulties on representing the complex non-linear flow fields. In this paper, a data-driven approach, Feature Similarity Measurement (FSM), is developed and proposed to resolve the challenges of modeling the interfacial forces closures. Case study is performed with two-phase flow scenarios where the high-fidelity experimental data is available. Within the Eulerian-Eulerian two-fluid framework, only momentum equations for gas and liquid phases are solved and reduced-order interfacial momentum closures are aided with FSM. Predictions of void fraction and velocity fields are analyzed and demonstrate the potential of machine learning-driven interfacial forces closures.

97 MATHEMATICS AND COMPUTING↗

Near-infrared spectroscopy of the Y0 WISEP J173835.52+273258.9 and the Y1 WISE J035000.32-565830.2: the importance of non-equilibrium chemistry

We present new near-infrared spectra, obtained at Gemini Observatory, for two Y dwarfs: WISE J035000.32−565830.2 (W0350) and WISEP J173835.52+273258.9 (W1738). A FLAMINGOS-2 R = 540 spectrum was obtained for W0350, covering 1.0<λ μm < 1.7, and a cross-dispersed Gemini near-infrared spectrograph R = 2800 spectrum was obtained for W1738, covering 0.993–1.087 μm, 1.191–1.305 μm, 1.589–1.631 μm, and 1.985–2.175 μm, in four orders. We also present revised YJH photometry for W1738, using new NIRI Y and J imaging, and a re-analysis of the previously published NIRI H-band images. We compare these data, together with previously published data for late-T and Y dwarfs, to cloud-free models of solar metallicity, calculated both in chemical equilibrium and with disequilibrium driven by vertical transport. We find that for the Y dwarfs, the non-equilibrium models reproduce the near-infrared data better than the equilibrium models. The remaining discrepancies suggest that fine-tuning the CH{sub 4}/CO and NH{sub 3}/N{sub 2} balance is needed. Improved trigonometric parallaxes would improve the analysis. Despite the uncertainties and discrepancies, the models reproduce the observed near-infrared spectra well. We find that for the Y0, W1738, T{sub eff}=425±25 K, and log g=4.0±0.25, and for the Y1, W0350, T{sub eff}=350±25 K, and log g=4.0±0.25. W1738 may be metal-rich. Based on evolutionary models, these temperatures and gravities correspond to a mass range for both Y dwarfs of 3–9 Jupiter masses, with W0350 being a cooler, slightly older, version of W1738; the age of W0350 is 0.3–3 Gyr, and the age of W1738 is 0.15–1 Gyr.

79 ASTRONOMY AND ASTROPHYSICS↗

Image Collection Simulation Using High-Resolution Atmospheric Modeling

A new method is described for simulating the passive remote sensing image collection of ground targets that includes effects from atmospheric physics and dynamics at fine spatial and temporal scales. The innovation in this research is the process of combining a high-resolution weather model with image collection simulation to attempt to account for heterogeneous and high-resolution atmospheric effects on image products. The atmosphere was modeled on a 3D voxel grid by a Large-Eddy Simulation (LES) driven by forcing data constrained by local ground-based and air-based observations. The spatial scale of the atmospheric model (10–100 m) came closer than conventional weather forecast scales (10–100 km) to approaching the scale of typical commercial multispectral imagery (2 m). This approach was demonstrated through a ground truth experiment conducted at the Department of Energy Atmospheric Radiation Measurement Southern Great Plains site. In this experiment, calibrated targets (colored spectral tarps) were placed on the ground, and the scene was imaged with WorldView-3 multispectral imagery at a resolution enabling the tarps to be visible in at least 9–12 image pixels. The image collection was simulated with Digital Imaging and Remote Sensing Image Generation (DIRSIG) software, using the 3D atmosphere from the LES model to generate a high-resolution cloud mask. The high-resolution atmospheric model-predicted cloud coverage was usually within 23% of the measured cloud cover. The simulated image products were comparable to the WorldView-3 satellite imagery in terms of the variations of cloud distributions and spectral properties of the ground targets in clear-sky regions, suggesting the potential utility of the proposed modeling framework in improving simulation capabilities, as well as testing and improving the operation of image collection processes.

54 ENVIRONMENTAL SCIENCES↗

Data-driven Detection and Early Prediction of Thermoacoustic Instability in a Multi-nozzle Combustor

Thermoacoustic instability (TAI) is a critical issue in modern lean-burn gas-turbine combustors, which is induced by a strong coupling between the resonant combustor acoustics and fluctuations in the heat release rate. This instability may lead to high-amplitude pressure waves that generate undesirable noise levels as well as fatigue stresses in mechanical structures of the combustor. The intense pressure fluctuations due to TAI may also cause large flow perturbations and possibly flow reversal that may lead to flame oscillations, flame liftoff, and even flame blow-out. Hence, there is a strong need for exercising control actions in a timely fashion to mitigate the TAI phenomena. Anomaly detection is an essential prerequisite to the design of a good controller and such a detector must be able to reliably predict a forthcoming TAI. To detect and predict the onset of a TAI from an ensemble of pressure time series, this paper investigates three data-driven methods: Fast Fourier transform (FFT), symbolic time series analysis (STSA), and hidden Markov modeling (HMM). The main focus of the paper is to make a comparative evaluation of these three anomaly detection methods for classification of the current regime of operation into stable and unstable categories as well as for real-time identification of precursors to impending instabilities with short-length time series of measured variables (e.g., pressure oscillations). Finally, the results, generated on experimental data from a multi-nozzle combustor apparatus, have been compared to evaluate the performance of FFT, STSA, and HMM methods for TAI analysis.

42 ENGINEERING↗