Search NASA⌕ Search

SEARCH · Search NASA

Results for “parameter optimization”

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 1,207 records · Page 67

Learning to Optimize Variational Quantum Circuits to Solve Combinatorial Problems

Quantum computing is a computational paradigm with the potential to outperform classical methods for a variety of problems. Proposed recently, the Quantum Approximate Optimization Algorithm (QAOA) is considered as one of the leading candidates for demonstrating quantum advantage in the near term. QAOA is a variational hybrid quantum-classical algorithm for approximately solving combinatorial optimization problems. The quality of the solution obtained by QAOA for a given problem instance depends on the performance of the classical optimizer used to optimize the variational parameters. In this paper, we formulate the problem of finding optimal QAOA parameters as a learning task in which the knowledge gained from solving training instances can be leveraged to find high-quality solutions for unseen test instances. To this end, we develop two machine-learning-based approaches. Our first approach adopts a reinforcement learning (RL) framework to learn a policy network to optimize QAOA circuits. Our second approach adopts a kernel density estimation (KDE) technique to learn a generative model of optimal QAOA parameters. In both approaches, the training procedure is performed on small-sized problem instances that can be simulated on a classical computer; yet the learned RL policy and the generative model can be used to efficiently solve larger problems. Furthermore, extensive simulations using the IBM Qiskit Aer quantum circuit simulator demonstrate that our proposed RL- and KDE-based approaches reduce the optimality gap by factors up to 30.15 when compared with other commonly used off-the-shelf optimizers.

97 MATHEMATICS AND COMPUTING↗

Optimization of spray breakup model parameters for predicting fuel spray and film characteristics in gasoline direct injection engines

This study investigated the behavior of gasoline direct injection (GDI) sprays using computational fluid dynamics (CFD). The authors developed an approach to identify optimal spray breakup model parameters by evaluating an error function across numerous simulations, with the goal of minimizing discrepancies from experimental data. Using the optimal setup, the simulated spray matched well with projected liquid volume distributions, liquid penetration, and spray width measured in a constant-pressure continuous-flow chamber. To further validate the approach, the same setup was tested across various fuels, injectors, and operating conditions. Subsequently, the optimal setup, along with a recently developed spray-wall interaction model, were applied to a direct-injection spark-ignited engine under late-injection conditions to predict and evaluate fuel film formation and evolution at varying engine coolant temperatures. Here, with the centrally mounted injector directing the spray toward the piston, simulations indicated that the spray tends to impinge on the piston surface. The proposed simulation framework also accurately captured the aggregate film area on the piston surface, aligning with previously published experimental results. Moreover, simulations showed that increasing the coolant temperature from cold start conditions (333 K) to warm conditions (363 K) reduced the fuel mass deposited on the piston by roughly 50%. Furthermore, for the spray-guided engine configuration studied in this work, the CFD model predicted minimal film deposition on the spark plug electrodes regardless of the coolant temperatures due to a relatively weak in-cylinder flow during the compression phase.

Computational fluid dynamics (CFD)↗

Selective-area growth of heavily n–doped GaAs nanostubs on Si(001) by molecular beam epitaxy

Using an aspect ratio trapping technique, we demonstrate molecular beam epitaxy of GaAs nanostubs on Si(001) substrates. Nanoholes in a SiO{sub 2} mask act as a template for GaAs-on-Si selective-area growth (SAG) of nanostubs 120 nm tall and ≤100 nm in diameter. We investigate the influence of growth parameters including substrate temperature and growth rate on SAG. Optimizing these parameters results in complete selectivity with GaAs growth only on the exposed Si(001). Due to the confined-geometry, strain and defects in the GaAs nanostubs are restricted in lateral dimensions, and surface energy is further minimized. We assess the electrical properties of the selectively grown GaAs nanostubs by fabricating heterogeneous p{sup +}–Si/n{sup +}–GaAs p–n diodes.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Data-driven quantum approximate optimization algorithm for power systems

Abstract Quantum technology provides a ground-breaking methodology to tackle challenging computational issues in power systems. It is especially promising for Distributed Energy Resources (DERs) dominant systems that have been widely developed to promote energy sustainability. In those systems, knowing the maximum sections of power and data delivery is essential for monitoring, operation, and control. However, high computational effort is required. By leveraging quantum resources, Quantum Approximate Optimization Algorithm (QAOA) provides a means to search for these sections efficiently. However, QAOA performance relies heavily on critical parameters, especially for weighted graphs. Here we present a data-driven QAOA, which transfers quasi-optimal parameters between weighted graphs based on the normalized graph density. We verify the strategy with 39,774 expectation value calculations. Without parameter optimization, our data-driven QAOA is comparable with the Goemans-Williamson algorithm. This work advances QAOA and pilots its practical application to power systems in noisy intermediate-scale quantum devices.

Jing, Hang (ORCID:0000000256343489)↗

Bayesian Optimization for Anything (BOA): An open-source framework for accessible, user-friendly Bayesian optimization

We introduce Bayesian Optimization for Anything (BOA), a high-level Bayesian Optimization (BO) framework and model wrapping toolkit, which presents a novel approach to simplifying BO, with the goal of making it more accessible and user-friendly, particularly for those with limited expertise in the field. BOA addresses common barriers in implementing BO, focusing on ease of use, reducing the need for deep domain knowledge, and cutting down on extensive coding requirements. A notable feature of BOA is its language-agnostic architecture, which facilitates broader application in various fields and to a wider audience. We showcase BOA's application through three examples: a high-dimensional optimization with parameters of the SWAT+ watershed model, a highly parallelized optimization of this intrinsically non-parallel model, and a multi-objective optimization of the FETCH Tree-Crown Hydrodynamics model. Furthermore, these test cases illustrate BOA's effectiveness in addressing complex optimization challenges in diverse scenarios.

54 ENVIRONMENTAL SCIENCES↗

Stabilizing dynamic subsea power cables using Bi-stable nonlinear energy sinks

This study investigates vibration mitigation of dynamic subsea cables through passive bi-stable nonlinear energy sinks (B-NESs). These devices suppress vibration energy in a broadband way, and can be regarded as extensions of classical linear tuned mass dampers (TMDs) which are narrowband devices. Through the open-source MoorDyn library, we simulated the vibrations of a vertical subsea cable equipped with a set of B-NESs. Multi-objective optimization was performed to detect the B-NES parameters for optimal mitigation of the cable vibrations. Advanced signal processing verified the efficacy of the optimized B-NESs not only to rapidly absorb and locally dissipate vibration energy, but also to nonlinearly scatter vibration energy from low to high frequencies within the cable itself. This last feature is especially beneficial for vibration mitigation of the undersea cable, as at higher frequencies the cable vibrations exhibit drastically reduced amplitudes and are more effectively dissipated by inherent structural damping and hydrodynamic radiation damping. This contrasts with traditional TMDs which can mitigate vibration only at a single frequency. Furthermore, our robustness study confirms the B-NES's effectiveness under even varying environmental conditions. Overall, the B-NES's capacity for broadband vibration mitigation renders it a promising retrofit solution for improving the performance and operational safety of dynamic power cables in offshore wind farms and other marine applications.

17 WIND ENERGY↗

Coherent Exciton Dynamics in the Presence of Underdamped Vibrations

Recent ultrafast optical experiments show that excitons in large biological light-harvesting complexes are coupled to molecular vibration modes. These high-frequency vibrations will not only affect the optical response, but also drive the exciton transport. Here, using a model dimer system, the frequency of the underdamped vibration is shown to have a strong effect on the exciton dynamics such that quantum coherent oscillations in the system can be present even in the case of strong noise. Two mechanisms are identified to be responsible for the enhanced transport efficiency: critical damping due to the tunable effective strength of the coupling to the bath, and resonance coupling where the vibrational frequency coincides with the energy gap in the system. The interplay of these two mechanisms determines parameters responsible for the most efficient transport, and these optimal control parameters are comparable to those in realistic light-harvesting complexes. Interestingly, oscillations in the excitonic coherence at resonance are suppressed in comparison to the case of an off-resonant vibration.

74 ATOMIC AND MOLECULAR PHYSICS↗

A.I. for nuclear physics

Nuclear science is concerned with the understanding of the nature of matter, its basic constituents and their interaction to form the elements and the properties we observe. This includes the forms of matter we see around us and also exotic forms such as those that existed in the first moments after the Big Bang and that exist today inside neutron stars. The techniques, tools, and expertise needed for nuclear physics (NP) research are therefore diverse in nature. State-of-the art accelerators are being developed to illuminate the dynamical basis of the core of the atom in terms of the fundamental constituents called quarks and gluons and to increase the number of isotopes with known properties. This scientific infrastructure is reaching scales and complexities that require computational methods for tasks such as anomaly detection in operational data. New methodologies are needed to detect anomalies and to optimize operating parameters, predict failures as well as to discover new optimization algorithms.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗

Skip-Connected Self-Recurrent Spiking Neural Networks with Joint Intrinsic Parameter and Synaptic Weight Training

As an important class of spiking neural networks (SNNs), recurrent spiking neural networks (RSNNs) possess great computational power and have been widely used for processing sequential data like audio and text. However, most RSNNs suffer from two problems. 1. Due to the lack of architectural guidance, random recurrent connectivity is often adopted, which does not guarantee good performance. 2. Training of RSNNs is in general challenging, bottlenecking achievable model accuracy. To address these problems, we propose a new type of RSNNs called Skip-Connected Self-Recurrent SNNs (ScSr-SNNs). Recurrence in ScSr-SNNs is introduced in a stereotyped manner by adding self-recurrent connections to spiking neurons. The SNNs with self-recurrent connections can realize recurrent behaviors similar to those of more complex RSNNs while the error gradients can be more straightforwardly calculated due to the mostly feedforward nature of the network. The network dynamics is enriched by skip connections between nonadjacent layers. Moreover, we propose a new backpropagation (BP) method called backpropagated intrinsic plasticity (BIP) to further boost the performance of ScSr-SNNs by training intrinsic model parameters. Unlike standard intrinsic plasticity rules that adjust the neuron's intrinsic parameters according to neuronal activity, the proposed BIP method optimizes intrinsic parameters based on the backpropagated error gradient of a well-defined global loss function in addition to synaptic weight training. Here, based on challenging speech, neuromorphic speech, and neuromorphic image datasets, the proposed ScSr-SNNs can boost performance by up to 2.85% compared with other types of RSNNs trained by state-of-the-art BP methods.

97 MATHEMATICS AND COMPUTING↗

Performance of a prototype atomic clock based on lin parallel lin coherent population trapping resonances in Rb atomic vapor

We report on the performance of the first table-top prototype atomic clock based on coherent population trapping (CPT) resonances with parallel linearly polarized optical fields (lin parallel lin configuration). Our apparatus uses a vertical-cavity surface-emitting laser (VCSEL) tuned to the D{sub 1} line of {sup 87}Rb with the current modulation at the {sup 87}Rb hyperfine frequency. We demonstrate cancellation of the first-order light shift by the proper choice of rf modulation power and further improve our prototype clock stability by optimizing the parameters of the microwave lock loop. Operating in these optimal conditions, we measured a short-term fractional frequency stability (Allan deviation) 2x10{sup -11}{tau}{sup -1/2} for observation times 1 s{<=}{tau}{<=}20 s. This value is limited by large VCSEL phase noise and environmental temperature fluctuation. Further improvements in frequency stability should be possible with an apparatus designed as a dedicated lin parallel lin CPT resonance clock with environmental impacts minimized.

74 ATOMIC AND MOLECULAR PHYSICS↗

Spatially-Variant Tikhonov Regularization for Double-Difference Waveform Inversion

Double-difference waveform inversion is a potential tool for quantitative monitoring for geologic carbon storage. It jointly inverts time-lapse seismic data for changes in reservoir geophysical properties. Due to the ill-posedness of waveform inversion, it is a great challenge to obtain reservoir changes accurately and efficiently, particularly when using time-lapse seismic reflection data. Regularization techniques can be utilized to address the issue of ill-posedness. The regularization parameter controls the smoothness of inversion results. A constant regularization parameter is normally used in waveform inversion, and an optimal regularization parameter has to be selected. The resulting inversion results are a trade off among regions with different smoothness or noise levels; therefore the images are either over regularized in some regions while under regularized in the others. In this paper, we employ a spatially-variant parameter in the Tikhonov regularization scheme used in double-difference waveform tomography to improve the inversion accuracy and robustness. We compare the results obtained using a spatially-variant parameter with those obtained using a constant regularization parameter and those produced without any regularization. We observe that, utilizing a spatially-variant regularization scheme, the target regions are well reconstructed while the noise is reduced in the other regions. We show that the spatially-variant regularization scheme provides the flexibility to regularize local regions based on the a priori information without increasing computational costs and the computer memory requirement.

ACCURACY↗

Guidelines and Parameter Selection for the Simulation of Progressive Delamination

Turon s methodology for determining optimal analysis parameters for the simulation of progressive delamination is reviewed. Recommended procedures for determining analysis parameters for efficient delamination growth predictions using the Abaqus/Standard cohesive element and relatively coarse meshes are provided for single and mixed-mode loading. The Abaqus cohesive element, COH3D8, and a user-defined cohesive element are used to develop finite element models of the double cantilever beam specimen, the end-notched flexure specimen, and the mixed-mode bending specimen to simulate progressive delamination growth in Mode I, Mode II, and mixed-mode fracture, respectively. The predicted responses are compared with their analytical solutions. The results show that for single-mode fracture, the predicted responses obtained with the Abaqus cohesive element correlate well with the analytical solutions. For mixed-mode fracture, it was found that the response predicted using COH3D8 elements depends on the damage evolution criterion that is used. The energy-based criterion overpredicts the peak loads and load-deflection response. The results predicted using a tabulated form of the BK criterion correlate well with the analytical solution and with the results predicted with the user-written element.

Song, Kyongchan↗

Melt Pool and Heat Treatment Optimization for the Fabrication of High-Strength and High-Toughness Additively Manufactured 4340 Steel

Additively manufactured (AM) components offer superior design flexibility compared to their conventionally manufactured counterparts, and optimizing processing parameters is key to achieving high-quality depositions with desirable and predictable mechanical properties. This study was focused on 4340 steel fabricated using laser powder bed fusion (LPBF), and 42 laser power and scan speed combinations have been systematically investigated to determine an optimized melt pool geometry that would ensure fully-dense parts. The AM material was compared with a wrought 4340 equivalent and studied in two customized heat treated conditions, optimized for strength and toughness, respectively. The microstructures of the as-fabricated and heat treated AM and wrought materials were characterized to assess differences introduced by the layer-by-layer fabrication process and subsequent heat treatment. Tensile properties of both materials were also evaluated and demonstrate that the AM materials offer equal or superior properties compared to the wrought equivalents. Differences in fracture surface morphologies indicate the distinct failure mechanisms associated with the materials’ characteristic microstructures, and the role of inclusions in the failures was studied to elucidate these differences. Complementary to the experimental investigations, the dataset was leveraged to make recommendations for future design of experiments to optimize AM build parameters in other material systems. A statistical Monte Carlo analysis was used to predict the interpolation error produced using reduced datasets and to enable informed processing parameters selection. These findings are discussed to make recommendations for the use of AM materials for high-integrity structural applications.

36 MATERIALS SCIENCE↗

Automated detection of part quality during two-photon lithography via deep learning

Two-photon lithography (TPL) is an additive manufacturing technique for fabricating three-dimensional objects with nanoscale features. A main challenge of TPL is the routine and labor-intensive task of finding suitable light dosage parameters, i.e. writing speed and laser intensity that induce photo-polymerization within a wide variety of candidate photo-curing polymers. Another challenge is the monitoring required during fabrication. In this work, we apply machine learning (ML) models to accelerate the process of identifying optimal light dosage parameters and automate the detection of part quality. We curate TPL videos of different parts fabricated under a range of light dosage parameters using different resins and train spatial-temporal ML models on this data. Our results show that ML models can detect TPL part quality with a 95.1% accuracy in milliseconds. We also evaluate classification failures and identify two operating modes: parameter optimization and part quality detection. Last but not least, we publicly release this labelled dataset so that it may serve as a useful benchmark to the community. Our approach to process optimization and part quality detection addresses important aspects of TPL industrialization, is applicable beyond TPL and should benefit other additive manufacturing techniques with similar barriers to operating at industrial scale.

36 MATERIALS SCIENCE↗

Parameter estimation techniques based on optimizing goodness-of-fit statistics for structural reliability

New methods are presented that utilize the optimization of goodness-of-fit statistics in order to estimate Weibull parameters from failure data. It is assumed that the underlying population is characterized by a three-parameter Weibull distribution. Goodness-of-fit tests are based on the empirical distribution function (EDF). The EDF is a step function, calculated using failure data, and represents an approximation of the cumulative distribution function for the underlying population. Statistics (such as the Kolmogorov-Smirnov statistic and the Anderson-Darling statistic) measure the discrepancy between the EDF and the cumulative distribution function (CDF). These statistics are minimized with respect to the three Weibull parameters. Due to nonlinearities encountered in the minimization process, Powell's numerical optimization procedure is applied to obtain the optimum value of the EDF. Numerical examples show the applicability of these new estimation methods. The results are compared to the estimates obtained with Cooper's nonlinear regression algorithm.

Starlinger, Alois↗

Quantum Alternating Operator Ansatz (QAOA) Phase Diagrams and Applications for Quantum Chemistry

Determining Hamiltonian ground states and energies is a challenging task with many possible approaches on quantum computers. While variational quantum eigensolvers are popular approaches for near term hardware, adiabatic state preparation is an alternative that does not require noisy optimization of parameters. Beyond adiabatic schedules, QAOA is an important method for optimization problems. In this work we modify QAOA to apply to finding ground states of molecules and empirically evaluate the modified algorithm on several molecules. This modification applies physical insights used in classical approximations to construct suitable QAOA operators and initial state. We find robust qualitative behavior for QAOA as a function of the number of steps and size of the parameters, and demonstrate this behavior also occurs in standard QAOA applied to combinatorial search. To this end we introduce QAOA phase diagrams that capture its performance and properties in various limits. In particular we show a region in which non-adiabatic schedules perform better than the adiabatic limit while employing lower quantum circuit depth. We further provide evidence our results and insights also apply to QAOA applications beyond chemistry.

Kremenetski, Vladimir↗

NIMBUS-7 CZCS. Coastal Zone Color Scanner Imagery for Selected Coastal Regions. North America - Europe. South America - Africa - Antarctica. Level 2 Photographic Product

The Nimbus-7 Coastal Zone Color Scanner (CZCS) is the first spacecraft instrument devoted to the measurement of ocean color. Although instruments on other satellites have sensed ocean color, their spectral bands, spatial resolution, and dynamic range were optimized for geographical or meteorological use. In the CZCS, every parameter is optimized for use over water to the exclusion of any other type of sensing. The signal-to-noise ratios in the spectral channels sensing reflected solar radiance are higher than those required in the past. These ratios need to be high because the ocean is such a poor reflecting surface that the majority of the signal seen by the reflected energy channels at spacecraft altitudes is backscattered solar radiation from the atmosphere rather than reflected solar energy from the ocean. The CZCS is a conventional multichannel scanning radiometer utilizing a rotating plane mirror at a 45 deg angle to the optic axis of a Cassegrain telescope. The mirror scans 360 deg; however, only 80 deg of data centered on the spacecraft nadir is collected for ocean color measurements. Spatial resolution at spacecraft nadir is 825x825 m with some degradation at the edges of the scan swath. The useful swath width from a spacecraft altitude of 955 km is 1600 km.

Source record↗

Development of methodologies for the estimation of thermal properties associated with aerospace vehicles

Thermal stress analyses are an important aspect in the development of aerospace vehicles such as the National Aero-Space Plane (NASP) and the High-Speed Civil Transport (HSCT) at NASA-LaRC. These analyses require knowledge of the temperature within the structures which consequently necessitates the need for thermal property data. The initial goal of this research effort was to develop a methodology for the estimation of thermal properties of aerospace structural materials at room temperature and to develop a procedure to optimize the estimation process. The estimation procedure was implemented utilizing a general purpose finite element code. In addition, an optimization procedure was developed and implemented to determine critical experimental parameters to optimize the estimation procedure. Finally, preliminary experiments were conducted at the Aircraft Structures Branch (ASB) laboratory.

Scott, Elaine P.↗