Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “recurrence”

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
NASA NTRS2005

Adaptive Filtering Using Recurrent Neural Networks

A method for adaptive (or, optionally, nonadaptive) filtering has been developed for estimating the states of complex process systems (e.g., chemical plants, factories, or manufacturing processes at some level of abstraction) from time series of measurements of system inputs and outputs. The method is based partly on the fundamental principles of the Kalman filter and partly on the use of recurrent neural networks. The standard Kalman filter involves an assumption of linearity of the mathematical model used to describe a process system. The extended Kalman filter accommodates a nonlinear process model but still requires linearization about the state estimate. Both the standard and extended Kalman filters involve the often unrealistic assumption that process and measurement noise are zero-mean, Gaussian, and white. In contrast, the present method does not involve any assumptions of linearity of process models or of the nature of process noise; on the contrary, few (if any) assumptions are made about process models, noise models, or the parameters of such models. In this regard, the method can be characterized as one of nonlinear, nonparametric filtering. The method exploits the unique ability of neural networks to approximate nonlinear functions. In a given case, the process model is limited mainly by limitations of the approximation ability of the neural networks chosen for that case. Moreover, despite the lack of assumptions regarding process noise, the method yields minimum- variance filters. In that they do not require statistical models of noise, the neural- network-based state filters of this method are comparable to conventional nonlinear least-squares estimators.

Parlos, Alexander G.↗
NASA NTRS2004

Algorithm for Training a Recurrent Multilayer Perceptron

An improved algorithm has been devised for training a recurrent multilayer perceptron (RMLP) for optimal performance in predicting the behavior of a complex, dynamic, and noisy system multiple time steps into the future. [An RMLP is a computational neural network with self-feedback and cross-talk (both delayed by one time step) among neurons in hidden layers]. Like other neural-network-training algorithms, this algorithm adjusts network biases and synaptic-connection weights according to a gradient-descent rule. The distinguishing feature of this algorithm is a combination of global feedback (the use of predictions as well as the current output value in computing the gradient at each time step) and recursiveness. The recursive aspect of the algorithm lies in the inclusion of the gradient of predictions at each time step with respect to the predictions at the preceding time step; this recursion enables the RMLP to learn the dynamics. It has been conjectured that carrying the recursion to even earlier time steps would enable the RMLP to represent a noisier, more complex system.

Parlos, Alexander G.↗
NASA NTRS2011

The White Dwarf Mass and the Accretion Rate of Recurrent Novae: An X-ray Perspective

We present recent results of quiescent X-ray observations of recurrent novae (RNe) and related objects. Several RNe are luminous hard X-ray sources in quiescence, consistent with accretion onto a near Chandrasekhar mass white dwarf. Detection of similar hard X-ray emissions in old novae and other cataclysmic variables may lead to identification of additional RN candidates. On the other hand, other RNe are found to be comparatively hard X-ray faint. We present several scenarios that may explain this dichotomy, which should be explored further.

Mukai, Koji↗
NASA NTRS2011

The Spectroscopic Evolution of the Symbiotic-like Recurrent Nova V407 Cygni During Its 2010 Outburst. 2. The Circumstellar Environment and the Aftermath

The nova outburst of V407 Cyg in 2010 Mar. 10 was the first observed for this star but its close resemblance to the well known symbiotic-like recurrent nova RS Oph suggests that it is also a member of this rare type of Galactic novae. The nova was the first detected at gamma-ray energies and is the first known nova explosion for this system. The extensive multiwavelength coverage of this outburst makes it an ideal comparison with the few other outbursts known for similar systems. We extend our previous analysis of the Mira and the expanding shock from the explosion to detail the time development of the photoionized Mira wind, circumstellar medium, and shocked circumstellar environment to derive their physical parameters and how they relate to large scale structure of the environment, extending the previous coverage to more than 500 days after outburst. We use optical spectra obtained at high resolution with the Nordic Optical Telescope (NOT) (R approx. =.45000 to 65000) and medium resolution Ondrejov Observatory (R approx. = 12000) data and compare the line variations with publicly available archival measurements at 30 GHz OVNR and at X-rays with Swift during the first four months of the outburst, through the end of the epoch of strong XR emission. We use nebular diagnostics and high resolution profile variations to derive the densities and locations of the extended emission. We find that the higher the ionization and/or the higher the excitation energy, the more closely the profiles resemble the He II/Ca V-type high velocity shock profile discussed in Paper I. This also accounts for the comparative development of the [N II] and [O III] isoelectronic transitions: the [O III] 4363A profile does not show the low velocity peaks while the excited [N II] 5754A does. If nitrogen is mainly N(+3) or higher in the shock, the upper state of the [N II] nebular lines will contribute but if the oxygen is O(+2) then this line is formed by recombination, masking the nebular contributor, and the lower states are collisionally quenched but emit from the low density surroundings. Absorption lines of Fe-peak ions formed in the Mira wind were visible as P Cyg profiles at low velocity before Day 69, around the time of the X-ray peak and we identified many absorption transitions without accompanying emission for metal lines. The H Balmer lines showed strong P Cyg absorption troughs that weakened during the 2010 observing period, through Day 128. The Fe-peak line profiles and flux variations were different for permitted and forbidden transitions: the E1 transitions were not visible after Day 128 but had shown a narrow peak superimposed on an extended (200 km/s) blue wing, while the M1 and E2 transitions persisted to Day 529, the last observation, and showed extended redshifted wings up of the same velocity. We distinguish the components from the shock, the photoionized environment, and the chromosphere and inner Mira wind using spectra taken more than one year after outburst. The multiple shells and radiative excitation phenomenology are similar to those recently cited for GRBs and SNIa .

Shore, S. N.↗
NASA NTRS2014

The 2011 Outburst of Recurrent Nova T Pyx: X-Ray Observations Expose the White Dwarf Mass and Ejection Dynamics

The recurrent nova T Pyx underwent its sixth historical outburst in 2011, and became the subject of an intensive multi-wavelength observational campaign.We analyze data from the Swift and Suzaku satellites to produce a detailed X-ray light curve augmented by epochs of spectral information. X-ray observations yield mostly non-detections in the first four months of outburst, but both a super-soft and hard X-ray component rise rapidly after Day 115. The super-soft X-ray component, attributable to the photosphere of the nuclear-burning white dwarf, is relatively cool (approximately 45 electron volts) and implies that the white dwarf in T Pyx is significantly below the Chandrasekhar mass (approximately 1 M). The late turn-on time of the super-soft component yields a large nova ejecta mass (approximately greater than 10(exp −5) solar mass), consistent with estimates at other wavelengths. The hard X-ray component is well fit by a approximately 1 kiloelectron volt thermal plasma, and is attributed to shocks internal to the 2011 nova ejecta. The presence of a strong oxygen line in this thermal plasma on Day 194 requires a significantly super-solar abundance of oxygen and implies that the ejecta are polluted by white dwarf material. The X-ray light curve can be explained by a dual-phase ejection, with a significant delay between the first and second ejection phases, and the second ejection finally released two months after outburst. A delayed ejection is consistent with optical and radio observations of T Pyx, but the physical mechanism producing such a delay remains a mystery.

nova T Pyx↗
NASA NTRS2016

Assessing Spontaneous Combustion Instability with Recurrence Quantification Analysis

Spontaneous instabilities can pose a significant challenge to verification of combustion stability, and characterizing its onset is an important avenue of improvement for stability assessments of liquid propellant rocket engines. Recurrence Quantification Analysis (RQA) is used here to explore nonlinear combustion dynamics that might give insight into instability. Multiple types of patterns representative of different dynamical states are identified within fluctuating chamber pressure data, and markers for impending instability are found. A class of metrics which describe these patterns is also calculated. RQA metrics are compared with and interpreted against another metric from nonlinear time series analysis, the Hurst exponent, to help better distinguish between stable and unstable operation.

Eberhart, Chad J.↗
NASA NTRS2018

Automatic Speech Recognition for Launch Control Center Communication Using Recurrent Neural Networks with Data Augmentation and Custom Language Model

Transcribing voice communications in NASA’s launch control center is important for information utilization. However, automatic speech recognition in this environment is particularly challenging due to the lack of training data, unfamiliar words in acronyms, multiple different speakers and accents, and conversational characteristics of speaking. We used bidirectional deep recurrent neural networks to train and test speech recognition performance. We showed that data augmentation and custom language models can improve speech recognition accuracy. Transcribing communications from the launch control center will help the machine analyze information and accelerate knowledge generation.

Chow, Edward↗
DOE OSTI2020

Physics-constrained Deep Recurrent Neural Models of Building Thermal Dynamics

We develop physics-constrained and control-oriented predictive deep learning models for the thermal dynamics of a real-world commercial office building. The proposed method is based on the systematic encoding of physics-based prior knowledge into a structured recurrent neural architecture. Specifically, our model mimics the structure of the building thermal dynamics model and leverages penalty methods to model inequality constraints. Additionally, we use constrained matrix parameterization based on the Perron-Frobenius theorem to bound the eigenvalues of the learned network weights. We interpret the stable eigenvalues as dissipativeness of the learned building thermal model. We demonstrate the effectiveness of the proposed approach on a dataset obtained from an office building with $20$ thermal zones.

Building Energy, structured neural networks↗
DOE OSTI2025

Conflict Detection in Open RAN with Recurrent Neural Networks Using Geometric Manifolds

Allowing third-party applications on Radio Access Network (RAN) Intelligent Controllers (RICs) within the OpenRAN (O-RAN) framework introduces conflicting interactions that are often difficult to detect in advance. These conflicts, occurring between third-party applications in the Near RealTime RIC (Near-RT RIC), known as xApps, can lead to performance degradation and instability in O-RAN if not identified early. Existing conflict detection and mitigation solutions in the literature assume that the conflicts are known beforehand, which is not always accurate due to the complex and often hidden relationships between control parameters and Key Performance Indicators (KPIs). In this paper, we propose a novel Recurrent Neural Network (RNN) to detect both known and unknown conflicts in O-RAN xApps as specified in the O-RAN standards. We model the xApps, control parameters, and KPIs with nodes and edges to create graph structures and use the hidden nonEuclidean geometric properties of the Riemannian manifold to train the RNN model. The performance of this proposed model is validated using evaluation metrics and compared with benchmarks. Results demonstrate that the proposed RNN model, leveraging Riemannian geometric properties, can achieve 100% of the F1-score provided by an optimal solution in just 20 iterations.

5G↗
DOE OSTI2025

Conflict Detection in Open RAN with Recurrent Neural Networks Using Geometric Manifolds

Allowing third-party applications on Radio Access Network (RAN) Intelligent Controllers (RICs) within the OpenRAN (O-RAN) framework introduces conflicting interactions that are often difficult to detect in advance. These conflicts, occurring between third-party applications in the Near RealTime RIC (Near-RT RIC), known as xApps, can lead to performance degradation and instability in O-RAN if not identified early. Existing conflict detection and mitigation solutions in the literature assume that the conflicts are known beforehand, which is not always accurate due to the complex and often hidden relationships between control parameters and Key Performance Indicators (KPIs). In this paper, we propose a novel Recurrent Neural Network (RNN) to detect both known and unknown conflicts in O-RAN xApps as specified in the O-RAN standards. We model the xApps, control parameters, and KPIs with nodes and edges to create graph structures and use the hidden nonEuclidean geometric properties of the Riemannian manifold to train the RNN model. The performance of this proposed model is validated using evaluation metrics and compared with benchmarks. Results demonstrate that the proposed RNN model, leveraging Riemannian geometric properties, can achieve 100% of the F1-score provided by an optimal solution in just 20 iterations.

5G↗
DOE OSTI2022

Data-driven modeling of coarse mesh turbulence for reactor transient analysis using convolutional recurrent neural networks

Advanced nuclear reactors often exhibit complex thermal-fluid phenomena during transients. To accurately capture such phenomena, a coarse-mesh three-dimensional (3-D) modeling capability is desired for modern nuclear-system code. In the coarse-mesh 3-D modeling of advanced-reactor transients that involve flow and heat transfer, accurately predicting the turbulent viscosity is a challenging task that requires an accurate and computationally efficient model to capture the unresolved fine-scale turbulence. In this work, we propose a data-driven coarse-mesh turbulence model based on local flow features for the transient analysis of thermal mixing and stratification in a sodium-cooled fast reactor. The model has a coarse mesh setup to ensure computational efficiency, while it is trained by fine-mesh computational fluid dynamics (CFD) data to ensure accuracy. A novel neural network architecture, combining a densely connected convolutional network and a long-short-term-memory network, is developed that can efficiently learn from the spatial temporal CFD transient simulation results. The neural network model was trained and optimized on a loss-of flow transient and demonstrated high accuracy in predicting the turbulent viscosity field during the whole transient. The trained model's generalization capability was also investigated on two other transients with different inlet conditions. The study demonstrates the potential of applying the proposed data-driven approach to support the coarse-mesh multi-dimensional modeling of advanced reactors.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗
DOE OSTI2022

Hybrid Recurrent Neural Network Modeling for Traffic Delay Prediction at Signalized Intersections Along an Urban Arterial

This paper studies the traffic delay prediction modeling for multiple signalized intersections along the Ala Moana Boulevard and Nimitz Highway in Hawaii. Several machine learning (ML) based approaches have been studied in the literature, and most of them focused on prediction accuracy rather than the end use of real-time control and implementation. These ML models tend to be very complex and non-linear in nature, making it challenging to achieve fast inferences and are computationally heavy for real-time signal control implementation. As such, in this paper, a simple yet accurate hybrid modeling method is proposed to predict traffic delay one-step ahead with the model made suitable for real-time implementation to control traffic flow. Since real-time road-side measurements are recorded in unstructured form, the paper also discusses other issues related to data extraction and the pre-processing process. Finally, a simple signal control loop is developed to demonstrate the proposed modeling approach, which has shown advantages in model accuracy and computation efficiency compared against several existing modeling methods.

42 ENGINEERING↗
DOE OSTI2025

Recurrent features of amplitudes in planar $\mathcal{N}$ = 4 super Yang-Mills theory

The planar three-gluon form factor for the chiral stress tensor operator in planar maximally supersymmetric Yang-Mills theory is an analog of the Higgs-to-three-gluon scattering amplitude in QCD. The amplitude (symbol) bootstrap program has provided a wealth of high-loop perturbative data about this form factor, with results up to eight loops available. The symbol of the form factor at L loops is given by words of length 2L in six letters with associated integer coefficients. In this paper, we analyze this data, describing patterns of zero coefficients and relations between coefficients. We find many sequences of words whose coefficients are given by closed-form expressions which we expect to be valid at any loop order. Moreover, motivated by our previous machine-learning analysis, we identify simple recursion relations that relate the coefficient of a word to the coefficients of particular lower-loop words. These results open an exciting door for understanding scattering amplitudes at all loop orders.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗