Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Neural Network”

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 721 records · Page 40

Identification and control of plasma vertical position using neural network in Damavand tokamak

In this work, a nonlinear model is introduced to determine the vertical position of the plasma column in Damavand tokamak. Using this model as a simulator, a nonlinear neural network controller has been designed. In the first stage, the electronic drive and sensory circuits of Damavand tokamak are modified. These circuits can control the vertical position of the plasma column inside the vacuum vessel. Since the vertical position of plasma is an unstable parameter, a direct closed loop system identification algorithm is performed. In the second stage, a nonlinear model is identified for plasma vertical position, based on the multilayer perceptron (MLP) neural network (NN) structure. Estimation of simulator parameters has been performed by back-propagation error algorithm using Levenberg-Marquardt gradient descent optimization technique. The model is verified through simulation of the whole closed loop system using both simulator and actual plant in similar conditions. As the final stage, a MLP neural network controller is designed for simulator model. In the last step, online training is performed to tune the controller parameters. Simulation results justify using of the NN controller for the actual plant.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

Neural Network Based Representation of UH-60A Pilot and Hub Accelerations

Neural network relationships between the full-scale, experimental hub accelerations and the corresponding pilot floor vertical vibration are studied. The present physics-based, quantitative effort represents an initial systematic study on the UH-60A Black Hawk hub accelerations. The NASA/Army UH-60A Airloads Program flight test database was used. A 'maneuver-effect-factor (MEF)', derived using the roll-angle and the pitch-rate, was used. Three neural network based representation-cases were considered. The pilot floor vertical vibration was considered in the first case and the hub accelerations were separately considered in the second case. The third case considered both the hub accelerations and the pilot floor vertical vibration. Neither the advance ratio nor the gross weight alone could be used to predict the pilot floor vertical vibration. However, the advance ratio and the gross weight together could be used to predict the pilot floor vertical vibration over the entire flight envelope. The hub accelerations data were modeled and found to be of very acceptable quality. The hub accelerations alone could not be used to predict the pilot floor vertical vibration. Thus, the hub accelerations alone do not drive the pilot floor vertical vibration. However, the hub accelerations, along with either the advance ratio or the gross weight or both, could be used to satisfactorily predict the pilot floor vertical vibration. The hub accelerations are clearly a factor in determining the pilot floor vertical vibration.

Kottapalli, Sesi↗

Structured Neural Network Modeling for Developing Digital Twins Models of Hydropower Generation Units

Dynamic modeling is a key part in the development of digital twin (DT) for dynamic systems. This is true for hydropower systems, where whole system modeling including penstock, turbine and generators, etc is important in realizing actuate modeling for the real systems. On the other hand, in response to the large variations of the power demand due to increased penetration of renewables such as wind and solar, hydropower systems are now required to operate in a large power generation range. This situation triggers the nonlinear characteristics of the generation unit with respect to its models. As such, it is imperative to use data driven modeling such as neural networks to learn the nonlinear dynamics of the hydropower generation unit. To achieve this objective, this study constructs a modeling and learning algorithm integrated with multiple structured neural network models for the modeling of turbine shaft speed, penstock pressure, and generator power output based on the generator power control setpoint, field current, and field voltage. In addition, the study uses the hydropower data from Tacoma Public Utilities to train and validate the proposed neural network algorithm. The results have shown that this structured neural network modeling approach can learn the system dynamics effectively by using the real-time data collected from the hydropower system with the desired modeling results.

Wang, Hong↗

64x64 Analog Input Array for 3-Dimensional Neural Network Processor

In pattern recognition and classification for spatio-temporal problems, one of the most challenging tasks is to provide a good and valid solution in real-time. Because of time constraints, software-based neural network approaches may not be suitable for practical use. Hardware solutions seem to be good candidates for this class of problems. Currently, the Three Dimensional Analog Neural Network (3-DANN) is an effective approach to solving spatio-temporal problems in three-dimensional hardware.

spatio-temporal 3-DANN Three Dimensional Analog Ne↗

An adaptive and stability-promoting layerwise training approach for sparse deep neural network architecture

This work presents a two-stage adaptive framework for progressively developing deep neural network (DNN) architectures that generalize well for a given training data set. In the first stage, a layerwise training approach is adopted where a new layer is added each time and trained independently by freezing parameters in the previous layers. We impose desirable structures on the DNN by employing manifold regularization, sparsity regularization, and physics-informed terms. We introduce a ε – δ – stability-promoting concept as a desirable property for a learning algorithm and show that employing manifold regularization yields a ε – δ stability-promoting algorithm. Further, we also derive the necessary conditions for the trainability of a newly added layer and investigate the training saturation problem. In the second stage of the algorithm (post-processing), a sequence of shallow networks is employed to extract information from the residual produced in the first stage, thereby improving the prediction accuracy. Numerical investigations on prototype regression and classification problems demonstrate that the proposed approach can outperform fully connected DNNs of the same size. Moreover, by equipping the physics-informed neural network (PINN) with the proposed adaptive architecture strategy to solve partial differential equations, we numerically show that adaptive PINNs not only are superior to standard PINNs but also produce interpretable hidden layers with provable stability. As a result, we also apply our architecture design strategy to solve inverse problems governed by elliptic partial differential equations.

42 ENGINEERING↗

Neural network recognition of chemical class information in mobility spectra obtained at high temperatures

A minimal neural network was applied to a large library of high-temperature mobility spectra drawn from 16 chemical classes including 154 substances with 2000 spectra at various concentrations. A genetic algorithm was used to create a representative subset of points from the mobility spectrum as input to a cascade-type back-propagation network. This network demonstrated that significant information specific to chemical class was located in the spectral region near the reactant ions. This network failed to generalize the solution to unfamiliar compounds necessitating the use of complete spectra in network processing. An extended back-propagation network classified unfamiliar chemicals by functional group with a mean for average values of 0.83 without sulfides and 0.79 with sulfides. Further experiments confirmed that chemical class information was resident in the spectral region near the reactant ions. Deconvolution of spectra demonstrated the presence of ions, merged with the reactant ion peaks that originated from introduced samples. The ability of the neural network to generalize the solution to unfamiliar compounds suggests that these ions are distinct and class specific.

NASA Discipline Environmental Health↗

Multifidelity Neural Network Formulations for Prediction of Reactive Molecular Potential Energy Surfaces

Here, this paper focuses on the development of multifidelity modeling approaches using neural network surrogates, where training data arising from multiple model forms and resolutions are integrated to predict high-fidelity response quantities of interest at lower cost. We focus on the context of quantum chemistry and the integration of information from multiple levels of theory. Important foundations include the use of symmetry function-based atomic energy vector constructions as feature vectors for representing structures across families of molecules and single-fidelity neural network training capabilities that learn the relationships needed to map feature vectors to potential energy predictions. These foundations are embedded within several multifidelity topologies that decompose the high-fidelity mapping into model-based components, including sequential formulations that admit a general nonlinear mapping across fidelities and discrepancy-based formulations that presume an additive decomposition. Methodologies are first explored and demonstrated on a pair of simple analytical test problems and then deployed for potential energy prediction for C 5 H 5 using B2PLYP-D3/6-311++G(d,p) for high-fidelity simulation data and Hartree–Fock 6-31G for low-fidelity data. For the common case of limited access to high-fidelity data, our computational results demonstrate that multifidelity neural network potential energy surface constructions achieve roughly an order of magnitude improvement, either in terms of test error reduction for equivalent total simulation cost or reduction in total cost for equivalent error.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

SuperCDMS Event Reconstruction Using Convolutional Neural Networks [Thesis]

The SuperCDMS experiment uses cryogenic silicon and germanium detectors to search for dark matter candidates such as WIMPs (Weakly Interacting Massive Particles) streaming through the Earth. Collisions in the silicon and germanium crystals are expected to produce phonons whose thermal signatures can be measured. This thesis first describes the integration of a new Signal Distribution Unit (SDU) to the SuperCDMS data acquisition system, which allows for synchronization of multiple detectors and electronic/mechanical noise characterization via accelerometer, antenna, and AC phase measurements. From SuperCDMS detector data it is necessary to reconstruct the energies of the particle events. This thesis explores the use of Convolutional Neural Networks (CNNs) to perform this reconstruction and finds that, although they perform well, changing the noise model breaks the model and requires the neural network to be retrained. In order to mitigate this issue, a new CNN model is proposed which includes the noise Power Spectral Density (PSD) of the data as an additional input to the CNN. While it proves to be effective as a denoising algorithm, it still fails for data with a different noise model. However, including data from multiple PSDs in the neural network training sample allows it to handle data with different types of noise while still maintaining the quality of the reconstruction. Nevertheless, neural networks trained even on multiple PSDs do not robustly handle data taken with PSDs dissimilar to those in the training sample, suggesting that CNNs may need to be retrained whenever the noise environment changes in a significant way.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

Neural Network Approaches for Mobile Spectroscopic Gamma-Ray Source Detection

Artificial neural networks (ANNs) for performing spectroscopic gamma-ray source identification have been previously introduced, primarily for applications in controlled laboratory settings. To understand the utility of these methods in scenarios and environments more relevant to nuclear safety and security, this work examines the use of ANNs for mobile detection, which involves highly variable gamma-ray background, low signal-to-noise ratio measurements, and low false alarm rates. Simulated data from a 2” × 4” × 16” NaI(Tl) detector are used in this work for demonstrating these concepts, and the minimum detectable activity (MDA) is used as a performance metric in assessing model performance.In addition to examining simultaneous detection and identification, binary spectral anomaly detection using autoencoders is introduced in this work, and benchmarked using detection methods based on Non-negative Matrix Factorization (NMF) and Principal Component Analysis (PCA). On average, the autoencoder provides a 12% and 23% improvement over NMF- and PCA-based detection methods, respectively. Additionally, source identification using ANNs is extended to leverage temporal dynamics by means of recurrent neural networks, and these time-dependent models outperform their time-independent counterparts by 17% for the analysis examined here. The paper concludes with a discussion on tradeoffs between the ANN-based approaches and the benchmark methods examined here.

Bilton, Kyle J. (ORCID:0000000184553689)↗

Accelerating Binarized Neural Networks via Bit-Tensor-Cores in Turing GPUs

Despite foreseeing tremendous speedups over conventional deep neural networks, the performance advantage of binarized neural networks (BNNs) has merely been showcased on general-purpose processors such as CPUs and GPUs. In fact, due to being unable to leverage bit-level-parallelism with a word-based architecture, GPUs have been criticized for extremely low utilization (1%) when executing BNNs. Consequently, the latest tensorcores in NVIDIA Turing GPUs start to experimentally support bit computation. In this work, we look into this brand new bit computation capability and characterize its unique features. We show that the stride of memory access can significantly affect performance delivery and a data-format co-design is highly desired to support the tensorcores for achieving superior performance than existing software solutions without tensorcores. We realize the tensorcore-accelerated BNN design, particularly the major functions for fully-connect and convolution layers — bit matrix multiplication and bit convolution. Evaluations on two NVIDIA Turing GPUs show that, with ResNet-18, our BTC-BNN design can process ImageNet at a rate of 5.6K images per second, 77% faster than state-of-the-art. Our BNN approach is released on https://github.com/pnnl/TCBNN.

Li, Ang↗

Livermore Big Artificial Neural Network Toolkit

LBANN is a toolkit that is designed to train artificial neural networks efficiently on high performance computing architectures. It is optimized to take advantages of key High Performance Computing features to accelerate neural network training. Specifically it is optimized for low-latency, high bandwidth interconnects, node-local NVRAM, node-local GPU accelerators, and high bandwidth parallel file systems. It is built on top of the open source Elemental distributed-memory dense and spars-direct linear algebra and optimization library that is released under the BSD license. The algorithms contained within LBANN are drawn from the academic literature and implemented to work within a distributed-memory framework.

Essen, BrianVan↗

Livermore Big Artificial Neural Network Toolkit

LBANN is a toolkit that is designed to train artificial neural networks efficiently on high performance computing architectures. It is optimized to take advantages of key High Performance Computing features to accelerate neural network training. Specifically it is optimized for low-latency, high bandwidth interconnects, node-local NVRAM, node-local GPU accelerators, and high bandwidth parallel file systems. It is built on top of the open source Elemental distributed-memory dense and spars-direct linear algebra and optimization library that is released under the BSD license. The algorithms contained within LBANN are drawn from the academic literature and implemented to work within a distributed-memory framework.

Essen, Brian Van↗

Search for standard model Higgs boson production in association with a W boson using a neural network discriminant at CDF

We present a search for standard model Higgs boson production in association with a W boson in proton-antiproton collisions (pp{yields}W{sup {+-}}H{yields}l{nu}bb) at a center of mass energy of 1.96 TeV. The search employs data collected with the CDF II detector that correspond to an integrated luminosity of approximately 1.9 fb{sup -1}. We select events consistent with a signature of a single charged lepton (e{sup {+-}}/{mu}{sup {+-}}), missing transverse energy, and two jets. Jets corresponding to bottom quarks are identified with a secondary vertex tagging method, a jet probability tagging method, and a neural network filter. We use kinematic information in an artificial neural network to improve discrimination between signal and background compared to previous analyses. The observed number of events and the neural network output distributions are consistent with the standard model background expectations, and we set 95% confidence level upper limits on the production cross section times branching fraction ranging from 1.2 to 1.1 pb or 7.5 to 102 times the standard model expectation for Higgs boson masses from 110 to 150 GeV/c{sup 2}, respectively.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

Search for Standard Model Higgs Boson Production in Association with a W Boson using a Neural Network

We present a search for standard model Higgs boson production in association with a W boson in proton-antiproton collisions (p{bar p} {yields} W{sup {+-}}H {yields} {ell}{nu}b{bar b}) at a center of mass energy of 1.96 TeV. The search employs data collected with the CDF II detector that correspond to an integrated luminosity of approximately 1.9 fb{sup -1}. We select events consistent with a signature of a single charged lepton (e{sup {+-}}/{mu}{sup {+-}}), missing transverse energy, and two jets. Jets corresponding to bottom quarks are identified with a secondary vertex tagging method, a jet probability tagging method, and a neural network filter. We use kinematic information in an artificial neural network to improve discrimination between signal and background compared to previous analyses. The observed number of events and the neural network output distributions are consistent with the standard model background expectations, and we set 95% confidence level upper limits on the production cross section times branching fraction ranging from 1.2 to 1.1 pb or 7.5 to 102 times the standard model expectation for Higgs boson masses from 110 to 150 GeV/c{sup 2}, respectively.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

Use of neural networks to analyze pulse shape data in low-background detectors

Pacific Northwest National Laboratory has accumulated years of data with ultra-low-background proportional counters collected in an on-site shallow underground laboratory. This large dataset of events is exploited to study the impact of using neural networks for data analysis compared to simple pulse shape discrimination (PSD). The PSD method can introduce false positives for overlapping event distributions; however, a neural network can separate and correctly classify these events. Here, this paper describes the training, testing, and validation of a neutral network, analysis of challenge datasets, and a comparison between the standard PSD approach and a dense, fully-connected neural network.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

A data-driven CO 2 leakage detection using seismic data and spatial–temporal densely connected convolutional neural networks

In carbon capture and sequestration (also known as carbon capture and storage, or CCS), developing effective monitoring methods is needed to detect and respond to CO 2 leakage. CO 2 leakage detection methods rely on geophysical observations and monitoring sensor network. However, traditional methods usually require the development of site-specific physical models and expert interpretation, and the effectiveness of these methods can be limited to different application locations, operational scenarios, and conditions. Here, we developed a novel data-driven leakage detection method based on densely connected convolutional neural networks. Our method is an end-to-end detection approach, that differs from conventional leakage monitoring methods by directly learning a mapping relationship between seismic data and the CO 2 leakage mass. To account for the spatial and temporal characteristics of seismic data, our novel networks architecture combines 1D and 2D convolutional neural networks. To overcome the computational expense of solving optimization problems, we apply a densely-connecting strategy in our network architecture that reduces the number of network parameters. Based on the features generated by our convolutional neural networks, we further incorporate a long short-term memory network to utilize time-sequential information, which further improves the detection accuracy. Lastly, we employ our detection method to synthetic seismic datasets generated based on flow simulations of a hypothetical CO 2 storage scenario with injection into a partially compartmentalized sandstone storage reservoir. To evaluate method performance, we conducted multiple experiments including a random leakage test, a sequential test, and a robustness test. Numerical results show that our CO 2 leakage detection method successfully detects the leakage and accurately predicts the leakage mass, suggesting that it has the potential for application in monitoring of real CO 2 storage sites.

58 GEOSCIENCES↗

Automated target recognition and tracking using an optical pattern recognition neural network

The on-going development of an automatic target recognition and tracking system at the Jet Propulsion Laboratory is presented. This system is an optical pattern recognition neural network (OPRNN) that is an integration of an innovative optical parallel processor and a feature extraction based neural net training algorithm. The parallel optical processor provides high speed and vast parallelism as well as full shift invariance. The neural network algorithm enables simultaneous discrimination of multiple noisy targets in spite of their scales, rotations, perspectives, and various deformations. This fully developed OPRNN system can be effectively utilized for the automated spacecraft recognition and tracking that will lead to success in the Automated Rendezvous and Capture (AR&C) of the unmanned Cargo Transfer Vehicle (CTV). One of the most powerful optical parallel processors for automatic target recognition is the multichannel correlator. With the inherent advantages of parallel processing capability and shift invariance, multiple objects can be simultaneously recognized and tracked using this multichannel correlator. This target tracking capability can be greatly enhanced by utilizing a powerful feature extraction based neural network training algorithm such as the neocognitron. The OPRNN, currently under investigation at JPL, is constructed with an optical multichannel correlator where holographic filters have been prepared using the neocognitron training algorithm. The computation speed of the neocognitron-type OPRNN is up to 10(exp 14) analog connections/sec that enabling the OPRNN to outperform its state-of-the-art electronics counterpart by at least two orders of magnitude.

Chao, Tien-Hsin↗

Neural network based architectures for aerospace applications

A brief history of the field of neural networks research is given and some simple concepts are described. In addition, some neural network based avionics research and development programs are reviewed. The need for the United States Air Force and NASA to assume a leadership role in supporting this technology is stressed.

Ricart, Richard↗