Search NASA⌕ Search

SEARCH · Search NASA

Results for “variational 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 1,189 records · Page 66

Enhancing generative molecular design via uncertainty-guided fine-tuning of variational autoencoders

In recent years, deep generative models have been successfully applied to various molecular design tasks, particularly in the life and materials sciences. One critical challenge for pre-trained generative molecular design (GMD) models is to fine-tune them to be better suited for downstream design tasks that aim at optimizing specific molecular properties. However, redesigning and training an existing effective generative model from scratch for each new design task are impractical. Furthermore, the black-box nature of typical downstream tasks that involve property prediction makes it nontrivial to optimize the generative model in a task-specific manner. In this work, we propose an uncertainty-guided fine-tuning strategy that can effectively enhance a pre-trained variational autoencoder (VAE) for GMD through performance feedback in an active learning setting. The strategy begins by quantifying the model uncertainty of the generative model using an efficient active subspace-based UQ (uncertainty quantification) scheme. Next, the decoder diversity within the characterized model uncertainty class is explored to expand the viable space of molecular generation. The low-dimensionality of the active subspace makes this exploration tractable using a black-box optimization scheme, which in turn enables us to identify and leverage a diverse set of high-performing models to generate enhanced molecules. Empirical results across six target molecular properties using multiple VAE-based generative models demonstrate that our uncertainty-guided fine-tuning strategy consistently leads to improved models that outperform the original pre-trained models.

97 MATHEMATICS AND COMPUTING↗

Internally cooled membrane-based absorber for dehumidification and water heating: Validated model and simulation study

This report presents a comprehensive model for advanced membrane-based absorber components accounting for three separate fluid streams, one of them providing internal cooling, with complex flow patterns. The model was implemented with various liquid desiccants and their properties (including LiCl, CaCl 2 , and [emim][OMS] ionic liquid [IL]). The model has been validated using data from two laboratory prototype absorbers, one with CaCl 2 as the working fluid and one with an IL as the working fluid. Entropy analysis was further carried out to verify the model and understand effect of internal cooling on the entropy variation of three fluid streams. The developed model and codes are expected to enable detailed configuration optimization and provide in-depth understanding of three-fluid heat and mass exchanger (HMX) performance. The paper also describes parametric studies of the HMX components in utilizing the latent heat removed in space cooling to heat water and numerically explores the unique benefit of its application in a semi-open absorption heat pump water heater. The results show that a single three-fluid HMX has the potential to achieve simultaneous dehumidification and water heating efficiently at cost effectiveness, particularly in hot and humid climates.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

The Jastrow antisymmetric geminal power in Hilbert space: Theory, benchmarking, and application to a novel transition state

The Jastrow-modified antisymmetric geminal power (JAGP) ansatz in Hilbert space successfully overcomes two key failings of other pairing theories, namely, a lack of inter-pair correlations and a lack of multiple resonance structures, while maintaining a polynomially scaling cost, variational energies, and size consistency. Here, we present efficient quantum Monte Carlo algorithms that evaluate and optimize the JAGP energy for a cost that scales as the fifth power of the system size. We demonstrate the JAGP’s ability to describe both static and dynamic correlation by applying it to bond stretching in H{sub 2}O, C{sub 2}, and N{sub 2} as well as to a novel, multi-reference transition state of ethene. JAGP’s accuracy in these systems outperforms even the most sophisticated single-reference methods and approaches that of exponentially scaling active space methods.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Scaling and merging time-resolved pink-beam diffraction with variational inference

Time-resolved x-ray crystallography (TR-X) at synchrotrons and free electron lasers is a promising technique for recording dynamics of molecules at atomic resolution. While experimental methods for TR-X have proliferated and matured, data analysis is often difficult. Extracting small, time-dependent changes in signal is frequently a bottleneck for practitioners. Recent work demonstrated this challenge can be addressed when merging redundant observations by a statistical technique known as variational inference (VI). However, the variational approach to time-resolved data analysis requires identification of successful hyperparameters in order to optimally extract signal. In this case study, we present a successful application of VI to time-resolved changes in an enzyme, DJ-1, upon mixing with a substrate molecule, methylglyoxal. We present a strategy to extract high signal-to-noise changes in electron density from these data. Furthermore, we conduct an ablation study, in which we systematically remove one hyperparameter at a time to demonstrate the impact of each hyperparameter choice on the success of our model. We expect this case study will serve as a practical example for how others may deploy VI in order to analyze their time-resolved diffraction data.

47 OTHER INSTRUMENTATION↗

Closed-Loop Wind Turbine Controllers for Active Wake Mixing Strategies

Wind turbine active wake mixing (AWM) is an exciting new field of research where dynamic actuation, usually on the blade pitch angles, is used to increase wind farm-wide power production. From a controls perspective, the current state-of-the-art AWM strategies are very simple: a dynamic, usually periodic, reference signal is prescribed to actuators in an open-loop (OL) fashion. The actuation is then presumed to have a certain desired effect on the system, i.e., the wind turbine and the flow it affects. However, this system is highly nonlinear and experiences disturbances in the form of wind variations that are not known a priori. As a result, the OL approach might not yield optimal results. In this article, a novel approach is presented, which closes the loop on AWM controllers. A feedback loop is implemented, which uses measurements of the individual blade bending moments that are widely available on modern wind turbines to determine the individual blade pitch angles. A proof of concept of this implementation is presented in this article, and a thorough comparison with the OL method is executed using high-fidelity flow simulations. These simulations show that the novel closed-loop controller does not substantially increase wake mixing but achieves similar performance as the OL controller while reducing fatigue loads on the controlled turbine.

17 WIND ENERGY↗

Joint Data Filtering and Labeling Using Gaussian Processes and Alternating Direction Method of Multipliers

Here, sequence labeling aims at assigning a label to every sample of a signal (or pixel of an image) while considering the sequentiality (or vicinity) of the samples. To perform this task, many works in the literature first filter and then label the data. Unfortunately, the filtering, which is performed independently from the labeling, is far from optimal and frequently makes the latter task harder. In this paper, a novel approach that trains a Gaussian process classifier and estimates the coefficients of an optimal filter jointly is presented. The new approach, based on Bayesian modeling and alternating direction method of multipliers (ADMMs) optimization, performs both tasks simultaneously. All unknowns are treated as stochastic variables, which are estimated using variational inference and filtering and labeling are linked with the use of ADMM. In the experimental section, synthetic and real experiments are presented to compare the proposed method with other existing approaches.

97 MATHEMATICS AND COMPUTING↗

Engineering quantitative stomatal trait variation and local adaptation potential by cis‐regulatory editing

Summary Cis‐regulatory element editing can generate quantitative trait variation that mitigates extreme phenotypes and harmful pleiotropy associated with coding sequence mutations. Here, we applied a multiplexed CRISPR/Cas9 approach, informed by bioinformatic datasets, to generate genotypic variation in the promoter ofOsSTOMAGEN, a positive regulator of rice stomatal density. Engineered genotypic variation corresponded to broad and continuous variation in stomatal density, ranging from 70% to 120% of wild‐type stomatal density. This panel of stomatal variants was leveraged in physiological assays to establish discrete relationships between stomatal morphological variation and stomatal conductance, carbon assimilation and intrinsic water use efficiency in steady‐state and fluctuating light conditions. Additionally, promoter alleles were subjected to vegetative drought regimes to assay the effects of the edited alleles on developmental response to drought. Notably, the capacity for drought‐responsive stomatal density reprogramming instomagenand two cis‐regulatory edited alleles was reduced. Collectively our data demonstrate that cis‐regulatory element editing can generate near‐isogenic trait variation that can be leveraged for establishing relationships between anatomy and physiology, providing a basis for optimizing traits across diverse environments.

Biotechnology & Applied Microbiology↗

Inpatient Opioid Use After Pancreatectomy: Opportunities for Reducing Initial Opioid Exposure in Cancer Surgery Patients

Background: Despite advances in enhanced surgical recovery programs, strategies limiting postoperative inpatient opioid exposure have not been optimized for pancreatic surgery. The primary aims of this study were to analyze the magnitude and variations in post-pancreatectomy opioid administration and to characterize predictors of low and high inpatient use. Methods: Clinical characteristics and inpatient oral morphine equivalents (OMEs) were downloaded from electronic records for consecutive pancreatectomy patients at a high-volume institution between March 2016 and August 2017. Regression analyses identified predictors of total OMEs as well as highest and lowest quartiles. Results: Pancreatectomy was performed for 158 patients (73% pancreaticoduodenectomy). Transversus abdominus plane (TAP) block was performed for 80% (n = 127) of these patients, almost always paired with intravenous patient-controlled analgesia (IV-PCA), whereas 15% received epidural alone. All the patients received scheduled non-opioid analgesics (median, 2). The median total OME administered was 423 mg (range 0–4362 mg). Higher total OME was associated with preoperative opioid prescriptions (p < 0.001), longer hospital length of stay (LOS; p < 0.001), and no epidural (p = 0.006). The lowest and best quartile cutoff was 180 mg of OME or less, whereas the highest and worst quartile cutoff began at 892.5 mg. After adjustment for inpatient team, only epidural use [odds ratio (OR) 0.3; p = 0.04] predicted lowest-quartile OME. Preoperative opioid prescriptions (OR 8.1; p < 0.001), longer operative time (OR 3.4; p = 0.05), and longer LOS (OR 1.1; p = 0.007) predicted highest-quartile OME. Conclusions: Preoperative opioid prescriptions and longer LOS were associated with increased inpatient OME, whereas epidural use reduced inpatient OME. Understanding the predictors of inpatient opioid use and the variables predicting the lowest and highest quartiles can inform decision-making regarding preoperative counseling, regional anesthetic block choice, and novel inpatient opioid weaning strategies to reduce initial postoperative opioid exposure.

62 RADIOLOGY AND NUCLEAR MEDICINE↗

Optimization of Hydride Rim Formation in Unirradiated Zr-4 Cladding

The purpose of this work is to build on the results reported in the M2 milestone M2FT 13PN0805051, document number FCRD-USED-2013-000151 (Hanson, 2013). In that work, it was demonstrated that unirradiated samples of zircaloy-4 cladding could be pre-hydrided at temperatures below 400°C in pure hydrogen gas and that the growth of hydrides on the surface could be controlled by changing the surface condition of the samples and form a desired hydride rim on the outside diameter of the cladding. The work performed at Pacific Northwest National Laboratory since the issuing of the M2 milestone has focused its efforts to optimize the formation of a hydride rim on available zircaloy-4 cladding samples by controlling temperature variation and gas flow control during pre-hydriding treatments. Surface conditioning of the outside surface was also examined as a variable. The results of test indicate that much of the variability in the hydride thickness is due to temperature variation occurring in the furnaces as well as how hydrogen gas flows across the sample surface. Efforts to examine other alloys, gas concentrations, and different surface conditioning plan to be pursed in the next FY as more cladding samples become available.

Cladding↗

Gelcasting low-carbon B4C for cermets (Allison intern SULI paper)

A boron carbide (B 4 C) gelcasting suspension involving melamine formaldehyde (MF) resin was developed to obtain green bodies with low carbon after pyrolysis. MF resin allows highly complicated B 4 C parts to be produced with little to no shrinkage when gelcast into molds. A MF binder content of ~7.5 wt % consistently produced green bodies with sufficient strength to withstand demolding and handling. Variations in B 4 C particle size, curing temperature, and pH were investigated in order to optimize the suspension for casting. Resin pyrolysis trials showed that 90.0 ± 0.9 wt % of resin mass was lost during pyrolysis, corresponding to < 1 wt % of binder residual in the overall green bodies. The minimal carbon content in the green bodies is favorable for the infiltration of aluminum into open pore spaces. Due to the high hardness of B 4 C and the ductility of aluminum, the ceramic-metal (cermet) composite formed is ideal for applications such as light-weight armors.

36 MATERIALS SCIENCE↗

Comparison of Tempered and Truncated Fractional Models

Tempered fractional operators are able to model effects that classical partial differential equations cannot quite capture. For example, this includes the super- and subdiffusion effects that are present in hydrology and geophysics models. However, tempered fractional operators are computationally intensive. We analyze a truncated variation of the fractional operators, which are less computationally intensive, in an effort to use them in place of the more complex tempered variation. In particular, we train parameters of the truncated operator using neural networks in order to optimize the difference of the actions of the two operators.

58 GEOSCIENCES↗

Large Area Picosecond PhotoDetector (LAPPD) definitive tests

The goal of the project was to determine the optimal functional parameters for a new LAPPD, determine the cloud charge radius and test for gain variation due to pore sharing. Also, cosmics tests with C4F8 heavy gas of the new LAPPD were planned.

43 PARTICLE ACCELERATORS↗

Duct liner optimization for turbomachinery noise sources

An acoustical field theory for axisymmetric, multisectioned lined ducts with uniform flow profiles was combined with a numerical minimization algorithm to predict optimal liner configurations having one, two, and three sections. Source models studied include a point source located on the axis of the duct and rotor/outlet-stator viscous wake interaction effects for a typical research compressor operating at an axial flow Mach number of about 0.4. For this latter source, optimal liners for equipartition-of energy, zero-phase, and least-attenuated-mode source variations were also calculated and compared with exact results. It is found that the potential benefits of liner segmentation for the attenuation of turbomachinery noise is greater than would be predicted from point source results. Furthermore, effective liner design requires precise knowledge of the circumferential and radial modal distributions.

Lester, H. C.↗

Discrete approximations to optimal trajectories using direct transcription and nonlinear programming

A recently developed method for solving optimal trajectory problems uses a piecewise-polynomial representation of the state and control variables, enforces the equations of motion via a collocation procedure, and thus approximates the original calculus-of-variations problem with a nonlinear-programming problem, which is solved numerically. This paper identifies this method as a direct transcription method and proceeds to investigate the relationship between the original optimal-control problem and the nonlinear-programming problem. The discretized adjoint equation of the collocation method is found to have deficient accuracy, and an alternate scheme which discretizes the equations of motion using an explicit Runge-Kutta parallel-shooting approach is developed. Both methods are applied to finite-thrust spacecraft trajectory problems, including a low-thrust escape spiral, a three-burn rendezvous, and a low-thrust transfer to the moon.

Enright, Paul J.↗

A simple analytical aerodynamic model of Langley Winged-Cone Aerospace Plane concept

A simple three DOF analytical aerodynamic model of the Langley Winged-Coned Aerospace Plane concept is presented in a form suitable for simulation, trajectory optimization, and guidance and control studies. The analytical model is especially suitable for methods based on variational calculus. Analytical expressions are presented for lift, drag, and pitching moment coefficients from subsonic to hypersonic Mach numbers and angles of attack up to +/- 20 deg. This analytical model has break points at Mach numbers of 1.0, 1.4, 4.0, and 6.0. Across these Mach number break points, the lift, drag, and pitching moment coefficients are made continuous but their derivatives are not. There are no break points in angle of attack. The effect of control surface deflection is not considered. The present analytical model compares well with the APAS calculations and wind tunnel test data for most angles of attack and Mach numbers.

Pamadi, Bandu N.↗

Liner Optimization Studies Using the Ducted Fan Noise Prediction Code TBIEM3D

In this paper we demonstrate the usefulness of the ducted fan noise prediction code TBIEM3D as a liner optimization design tool. Boundary conditions on the interior duct wall allow for hard walls or a locally reacting liner with axially segmented, circumferentially uniform impedance. Two liner optimization studies are considered in which farfield noise attenuation due to the presence of a liner is maximized by adjusting the liner impedance. In the first example, the dependence of optimal liner impedance on frequency and liner length is examined. Results show that both the optimal impedance and attenuation levels are significantly influenced by liner length and frequency. In the second example, TBIEM3D is used to compare radiated sound pressure levels between optimal and non-optimal liner cases at conditions designed to simulate take-off. It is shown that significant noise reduction is achieved for most of the sound field by selecting the optimal or near optimal liner impedance. Our results also indicate that there is relatively large region of the impedance plane over which optimal or near optimal liner behavior is attainable. This is an important conclusion for the designer since there are variations in liner characteristics due to manufacturing imprecisions.

Dunn, M. H.↗

Flight Test of an Intelligent Flight-Control System

The F-15 Advanced Controls Technology for Integrated Vehicles (ACTIVE) airplane (see figure) was the test bed for a flight test of an intelligent flight control system (IFCS). This IFCS utilizes a neural network to determine critical stability and control derivatives for a control law, the real-time gains of which are computed by an algorithm that solves the Riccati equation. These derivatives are also used to identify the parameters of a dynamic model of the airplane. The model is used in a model-following portion of the control law, in order to provide specific vehicle handling characteristics. The flight test of the IFCS marks the initiation of the Intelligent Flight Control System Advanced Concept Program (IFCS ACP), which is a collaboration between NASA and Boeing Phantom Works. The goals of the IFCS ACP are to (1) develop the concept of a flight-control system that uses neural-network technology to identify aircraft characteristics to provide optimal aircraft performance, (2) develop a self-training neural network to update estimates of aircraft properties in flight, and (3) demonstrate the aforementioned concepts on the F-15 ACTIVE airplane in flight. The activities of the initial IFCS ACP were divided into three Phases, each devoted to the attainment of a different objective. The objective of Phase I was to develop a pre-trained neural network to store and recall the wind-tunnel-based stability and control derivatives of the vehicle. The objective of Phase II was to develop a neural network that can learn how to adjust the stability and control derivatives to account for failures or modeling deficiencies. The objective of Phase III was to develop a flight control system that uses the neural network outputs as a basis for controlling the aircraft. The flight test of the IFCS was performed in stages. In the first stage, the Phase I version of the pre-trained neural network was flown in a passive mode. The neural network software was running using flight data inputs with the outputs provided to instrumentation only. The IFCS was not used to control the airplane. In another stage of the flight test, the Phase I pre-trained neural network was integrated into a Phase III version of the flight control system. The Phase I pretrained neural network provided realtime stability and control derivatives to a Phase III controller that was based on a stochastic optimal feedforward and feedback technique (SOFFT). This combined Phase I/III system was operated together with the research flight-control system (RFCS) of the F-15 ACTIVE during the flight test. The RFCS enables the pilot to switch quickly from the experimental- research flight mode back to the safe conventional mode. These initial IFCS ACP flight tests were completed in April 1999. The Phase I/III flight test milestone was to demonstrate, across a range of subsonic and supersonic flight conditions, that the pre-trained neural network could be used to supply real-time aerodynamic stability and control derivatives to the closed-loop optimal SOFFT flight controller. Additional objectives attained in the flight test included (1) flight qualification of a neural-network-based control system; (2) the use of a combined neural-network/closed-loop optimal flight-control system to obtain level-one handling qualities; and (3) demonstration, through variation of control gains, that different handling qualities can be achieved by setting new target parameters. In addition, data for the Phase-II (on-line-learning) neural network were collected, during the use of stacked-frequency- sweep excitation, for post-flight analysis. Initial analysis of these data showed the potential for future flight tests that will incorporate the real-time identification and on-line learning aspects of the IFCS.

Davidson, Ron↗