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 1,171 records · Page 65

Active learning using hybrid surrogate tool life modeling for machining process optimization

Here, this paper describes an active learning approach for part-to-part iterative machining process optimization using a hybrid surrogate tool life model. A probabilistic interpolating tool life model is developed by combining the empirical Taylor-type tool life equation and the model fit error. The probabilistic tool life model is then used to calculate the machining cost per part distribution. The optimal machining parameters are selected using an expected improvement in machining cost per part criterion. The method is validated numerically using experimental results; the results show a median convergence error of 2.2% after three tests over 400 simulations. The method is validated experimentally on two industrial applications for Ti-6Al-4V roughing resulting in a cost per part reduction greater than 23% after two tests. The described method is a robust solution for rapid convergence to optimal machining parameters in an industrial production environment.

Active learning↗

SU-E-T-478: IMRT Delivery Parameter Dependence of Dose-Mass Optimization

Purpose: To compare DMH and DVH optimization sensitivity to changes in IMRT delivery parameters. Methods: Two lung and two head and neck (HN) cases were retrospectively optimized using DVH and DMH optimization. For both optimization approaches, changes to two parameters were studied: number of IMRT segments (5 and 10 per beam) and the minimum segment area (2 and 6 cm2). The number of beams, beam angles, and minimum MUs per segment were the same for both optimizations approaches for each patient. During optimization, doses to the organs at risk (OARs) were iteratively lowered until the standard deviation across the PTV was above ∼3.0%. For each patient DVH and DMH plans were normalized such that 95% of the PTV received the same dose. Plan quality was evaluated by dose indices (DIs), which represent the dose delivered to a certain anatomical structure volume. For the lung cases, DIs assessed included: 1% cord, 33% heart, both lungs 20% and 30%, and 50% esophagus. In the HN cases: 1% cord, 1% brainstem, left/right parotids 50%, 50% larynx, and 50% esophagus. Results: When increasing the number of segments, while keeping a small segment area (2cm2), the average percent change of all DIs for DVH/DMH optimizations for each patient were: −4.66/4.71, 3.21/3.46, −9.62/21.69 and −3.28/−7.62. For a large segment area (6cm2): −0.26/−1.46, −5.04/−1.92, −5.23/−2.19 and 4.12/19.63. Results from increasing segment area while keeping a small number of segments (5segments/beam) were: 1.41/7.90, 8.17/11.66, 0.09/33.58 and −4.83/−11.60 for each case. For large number of segments (10 segments/beam): 8.35/1.30, −0.91/5.77, 6.29/7.08 and 2.62/5.16. Conclusion: This preliminary study showed case dependent results. Changes in IMRT parameters did not show consistent DI changes for either optimization approach. A larger population of patients is warranted for such comparison.

60 APPLIED LIFE SCIENCES↗

Neural network accelerator for quantum control

Efficient quantum control is necessary for practical quantum computing implementations with current technologies. Conventional algorithms for determining optimal control parameters are computationally expensive, largely excluding them from use outside of the simulation. Existing hardware solutions structured as lookup tables are imprecise and costly. By designing a machine learning model to approximate the results of traditional tools, a more efficient method can be produced. Such a model can then be synthesized into a hardware accelerator for use in quantum systems. In this study, we demonstrate a machine learning algorithm for predicting optimal pulse parameters. This algorithm is lightweight enough to fit on a low-resource FPGA and perform inference with a latency of 175 ns and pipeline interval of 5 ns with > 0.99 gate fidelity. In the long term, such an accelerator could be used near quantum computing hardware where traditional computers cannot operate, enabling quantum control at a reasonable cost at low latencies without incurring large data bandwidths outside of the cryogenic environment.

43 PARTICLE ACCELERATORS↗

Generalized method for the optimization of pulse shape discrimination parameters

Organic scintillators exhibit fast timing, high detection efficiency for fast neutrons and pulse shape discrimination (PSD) capability. PSD is essential in mixed radiation fields, where different types of radiation need to be detected and discriminated. In neutron measurements for nuclear security and non proliferation effective PSD is crucial, because a weak neutron signature needs to be detected in the presence of a strong gamma-ray background. Here, the most commonly used deterministic PSD technique is charge integration (CI). This method requires the optimization of specific parameters to obtain the best gamma-neutron separation. These parameters depend on the scintillating material and light readout device and typically require a lengthy optimization process and a calibration reference measurement with a mixed source. In this paper, we propose a new method based on the scintillation fluorescence physics that enables to find the optimum PSD integration gates using only a gamma-ray emitter. We demonstrate our method using three organic scintillation detectors: deuterated trans-stilbene, small-molecule organic glass, and EJ-309. In all the investigated cases, our method allowed finding the optimum PSD CI parameters without the need of iterative optimization.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

Optimization of the Radiation Parameters on Technological Installations

To ensure that technological installations using radionuclide sources of radiation operate at top efficiency, the effect of the irradiation system parameters on the radiation utilization factor was studied. It was found that the radiation utilization factor depends not on individual parameters of the irradiation system but rather on a certain combination of these parameters. A dimensionless criterion was determined on the basis of these investigations – a complex based on a set of parameters of installations with maximum radiation utilization factor. The radiation utilization factor was determined as a function of the irradiator–irradiated object distance. The numerical range of the criterion corresponding to the maximum radiation utilization factor is shown. Data by means of Monte Carlo calculations were selected by performing experimental studies using independent dosimetric systems.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗

Optimal conditions for efficient flow-electrode capacitive deionization

One of the current barriers to achieving fast and stable performance for flow-electrode capacitive deionization (FCDI) is determining optimal operating parameters. To date, however, no consensus has been reached for universal conditions for FCDI. Through experimental and modeling approaches in this study, we systematically evaluated the influence of applied potential (V = 1.2–2.4 V) and electrolyte concentration (C0 = 0.05–0.5 M) on the FCDI and electrodialysis (ED) desalination processes. Evaluation indicators include the concentration decrease in the desalinated solution, salt removal rates, pH fluctuations, charge efficiency, and energy consumption. Results demonstrated that the dynamic curves of concentration decrease at 2.0 V nearly overlapped with the response at 1.6 V at certain electrolyte concentrations, while the salt removal rates at 0.2 M salt concentration were the best among all concentrations tested at a range of applied potential. Therefore, it was thus concluded that the optimum conditions for FCDI operation are 1.6 V applied potential and 0.2 M initial salt concentration, under which faradaic reactions are not being triggered, and concentration polarization does not significantly affect ion transfer. Furthermore, a comparative study between FCDI and ED indicated that ED has a different dependence on the electrolyte concentration and applied potential, in which the desalination can be linearly enhanced with increasing potential but greatly limited at high concentrations. Due to the presence of carbon particles in FCDI, the enhanced charge/ion transfer is probably the main reason for the different desalination performance of FCDI and ED. Overall, the optimal operating parameters obtained in this work could be used as basic test conditions for further development of new carbon-based materials for FCDI.

42 ENGINEERING↗

Optimization of AC Dipole Parameters for the Mu2e Extinction System

The Mu2e experiment is being planned at Fermilab to measure the rate for muons to convert to electrons in the field of an atomic nucleus with unprecedented precision. This experiment uses an 8 GeV primary proton beam consisting of short ({approx}200 nsec FW) bunches, separated by 1.7 {mu}sec. It is vital that out-of-bunch beam be suppressed at the level of 10{sup -10} or less. This poster describes the parametric analysis which was done to determine the optimum harmonics and magnet specifications for this system, as well as the implications for the beam line optics.

43 PARTICLE ACCELERATORS↗

A new stepwise carbon cycle data assimilation system using multiple data streams to constrain the simulated land surface carbon cycle

Here, large uncertainties in land surface models (LSMs) simulations still arise from inaccurate forcing, poor description of land surface heterogeneity (soil and vegetation properties), incorrect model parameter values and incomplete representation of biogeochemical processes. The recent increase in the number and type of carbon cycle-related observations, including both in situ and remote sensing measurements, has opened a new road to optimize model parameters via robust statistical model–data integration techniques, in order to reduce the uncertainties of simulated carbon fluxes and stocks. In this study we present a carbon cycle data assimilation system that assimilates three major data streams, namely the Moderate Resolution Imaging Spectroradiometer (MODIS)-Normalized Difference Vegetation Index (NDVI) observations of vegetation activity, net ecosystem exchange (NEE) and latent heat (LE) flux measurements at more than 70 sites (FLUXNET), as well as atmospheric CO 2 concentrations at 53 surface stations, in order to optimize the main parameters (around 180 parameters in total) of the Organizing Carbon and Hydrology in Dynamics Ecosystems (ORCHIDEE) LSM (version 1.9.5 used for the Coupled Model Intercomparison Project Phase 5 (CMIP5) simulations). The system relies on a stepwise approach that assimilates each data stream in turn, propagating the information gained on the parameters from one step to the next. Overall, the ORCHIDEE model is able to achieve a consistent fit to all three data streams, which suggests that current LSMs have reached the level of development to assimilate these observations. The assimilation of MODIS-NDVI (step 1) reduced the growing season length in ORCHIDEE for temperate and boreal ecosystems, thus decreasing the global mean annual gross primary production (GPP). Using FLUXNET data (step 2) led to large improvements in the seasonal cycle of the NEE and LE fluxes for all ecosystems (i.e., increased amplitude for temperate ecosystems). The assimilation of atmospheric CO 2 , using the general circulation model (GCM) of the Laboratoire de Météorologie Dynamique (LMDz; step 3), provides an overall constraint (i.e., constraint on large-scale net CO 2 fluxes), resulting in an improvement of the fit to the observed atmospheric CO 2 growth rate. Thus, the optimized model predicts a land C (carbon) sink of around 2.2 PgC yr -1 (for the 2000–2009 period), which is more compatible with current estimates from the Global Carbon Project (GCP) than the prior value. The consistency of the stepwise approach is evaluated with back-compatibility checks. The final optimized model (after step 3) does not significantly degrade the fit to MODIS-NDVI and FLUXNET data that were assimilated in the first two steps, suggesting that a stepwise approach can be used instead of the more “challenging” implementation of a simultaneous optimization in which all data streams are assimilated together. Most parameters, including the scalar of the initial soil carbon pool size, changed during the optimization with a large error reduction. This work opens new perspectives for better predictions of the land carbon budgets.

58 GEOSCIENCES↗

Advanced Interactive Display Formats for Terminal Area Traffic Control

This research project deals with an on-line dynamic method for automated viewing parameter management in perspective displays. Perspective images are optimized such that a human observer will perceive relevant spatial geometrical features with minimal errors. In order to compute the errors at which observers reconstruct spatial features from perspective images, a visual spatial-perception model was formulated. The model was employed as the basis of an optimization scheme aimed at seeking the optimal projection parameter setting. These ideas are implemented in the context of an air traffic control (ATC) application. A concept, referred to as an active display system, was developed. This system uses heuristic rules to identify relevant geometrical features of the three-dimensional air traffic situation. Agile, on-line optimization was achieved by a specially developed and custom-tailored genetic algorithm (GA), which was to deal with the multi-modal characteristics of the objective function and exploit its time-evolving nature.

Grunwald, Arthur J.↗

Advanced Interactive Display Formats for Terminal Area Traffic Control

This research project deals with an on-line dynamic method for automated viewing parameter management in perspective displays. Perspective images are optimized such that a human observer will perceive relevant spatial geometrical features with minimal errors. In order to compute the errors at which observers reconstruct spatial features from perspective images, a visual spatial-perception model was formulated. The model was employed as the basis of an optimization scheme aimed at seeking the optimal projection parameter setting. These ideas are implemented in the context of an air traffic control (ATC) application. A concept, referred to as an active display system, was developed. This system uses heuristic rules to identify relevant geometrical features of the three-dimensional air traffic situation. Agile, on-line optimization was achieved by a specially developed and custom-tailored genetic algorithm (GA), which was to deal with the multi-modal characteristics of the objective function and exploit its time-evolving nature.

Grunwald, Arthur J.↗

Experimental tests of a superposition hypothesis to explain the relationship between the vestibuloocular reflex and smooth pursuit during horizontal combined eye-head tracking in humans

1. We used a modeling approach to test the hypothesis that, in humans, the smooth pursuit (SP) system provides the primary signal for cancelling the vestibuloocular reflex (VOR) during combined eye-head tracking (CEHT) of a target moving smoothly in the horizontal plane. Separate models for SP and the VOR were developed. The optimal values of parameters of the two models were calculated using measured responses of four subjects to trials of SP and the visually enhanced VOR. After optimal parameter values were specified, each model generated waveforms that accurately reflected the subjects' responses to SP and vestibular stimuli. The models were then combined into a CEHT model wherein the final eye movement command signal was generated as the linear summation of the signals from the SP and VOR pathways. 2. The SP-VOR superposition hypothesis was tested using two types of CEHT stimuli, both of which involved passive rotation of subjects in a vestibular chair. The first stimulus consisted of a "chair brake" or sudden stop of the subject's head during CEHT; the visual target continued to move. The second stimulus consisted of a sudden change from the visually enhanced VOR to CEHT ("delayed target onset" paradigm); as the vestibular chair rotated past the angular position of the stationary visual stimulus, the latter started to move in synchrony with the chair. Data collected during experiments that employed these stimuli were compared quantitatively with predictions made by the CEHT model. 3. During CEHT, when the chair was suddenly and unexpectedly stopped, the eye promptly began to move in the orbit to track the moving target. Initially, gaze velocity did not completely match target velocity, however; this finally occurred approximately 100 ms after the brake onset. The model did predict the prompt onset of eye-in-orbit motion after the brake, but it did not predict that gaze velocity would initially be only approximately 70% of target velocity. One possible explanation for this discrepancy is that VOR gain can be dynamically modulated and, during sustained CEHT, it may assume a lower value. Consequently, during CEHT, a smaller-amplitude SP signal would be needed to cancel the lower-gain VOR. This reduction of the SP signal could account for the attenuated tracking response observed immediately after the brake. We found evidence for the dynamic modulation of VOR gain by noting differences in responses to the onset and offset of head rotation in trials of the visually enhanced VOR.(ABSTRACT TRUNCATED AT 400 WORDS).

Non-NASA Center↗

Multiobjective optimization in structural design with uncertain parameters and stochastic processes

The application of multiobjective optimization techniques to structural design problems involving uncertain parameters and random processes is studied. The design of a cantilever beam with a tip mass subjected to a stochastic base excitation is considered for illustration. Several of the problem parameters are assumed to be random variables and the structural mass, fatigue damage, and negative of natural frequency of vibration are considered for minimization. The solution of this three-criteria design problem is found by using global criterion, utility function, game theory, goal programming, goal attainment, bounded objective function, and lexicographic methods. It is observed that the game theory approach is superior in finding a better optimum solution, assuming the proper balance of the various objective functions. The procedures used in the present investigation are expected to be useful in the design of general dynamic systems involving uncertain parameters, stochastic process, and multiple objectives.

Rao, S. S.↗

Optimization of Hot Deformation Parameters and Constitutive Analysis for As-Cast Mg-5Li-3Zn-0.3Y Alloy Using Processing Maps

Hot deformation behavior of as-cast Mg-5Li-3Zn-0.3Y (wt.%) alloy has been studied by establishing processing maps covering the range of 200-450 °C in temperature and 0.001-10 s{sup −1} in strain rate. The map corresponding to the strain of 0.7 reveals that a domain of dynamic recrystallization occurred at 340-400 °C/0.01-0.158 s{sup −1} with its peak efficiency of approximately 31%. Analysis of the flow behaviors and microstructural observations indicated that dynamic recrystallization occurred in the domain, in which fine microstructure can be achieved. The above conditions represent the optimum processing parameters, and the flow instability regimes are also recognized from the maps. The average deformation in the deformed range was 136.414 kJ/mol which is close to that for lattice self-diffusion in magnesium and far smaller than that for as-cast Mg-5.75Zn-0.73Y-0.36Zr (wt.%) alloy. It is believed that a lower activation energy for Mg-5Li-3Zn-0.3Y alloy can partly result from both the weak particle pinning effects and adding Li into the Mg-Zn-Y-Zr alloy. Results of extrusion experiments conducted at 350 °C show that the alloy can be easily extruded with good mechanical properties of the yield strength of 132 MPa, the fracture strength of 215 MPa and the elongation of 20.9%, and the grains can be well refined.

36 MATERIALS SCIENCE↗

Method for estimating optimal spectral and energy parameters of laser irradiation in photodynamic therapy of biological tissue

We have solved the problem of layer-by-layer laser-light dosimetry in biological tissues and of selecting an individual therapeutic dose in laser therapy. A method is proposed for real-time monitoring of the radiation density in tissue layers in vivo, concentrations of its endogenous (natural) and exogenous (specially administered) chromophores, as well as in-depth distributions of the spectrum of light action on these chromophores. As the background information use is made of the spectrum of diffuse light reflected from a patient's tissue, measured by a fibre-optic spectrophotometer. The measured spectrum is quantitatively analysed by the method of approximating functions for fluxes of light multiply scattered in tissue and by a semi-analytical method for calculating the in-depth distribution of the light flux in a multi-layered medium. We have shown the possibility of employing the developed method for monitoring photosensitizer and oxyhaemoglobin concentrations in tissue, light power absorbed by chromophores in tissue layers at different depths and laser-induced changes in the tissue morphology (vascular volume content and ratios of various forms of haemoglobin) during photodynamic therapy. (biophotonics)

60 APPLIED LIFE SCIENCES↗