SEARCH · Engineering Papers

Results for “ITERATIVE 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.

SourceResearch connectionFrom yearThrough year

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 217 records · Page 12

All sources NASA NTRS DOE OSTI

DOE OSTI2017

Visualization for Classification in Deep Neural Networks

Recently, the techniques based on Deep Neural Networks (DNNs) have achieved a great performance in classification tasks in a wide range of applications, such as image recognition and natural language processing. However, DNN developers face a lot of trial and error during the development process and spend their efforts in debugging their network model through analyzing and understanding the classification results. As such, tools are needed that help the developers not only understand the results, but also suggest the ways to improve their model. In this paper, we propose a visual analytics tool for visualizing the classification results during the iterative development pipeline of a DNN model. Our tool enables exploring the classification results from any type of neural network models, identifying misclassified samples, examining the predicted score distributions of samples, and showing how the outcomes progressively change during the training process.

Chae, Junghoon↗

DOE OSTI2023

Deep learning workflow for the inverse design of molecules with specific optoelectronic properties

The inverse design of novel molecules with a desirable optoelectronic property requires consideration of the vast chemical spaces associated with varying chemical composition and molecular size. First principles-based property predictions have become increasingly helpful for assisting the selection of promising candidate chemical species for subsequent experimental validation. However, a brute-force computational screening of the entire chemical space is decidedly impossible. To alleviate the computational burden and accelerate rational molecular design, we here present an iterative deep learning workflow that combines (i) the density-functional tight-binding method for dynamic generation of property training data, (ii) a graph convolutional neural network surrogate model for rapid and reliable predictions of chemical and physical properties, and (iii) a masked language model. As proof of principle, we employ our workflow in the iterative generation of novel molecules with a target energy gap between the highest occupied molecular orbital (HOMO) and the lowest unoccupied molecular orbital (LUMO).

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2022

Deep-learning-guided high-resolution subsurface reflectivity imaging with application to ground-penetrating radar data

Subsurface reflectivity imaging is one of the most important geophysical characterization methods for revealing subsurface structures. In many cases, accurate subsurface reflectivity imaging is challenging because of, for example, random or coherent noise in the data and sparse source-receiver observation geometry. Here, we develop a deep-learning-guided iterative imaging method to improve subsurface structure imaging. Specifically, we train a supervised neural network to infer a noise-free, high-resolution image from a noisy, low-resolution image and use this estimated image as guidance to regularize least-squares imaging. We develop a systematic method to generate high-quality synthetic training data (data-label pairs) to train the guidance neural network. The trained neural network can provide high-fidelity predictions even for field-data images that are not in the training data. We validate our new imaging method using one synthetic and two field ground-penetrating radar data examples, and find that our method can produce clean, high-resolution subsurface reflectivity images where existing single-pass and least-squares imaging methods fail due to noise and insufficient data coverage.

58 GEOSCIENCES↗

DOE OSTI2021

Neural Lyapunov Control for Power System Transient Stability: A Deep Learning-Based Approach

We report that power system control and transient stability analysis play essential roles in secure system operation. Control of power systems typically involves highly nonlinear and complex dynamics. Most of the existing works address such problems with additional assumptions in system dynamics, leading to a requirement for a complete and general solution. This paper, therefore, proposes a novel control framework for various power system control and stability problems leveraging a learning-based approach. The proposed framework includes a two-module structure that iteratively and jointly learns the candidate Lyapunov function and control law via deep neural networks in a learning module. Meanwhile, it guides the learning procedure towards valid results satisfying Lyapunov conditions in a falsification module. The introduced termination criteria ensure provable system stability. This control framework is verified through several studies handling different types of power system control problems. The results show that the proposed framework is generalizable and can simplify the control design for complex power systems with the stability guarantee and enlarged region of attraction.

24 POWER TRANSMISSION AND DISTRIBUTION↗

NASA NTRS1992

Neural network error correction for solving coupled ordinary differential equations

A neural network is presented to learn errors generated by a numerical algorithm for solving coupled nonlinear differential equations. The method is based on using a neural network to correctly learn the error generated by, for example, Runge-Kutta on a model molecular dynamics (MD) problem. The neural network programs used in this study were developed by NASA. Comparisons are made for training the neural network using backpropagation and a new method which was found to converge with fewer iterations. The neural net programs, the MD model and the calculations are discussed.

Shelton, R. O.↗

NASA NTRS2001

A Global Optimization Methodology for Rocket Propulsion Applications

While the response surface method is an effective method in engineering optimization, its accuracy is often affected by the use of limited amount of data points for model construction. In this chapter, the issues related to the accuracy of the RS approximations and possible ways of improving the RS model using appropriate treatments, including the iteratively re-weighted least square (IRLS) technique and the radial-basis neural networks, are investigated. A main interest is to identify ways to offer added capabilities for the RS method to be able to at least selectively improve the accuracy in regions of importance. An example is to target the high efficiency region of a fluid machinery design space so that the predictive power of the RS can be maximized when it matters most. Analytical models based on polynomials, with controlled level of noise, are used to assess the performance of these techniques.

Source record↗

DOE OSTI2022

Modeling Integrated Electricity and Natural Gas Systems Using Co-Simulation

One approach for modeling natural gas and electricity networks as integrated systems is co-simulation, which involves separating a single model into subproblems that are solved iteratively, with coupling information passed between then. Although a fully integrated problem might be more direct, co-simulation offers advantages in that it can sometimes be more computationally tractable, and can also leverage existing simulation tools to represent the subproblems. Here we present a framework for co-simulation of natural gas and electricity networks using the HELICS tool. We provide validation of the co-simulation approach by comparing results with a fully integrated approach that uses the same simulation tool, finding that the two approaches provide identical results for a range of test cases. We then present results from natural gas and electricity network co-simulation case study of the Colorado Front Range, with an evaluation of the present-day system and also a future system with higher shares of renewable power. The presentation will close with a discussion on future avenues of work for co-simulation of gas and power networks.

co-simulation↗

DOE OSTI2019

Mapping Arbitrarily Sparse Two-body Interactions on One-dimensional Quantum Circuits

We consider an assignment problem arising in Fermionic-swap based mapping of the one-body and two-body interaction terms in simulating time evolution of a sparse second-quantized electronic structure Hamiltonian on a quantum computer. Relative efficiency of different swap networks depends on the relative costs of performing a swap layer and computing a Hamiltonian interaction term. Under the assumption that the interaction term cost dominates the computation, we develop an iterative algorithm that uses hypergraph optimal linear arrangement (HOLA) and partial distance-2 coloring to exploit arbitrary sparsity in the Hamiltonian for efficient computation using swap networks. Using a set of 122 problems from computational chemistry, we demonstrate performance improvements up to 100% relative to the state-of-the-art approaches for one-body terms and up to 86% utilization for two-body terms relative to a theoretical peak utilization. To the best of our knowledge, this is the first study to exploit arbitrary sparsity in orbital interactions for efficient computation on one-dimensional qubit connectivity layouts. The proposed algorithms form the basis for extension to map general k-body interactions that arise in other domains onto generalized qubit connectivity layouts.

quantum computing, Combinatorial Optimization, hyp↗

DOE OSTI2022

An MILP-Based Distributed Energy Management for Coordination of Networked Microgrids

An MILP-based distributed energy management for the coordination of networked microgrids is proposed in this paper. Multiple microgrids and the utility grid are coordinated through iteratively adjusted price signals. Based on the price signals received, the microgrid controllers (MCs) and distribution management system (DMS) update their schedules separately. Then, the price signals are updated according to the generation–load mismatch and distributed to MCs and DMS for the next iteration. The iteration continues until the generation–load mismatch is small enough, i.e., the generation and load are balanced under agreed price signals. Through the proposed distributed energy management, various microgrids and the utility grid with different economic, resilient, emission and socio-economic objectives are coordinated with generation–load balance guaranteed and the microgrid customers’ privacy preserved. In particular, a piecewise linearization technique is employed to approximate the augmented Lagrange term in the alternating direction method of multipliers (ADMM) algorithm. Thus, the subproblems are transformed into mixed integer linear programming (MILP) problems and efficiently solved by open-source MILP solvers, which would accelerate the adoption and deployment of microgrids and promote clean energy. The proposed MILP-based distributed energy management is demonstrated through various case studies on a networked microgrids test system with three microgrids.

24 POWER TRANSMISSION AND DISTRIBUTION↗

NASA NTRS1989

Monte Carlo simulations of the diurnal variation in seismic detection rate of sporadic meteoroid impacts on the moon

The rate of detection of meteoroid impacts on the moon by the lunar seismic network shows a characteristic diurnal variation. Assuming that these meteoroids have a flux and a preimpact orbital distribution similar to that of fireballs observed by terrestrial camera networks, one can compute the expected diurnal variation for a given set of parameters that describe the seismic wave generation and transmission on the moon. An iterative process to match the theoretical variation with the observed one has led us to the following results: (1) the majority of the detected impact events occur within a closer range of the network than was believed earlier. This results in higher meteoroid flux estimates from lunar seismic data that agree with the terrestrially measured flux. (2) For meteoroid masses smaller than 1000 g, seismic amplitude is approximately proportional to the one-fifth power of the impact speed; for larger masses it is approximately proportional to the eight-fifth power, provided that the terrestrial meteor data used for analysis are not biased. (3) Seismic efficiency of meteoroids smaller than 1000 g is significantly less than that of large meteoroids. (4) Using orbits of fireballs that represent meteorites, we predict that the share of meteorites among the detected impacts is approximately 15 percent assuming that seismic efficiency of the high-density meteorites is the same as that of average meteoroids. A greatly increased seismic efficiency for these high-density objects is not likely.

Oberst, J.↗

DOE OSTI2016

Development of an Ultrasonic Phased Array System for Wellbore Integrity Evaluation and Near-Wellbore Fracture Network Mapping of Injection and Production Wells in Geothermal Energy Systems

This paper documents our progress developing an ultrasound phased array system in combination with a model-based iterative reconstruction (MBIR) algorithm to inspect the health of and characterize the composition of the near-wellbore region for geothermal reservoirs. The main goal for this system is to provide a near-wellbore in-situ characterization capability that will significantly improve wellbore integrity evaluation and near well-bore fracture network mapping. A more detailed image of the fracture network near the wellbore in particular will enable the selection of optimal locations for stimulation along the wellbore, provide critical data that can be used to improve stimulation design, and provide a means for measuring evolution of the fracture network to support long term management of reservoir operations. Development of such a measurement capability supports current hydrothermal operations as well as the successful demonstration of Engineered Geothermal Systems (EGS). The paper will include the design of the phased array system, the performance specifications, and characterization methodology. In addition, we will describe the MBIR forward model derived for the phased array system and the propagation of compressional waves through a pseudo-homogenous medium.

MBIR↗

DOE OSTI2023

Radionuclide measurements of the international monitoring system

The International Monitoring System (IMS) is a unique global network of sensors, tuned to measure various phenomenology, with the common goal of detecting a nuclear explosion anywhere in the world. One component of this network collects measurements of radioactive particulates and gases (collectively known as radionuclides) present in the atmosphere; through this, compliance with the Comprehensive Nuclear-Test-Ban Treaty (CTBT) can be verified. The radionuclide sub-network consists of 120 sensors across 80 locations, supported by 16 measurement laboratories. All radionuclide stations make use of a form of γ-ray spectroscopy to measure radionuclides from samples; this remains largely unchanged since the network was first established 25 years ago. Advances in sampling and spectroscopy systems can yield improvements to the sensitivity of the network to detect a nuclear explosion. This paper summarises the status of the IMS radionuclide network, the current suite of technology used and reviews new technology that could enhance future iterations, potentially improving the verification power of the IMS.

Goodwin, Matthew A.↗

DOE OSTI2017

Target tracking via recursive Bayesian state estimation in radar networks

Modern cognitive radar networks incorporating intelligent and cognitive support-modules can actively adjust the radar-target geometry and optimally select a subset of radars to track the target of interest. Based on the theories of dynamic graphical models (DGM) and recursive Bayesian state estimation (RBSE), we propose a framework for single target tracking in mobile and cooperative radar networks, jointly considering path planning and radar selection. We formulate the tracking procedure as two iterative steps: (i) solving a combinatorial problem based on the expected cross-entropy measure to select the optimal subset of radars and their locations, and (ii) tracking the target using RBSE technique. We simulate the proposed framework using an illustrative example in 2-D space and demonstrate the tracking performance.

Xiang, Yijian↗

NASA NTRS1992

Accelerating Learning By Neural Networks

Electronic neural networks made to learn faster by use of terminal teacher forcing. Method of supervised learning involves addition of teacher forcing functions to excitations fed as inputs to output neurons. Initially, teacher forcing functions are strong enough to force outputs to desired values; subsequently, these functions decay with time. When learning successfully completed, terminal teacher forcing vanishes, and dynamics or neural network become equivalent to those of conventional neural network. Simulated neural network with terminal teacher forcing learned to produce close approximation of circular trajectory in 400 iterations.

Toomarian, Nikzad↗

DOE OSTI2015

Nominal Device Support for ATCA Shelf Manager

The ATCA standard specifies a mandatory Shelf Manager (ShM) unit which is a key element for the system operation. It includes the Intelligent Platform Management Controller (IPMC) which monitors the system health, retrieves inventory information and controls the Field Replaceable Units (FRUs). These elements enable the intelligent health monitoring, providing high-availability and safety operation, ensuring the correct system operation. For critical systems like ones of tokamak ITER these features are mandatory to support the long pulse operation. The Nominal Device Support (NDS) was designed and developed for the ITER CODAC Core System (CCS), which will be the responsible for plant Instrumentation and Control (I and C), supervising and monitoring on ITER. It generalizes the Enhanced Physics and Industrial Control System (EPICS) device support interface for Data Acquisition (DAQ) and timing devices. However the support for health management features and ATCA ShM are not yet provided. This paper presents the implementation and test of a NDS for the ATCA ShM, using the ITER Fast Plant System Controller (FPSC) prototype environment. This prototype is fully compatible with the ITER CCS and uses the EPICS Channel Access (CA) protocol as the interface with the Plant Operation Network (PON). The implemented solution running in an EPICS Input / Output Controller (IOC) provides Process Variables (PV) to the PON network with the system information. These PVs can be used for control and monitoring by all CA clients, such as EPICS user interface clients and alarm systems. The results are presented, demonstrating the fully integration and the usability of this solution. (authors)

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

DOE OSTI2020

GNET2: an R package for constructing gene regulatory networks from transcriptomic data

Abstract Motivation The Gene Network Estimation Tool (GNET) is designed to build gene regulatory networks (GRNs) from transcriptomic gene expression data with a probabilistic graphical model. The data preprocessing, model construction and visualization modules of the original GNET software were developed on different programming platforms, which were inconvenient for users to deploy and use. Results Here, we present GNET2, an improved implementation of GNET as an integrated R package. GNET2 provides more flexibility for parameter initialization and regulatory module construction based on the core iterative modeling process of the original algorithm. The data exchange interface of GNET2 is handled within an R session automatically. Given the growing demand for regulatory network reconstruction from transcriptomic data, GNET2 offers a convenient option for GRN inference on large datasets. Availability and implementation The source code of GNET2 is available at https://github.com/jianlin-cheng/GNET2. Supplementary information Supplementary data are available at Bioinformatics online.

59 BASIC BIOLOGICAL SCIENCES↗

DOE OSTI2026

Multidimensional Distributional Neural Network Output Demonstrated in Super‐Resolution of Surface Wind Speed

Accurate quantification of uncertainty in neural network predictions remains a central challenge for scientific applications involving high-dimensional, correlated data. While existing methods capture either aleatoric or epistemic uncertainty, few offer closed-form, multidimensional distributions that preserve spatial correlation while remaining computationally tractable. In this work, we present a framework for training neural networks with a multidimensional Gaussian loss, generating a closed-form predictive distribution over outputs informed by non-identically distributed training data. Our approach captures aleatoric uncertainty by iteratively estimating the means and covariance matrices, and is demonstrated on a super-resolution example out-of-training-sample. We leverage a Fourier representation of the covariance matrix to stabilize network training and preserve spatial correlation. We introduce a novel regularization strategy—referred to as information sharing—that interpolates between image-specific and global covariance estimates, enabling convergence of the super-resolution downscaling network trained on image-specific distributional loss functions. This framework allows for efficient sampling, explicit correlation modeling, and extensions to more complex distribution families all without disrupting prediction performance. We demonstrate the method on a surface wind speed downscaling task and discuss its broader applicability to uncertainty-aware prediction in scientific models.

17 WIND ENERGY↗

DOE OSTI2020

Active Learning A Neural Network Model For Gold Clusters & Bulk From Sparse First Principles Training Data

Small metal clusters are of fundamental scientific interest and of tremendous significance in catalysis. These nanoscale clusters display diverse geometries and structural motifs depending on the cluster size; a knowledge of this size-dependent structural motifs and their dynamical evolution has been of longstanding interest. Given the high computational cost of first-principles calculations, molecular modeling and atomistic simulations such as molecular dynamics (MD) has proven to be an important complementary tool to aid this understanding. Classical MD typically employ predefined functional forms which limits their ability to capture such complex size-dependent structural and dynamical transformation. Neural Network (NN) based potentials represent flexible alternatives and in principle, well-trained NN potentials can provide high level of flexibility, transferability and accuracy on-par with the reference model used for training. A major challenge, however, is that NN models are interpolative and requires large quantities (similar to 10 4 or greater) of training data to ensure that the model adequately samples the energy landscape both near and far-from-equilibrium. A highly desirable goal is minimize the number of training data, especially if the underlying reference model is first-principles based and hence expensive. In this work, we introduce an active learning (AL) scheme that trains a NN model on-the-fly with minimal amount of first-principles based training data. Our AL workflow is initiated with a sparse training dataset (similar to 1 to 5 data points) and is updated on-the-fly via a Nested Ensemble Monte Carlo scheme that iteratively queries the energy landscape in regions of failure and updates the training pool to improve the network performance. Using a representative system of gold clusters, we demonstrate that our AL workflow can train a NN with similar to 500 total reference calculations. Using an extensive DFT test set of similar to 1100 configurations, we show that our AL-NN is able to accurately predict both the DFT energies and the forces for clusters of a myriad of different sizes. Our NN predictions are within 30 meV/atom and 40 meV/angstrom of the reference DFT calculations. Moreover, our AL-NN model also adequately captures the various size-dependent structural and dynamical properties of gold clusters in excellent agreement with DFT calculations and available experiments. We finally show that our AL-NN model also captures bulk properties reasonably well, even though they were not included in the training data.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗