Search NASA⌕ Search

SEARCH · Search NASA

Results for “gradient methods”

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 361 records · Page 20

Hybrid optimization schemes for simulation-based problems.

The inclusion of computer simulations in the study and design of complex engineering systems has created a need for efficient approaches to simulation-based optimization. For example, in water resources management problems, optimization problems regularly consist of objective functions and constraints that rely on output from a PDE-based simulator. Various assumptions can be made to simplify either the objective function or the physical system so that gradient-based methods apply, however the incorporation of realistic objection functions can be accomplished given the availability of derivative-free optimization methods. A wide variety of derivative-free methods exist and each method has both advantages and disadvantages. Therefore, to address such problems, we propose a hybrid approach, which allows the combining of beneficial elements of multiple methods in order to more efficiently search the design space. Specifically, in this paper, we illustrate the capabilities of two novel algorithms; one which hybridizes pattern search optimization with Gaussian Process emulation and the other which hybridizes pattern search and a genetic algorithm. We describe the hybrid methods and give some numerical results for a hydrological application which illustrate that the hybrids find an optimal solution under conditions for which traditional optimal search methods fail.

59 BASIC BIOLOGICAL SCIENCES↗

Multi-Level Optimal Power Flow Solver in Large Distribution Networks: Preprint

Solving optimal power flow (OPF) problem for large distribution networks incurs high computational complexity. We consider a large multi-phase distribution networks of tree topology with deep penetration of active devices. We divide the network into collaborating areas featuring subtree topology and subareas featuring subsubtree topology. We design a multi-level implementation of the primal-dual gradient algorithm for solving the voltage regulation OPF problems while preserving nodal voltage information and topological information within areas and subareas. Numerical results on a 4,521-node system verifies that the proposed algorithm can significantly improve computational speed without compromising any optimality.

41 EE - Solar Energy Technologies Office (EE-4S)↗

Apparatus, methods and system for temperature gradient aging with in-situ electrical monitoring

An apparatus for making in-situ electrical measurements of a sample while the sample is subjected to a temperature gradient between a first side of the sample and a second side of the sample. The apparatus comprises a first heater and a second heater that are positioned in contact with the first side of the sample and the second side of the sample, respectively. The first heater and the second heater comprise respective faces that each have electrical contacts embedded therein. Electrical measurements pertaining to the sample can be made by way of the electrical contacts embedded in the faces. A temperature at either side of the sample can be monitored by way of thermocouples positioned inside the heaters in proximity to the faces of the heaters.

Reyes, Karla Rosa↗

Non-linear Multidimensional Optimization for use in Wire Scanner Fitting

To ensure experiment efficiency and quality from the Continuous Electron Beam Accelerator at Jefferson Lab, beam energy, size, and position must be measured. Wire scanners are devices inserted into the beamline to produce measurements which are used to obtain beam properties. Extracting physical information from the wire scanner measurements begins by fitting Gaussian curves to the data. This study focuses on optimizing and automating this curve-fitting procedure. We use a hybrid approach combining the efficiency of Newton Conjugate Gradient (NCG) method with the global convergence of three nature-inspired (NI) optimization approaches: genetic algorithm, differential evolution, and particle-swarm. In this Python-implemented approach, augmenting the locally-convergent NCG with one of the globally-convergent methods ensures the quality, robustness, and automation of curve-fitting. After comparing the methods, we establish that given an initial data-derived guess, each finds a solution with the same chi-square- a measurement of the agreement of the fit to the data. NCG is the fastest method, so it is the first to attempt data-fitting. The curve-fitting procedure escalates to one of the globally-convergent NI methods only if NCG fails, thereby ensuring a successful fit. This method allows for the most optimal signal fit and can be easily applied to similar problems.

Terzic, Balsa↗

Learning with Adaptive Conservativeness for Distributionally Robust Optimization: Incentive Design for Voltage Regulation: Preprint

Information asymmetry between the Distribution System Operator (DSO) and Distributed Energy Resource Aggregators (DERAs) obstructs designing effective incentives for voltage regulation. To capture this effect, we employ a Stackelberg game-theoretic framework, where the DSO seeks to overcome the information asymmetry and refine its incentive strategies by learning from DERA behavior over multiple iterations. We introduce a model-based online learning algorithm for the DSO, aimed at inferring the relationship between incentives and DERA responses. Given the uncertain nature of these responses, we also propose a distributionally robust incentive design model to control the probability of voltage regulation failure and then reformulate it into a convex problem. This model allows the DSO to periodically revise distribution assumptions on uncertain parameters in the decision model of the DERA. Finally, we present a gradient-based method that permits the DSO to adaptively modify its conservativeness level, measured by the size of a Wasserstein metric-based ambiguity set, according to historical voltage regulation performance. The effectiveness of our proposed method is demonstrated through numerical experiments.

distribution system operator↗

Material properties effects on the detonation spreading and propagation of diaminoazoxyfurazan (DAAF)

Recent dynamic testing of Diaminoazoxyfurazan (DAAF) has focused on understanding the material properties affecting the detonation propagation, spreading, behavior and symmetry. Small scale gap testing and wedge testing focus on the sensitivity to shock with the gap test including the effects of particle size and density. Floret testing investigates the detonation spreading as it is affected by particle size, density, and binder content. The polyrho testing illustrates the effects of density and binder content on the detonation velocity. Finally the detonation spreading effect can be most dramatically seen in the Mushroom and Onionskin tests where the variations due to density gradients, pressing methods and geometry can be seen on the wave breakout behavior.

36 MATERIALS SCIENCE↗

Optically pumped gradient magnetometer

A method is provided for sensing a magnetic field in a magnetic gradiometer of the kind in which pump light and light constituting an optical carrier traverse first and second atomic vapor cells that contain host atoms and that are separated from each other by a known distance. According to such method, the host atoms are prepared in a coherent superposition of two quantum states that differ in energy by an amount that is sensitive to an ambient magnetic field. Modulation of the optical carrier in the respective cells gives rise to sidebands that interfere to generate a beat frequency indicative of the magnetic field gradient. The host atoms are prepared at least in a mode that allows measurement of ambient magnetic field components perpendicular to the axis of the pump light. In such mode, the host atoms are spin-polarized by pump light while subjected to a controlled magnetic field directed parallel to the pump beam, and then the controlled magnetic field is adiabatically extinguished.

47 OTHER INSTRUMENTATION↗

Deformation accommodating periodic computational domain for a uniform velocity gradient

Many multiscale methods for granular materials use periodic computational domains to consider particle scale interactions and then calculate the stress to drive the continuum scale calculations. For problems involving large material deformations, the computation domain often needs to be reinitialized because of the distortion, causing the loss of the history information of the system. This work introduces an algorithm to accommodate a large deformation of the material while maintaining the computational domain cuboid to avoid the domain reinitialization during the computation. The algorithm uses a rotating frame of reference, in which the velocity gradient can be represented by an upper triangular matrix. The deformation caused by the upper triangular matrix is treated by the image system implied in the periodicity to maintain the computational domain cuboid. The effect from the rotation of the reference frame is considered using the inertial forces. Finally, simulations of simple and pure shear motions are carried out to illustrate the algorithm.

42 ENGINEERING↗

Comparative analysis of plasticity-based GND density estimation methods in crystal plasticity finite element models

In crystal plasticity finite element (CPFE) simulations, accurately quantifying geometrically necessary dislocations (GNDs) is critical for capturing strain gradients in polycrystals. We compare different methods for quantifying GNDs, all of which originate from the Nye tensor, which is computed as the curl of the plastic deformation gradient. The projection technique directly decomposes the Nye tensor onto individual screw and edge dislocation components to compute GNDs. This approach requires converting a nine-component Nye tensor into densities for a larger number of dislocation systems, a fundamentally underdetermined (non-unique) process, which is resolved using L2 minimization. In contrast, when employing CPFE analysis, one could directly compute dislocation densities on each slip system using shear gradients. Projection and slip gradient methods are compared with respect to their prediction of GNDs with changing grain size, strain, and grain neighborhoods, including multigrain junctions. Although these techniques match analytical GND densities for single slip, single crystal deformation, and are consistent with anticipated overall GND trends, we find that the GND densities from projection techniques are significantly lower than those predicted from CPFE-based slip gradients in polycrystals. A suggested improvement of only using the active dislocation systems in the projection technique almost entirely resolved this mismatch.

Crystal plasticity↗

A diagrammatic approach for automatically deriving analytical gradients of tensor hyper-contracted electronic structure methods

In this work, we introduce a diagrammatic approach to facilitate the automatic derivation of analytical nuclear gradients for tensor hyper-contraction (THC) based electronic structure methods. The automatically derived gradients are guaranteed to have the same scaling in terms of both operation count and memory footprint as the underlying energy calculations, and the computation of a gradient is roughly three times as costly as the underlying energy. The new diagrammatic approach enables the first cubic scaling implementation of nuclear derivatives for THC tensors fitted in molecular orbital basis (MO-THC). Furthermore, application of this new approach to THC-MP2 analytical gradients leads to an implementation, which is at least four times faster than the previously reported, manually derived implementation. Finally, we apply the new approach to the 14 tensor contraction patterns appearing in the supporting subspace formulation of multireference perturbation theory, laying the foundation for developments of analytical nuclear gradients and nonadiabatic coupling vectors for multi-state CASPT2.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Volt-VAR Optimization in Distribution Networks Using Twin Delayed Deep Reinforcement Learning

Modern distribution grids are undergoing new challenges due to the stochastic nature of distributed energy resources (DERs). High penetration of DERs has a significant impact on Volt-VAR profile and system power losses. This work proposes a deep reinforcement learning (DRL)-based Volt-VAR optimization approach for improving voltage profile and reducing system power loss under high penetration of distributed energy resources, such as battery energy storage and solar photovoltaic units in distribution grids. The twin delayed deep deterministic policy gradient (TD3) method-based DRL agent is proposed to configure optimal set-points of reactive power outputs of fast responding smart inverters. The agent schedules the reactive power of inverters according to their physical capabilities, such as minimum allowed power factor, e.g., 0.9 leading/lagging. The reward function of the proposed DRL scheme is designed carefully to ensure a proper voltage profile of the grids with effective scheduling of reactive power outputs from inverters. The performance of the proposed model is verified on modified IEEE 34- and 123-bus systems and compared with base case with no reactive supply by inverters, and local droop Volt-VAR control approach. The results show that the proposed method performs better than the local droop control and deep deterministic policy gradient (DDPG)-based DRL method for reducing voltage fluctuation and minimizing power loss.

Hossain, Rakib↗

Scalable Bayesian Physics-Informed Kolmogorov-Arnold Networks

Uncertainty quantification (UQ) plays a pivotal role in scientific machine learning, especially when surrogate models are used to approximate complex systems. Although multilayer perceptions (MLPs) are commonly employed as surrogates, they often suffer from overfitting due to their large number of parameters. Kolmogorov-Arnold networks (KANs) offer an alternative solution with fewer parameters. However, gradient-based inference methods, such as Hamiltonian Monte Carlo (HMC), may result in computational inefficiency when applied to KANs, especially for large-scale datasets, due to the high cost of back-propagation. To address these challenges, we propose a novel approach, combining the dropout Tikhonov ensemble Kalman inversion (DTEKI) with Chebyshev KANs. This gradient-free method effectively mitigates overfitting and enhances numerical stability. In addition, we incorporate the active subspace method to reduce the parameter-space dimensionality, allowing us to improve the accuracy of predictions and obtain more reliable uncertainty estimates. Extensive experiments demonstrate the efficacy of our approach in various test cases, including scenarios with large datasets and high noise levels. Our results show that the new method achieves comparable or better accuracy, much higher efficiency as well as stability compared to HMC, in addition to scalability. Moreover, by leveraging the low-dimensional parameter subspace, our method preserves prediction accuracy while substantially reducing further the computational cost.

97 MATHEMATICS AND COMPUTING↗

Color optimization of red OLEDs via periodic and gradient deposition rate of fluorescent dopants

We have investigated a method to enhance the light emission efficiency and color tuning for mixture of dyes with varying the growth parameters, during Alq{sub 3}:Nile Red deposition. We used solvent bath and green laser (532 nm) for improvement of the performance of devices. This paper presents two kind of evaporation method, periodic and gradient. We observed that periodic emission layers can greatly improve the device performances such as electroluminescence, chromaticity coordinates, luminance, and reduced the turn on voltage. The periodic 3 structure with evaporation rate of (0.1 → 3 → 2 → 3 → 2 → 0.1 A°/s) has a luminance of 3000 cd/m{sup 2} and maximum efficiency of 1.8 cd/A at 18 V which are the highest values among devices.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Charge gradient microscopy

A method for rapid imaging of a material specimen includes positioning a tip to contact the material specimen, and applying a force to a surface of the material specimen via the tip. In addition, the method includes moving the tip across the surface of the material specimen while removing electrical charge therefrom, generating a signal produced by contact between the tip and the surface, and detecting, based on the data, the removed electrical charge induced through the tip during movement of the tip across the surface. The method further includes measuring the detected electrical charge.

36 MATERIALS SCIENCE↗

Analytical energy gradient based on spin-free infinite-order Douglas-Kroll-Hess method with local unitary transformation

In this study, the analytical energy gradient for the spin-free infinite-order Douglas-Kroll-Hess (IODKH) method at the levels of the Hartree-Fock (HF), density functional theory (DFT), and second-order Møller-Plesset perturbation theory (MP2) is developed. Furthermore, adopting the local unitary transformation (LUT) scheme for the IODKH method improves the efficiency in computation of the analytical energy gradient. Numerical assessments of the present gradient method are performed at the HF, DFT, and MP2 levels for the IODKH with and without the LUT scheme. The accuracies are examined for diatomic molecules such as hydrogen halides, halogen dimers, coinage metal (Cu, Ag, and Au) halides, and coinage metal dimers, and 20 metal complexes, including the fourth–sixth row transition metals. In addition, the efficiencies are investigated for one-, two-, and three-dimensional silver clusters. The numerical results confirm the accuracy and efficiency of the present method.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Gradient Coding With Iterative Block Leverage Score Sampling

Gradient coding is a method for mitigating straggling servers in a centralized computing network that uses erasure-coding techniques to distributively carry out first-order optimization methods. Randomized numerical linear algebra uses randomization to develop improved algorithms for large-scale linear algebra computations. In this study, we propose a method for distributed optimization that combines gradient coding and randomized numerical linear algebra. The proposed method uses a randomized ℓ 2 -subspace embedding and a gradient coding technique to distribute blocks of data to the computational nodes of a centralized network, and at each iteration the central server only requires a small number of computations to obtain the steepest descent update. The novelty of our approach is that the data is replicated according to importance scores, called block leverage scores, in contrast to most gradient coding approaches that uniformly replicate the data blocks. Furthermore, we do not require a decoding step at each iteration, avoiding a bottleneck in previous gradient coding schemes. We show that our approach results in a valid ℓ 2 -subspace embedding, and that our resulting approximation converges to the optimal solution.

97 MATHEMATICS AND COMPUTING↗