Using Small-Angle Scattering Data and Parametric Machine Learning to Optimize Force Field Parameters for Intrinsically Disordered Proteins
Not Available
SEARCH · Search NASA
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.
Not Available
NASA is presently engaged in the design and development of a water reclamation system for the future space station. A major concern in processing water is the control of microbial contamination. As a means of developing an optimal microbial control strategy, studies were undertaken to determine the type and amount of contamination which could be expected in these systems under a variety of changing environmental conditions. A laboratory-based Taguchi optimization experiment was conducted to determine the ideal settings for 15 parameters which influence the survival of six bacterial species in aquatic systems. The experiment demonstrated that the bacterial survival period could be decreased significantly by optimizing environmental conditions.
Method for reducing sensitivity of optimal nonlinear systems to parameter uncertainty
In the present investigation, an in-situ multi-component reinforced aluminum based metal matrix composite was fabricated by the combination of self-propagating high-temperature synthesis and direct metal laser sintering process. The different mixtures of Al, TiO{sub 2} and B{sub 4}C powders were used to initiate and maintain the self-propagating high-temperature synthesis by laser during the sintering process. It was found from the X-ray diffraction analysis and scanning electron microscopy that the reinforcements like Al{sub 2}O{sub 3}, TiC, and TiB{sub 2} were formed in the composite. The scanning electron microscopy revealed the distribution of the reinforcement phases in the composite and phase identities. The variable parameters such as powder layer thickness, laser power, scanning speed, hatching distance and composition of the powder mixture were optimized for higher density, lower porosity and higher microhardness using Taguchi method. Experimental investigation shows that the density of the specimen mainly depends upon the hatching distance, composition and layer thickness. On the other hand, hatching distance, layer thickness and laser power are the significant parameters which influence the porosity. The composition, laser power and layer thickness are the key influencing parameters for microhardness. - Highlights: • The reinforcements such as Al{sub 2}O{sub 3}, TiC, and TiB{sub 2} were produced in Al-MMC through SHS. • The density is mainly influenced by the material composition and hatching distance. • Hatching distance is the major influencing parameter on porosity. • The material composition is the significant parameter to enhance the microhardness. • The SEM micrographs reveal the distribution of TiC, TiB{sub 2} and Al{sub 2}O{sub 3} in the composite.
Not provided.
Linear optimal control in systems with uncertain parameters, noting application to design of compensating network for flexible booster for uncertain value of first bending mode
Design optimization for fractional factorial experiments by finite decision problem based on prior information
The first twelve system state variables are presented with the necessary mathematical developments for incorporating them into the filter/smoother algorithm. Other state variables, i.e., aerodynamic coefficients can be easily incorporated into the estimation algorithm, representing uncertain parameters, but for initial checkout purposes are treated as known quantities. An approach for incorporating the NASA propulsion predictive model results into the optimal estimation algorithm was identified. This approach utilizes numerical derivatives and nominal predictions within the algorithm with global iterations of the algorithm. The iterative process is terminated when the quality of the estimates provided no longer significantly improves.
Purpose: To develop a fast and robust technique for the determination of optimized photon spectra for flattening filter free (FFF) beams to be applied in convolution/superposition dose calculations. Methods: A two-step optimization method was developed to derive optimal photon spectra for FFF beams. In the first step, a simple functional form of the photon spectra proposed by Ali ['Functional forms for photon spectra of clinical linacs,' Phys. Med. Biol. 57, 31-50 (2011)] is used to determine generalized shapes of the photon spectra. In this method, the photon spectra were defined for the ranges of field sizes to consider the variations of the contributions of scattered photons with field size. Percent depth doses (PDDs) for each field size were measured and calculated to define a cost function, and a collapsed cone convolution (CCC) algorithm was used to calculate the PDDs. In the second step, the generalized functional form of the photon spectra was fine-tuned in a process whereby the weights of photon fluence became the optimizing free parameters. A line search method was used for the optimization and first order derivatives with respect to the optimizing parameters were derived from the CCC algorithm to enhance the speed of the optimization. The derived photon spectra were evaluated, and the dose distributions using the optimized spectra were validated. Results: The optimal spectra demonstrate small variations with field size for the 6 MV FFF beam and relatively large variations for the 10 MV FFF beam. The mean energies of the optimized 6 MV FFF spectra were decreased from 1.31 MeV for a 3 Multiplication-Sign 3 cm{sup 2} field to 1.21 MeV for a 40 Multiplication-Sign 40 cm{sup 2} field, and from 2.33 MeV at 3 Multiplication-Sign 3 cm{sup 2} to 2.18 MeV at 40 Multiplication-Sign 40 cm{sup 2} for the 10 MV FFF beam. The developed method could significantly improve the agreement between the calculated and measured PDDs. Root mean square differences on the optimized PDDs were observed to be 0.41% (3 Multiplication-Sign 3 cm{sup 2}) down to 0.21% (40 Multiplication-Sign 40 cm{sup 2}) for the 6 MV FFF beam, and 0.35% (3 Multiplication-Sign 3 cm{sup 2}) down to 0.29% (40 Multiplication-Sign 40 cm{sup 2}) for the 10 MV FFF beam. The first order derivatives from the functional form were found to improve the speed of computational time up to 20 times compared to the other techniques. Conclusions: The derived photon spectra resulted in good agreements with measured PDDs over the range of field sizes investigated. The suggested method is easily applicable to commercial radiation treatment planning systems since it only requires measured PDDs as input.
Protocol optimization is critical in Computed Tomography (CT) for achieving desired diagnostic image quality while minimizing radiation dose. Due to the inter-effect of influencing CT parameters, traditional optimization methods rely on the testing of exhaustive combinations of these parameters. This poses a notable limitation due to the impracticality of exhaustive parameter testing. This study introduces a novel methodology leveraging Virtual Imaging Trials (VITs) and reinforcement learning to more efficiently optimize CT protocols. Computational phantoms with liver lesions were imaged using a validated CT simulator and reconstructed with a novel CT reconstruction Toolkit. The optimization parameter space included tube voltage, tube current, reconstruction kernel, slice thickness, and pixel size. The optimization process was done using a Proximal Policy Optimization (PPO) agent which was trained to maximize the Detectability Index (d’) of the liver lesion for each reconstructed image. Results showed that our reinforcement learning approach found the absolute maximum d’ across the test cases while requiring 79.7% fewer steps compared to an exhaustive search, demonstrating both accuracy and computational efficiency, offering a efficient and robust framework for CT protocol optimization. The flexibility of the proposed technique allows for use of varying image quality metrics as the objective metric to maximize for. Our findings highlight the advantages of combining VIT and reinforcement learning for CT protocol management.
In low earth orbit (LEO) satellite applications spacecraft power is provided by photovoltaic cells and batteries. To overcome battery shortcomings the University of Maryland, working in cooperation with NASA/GSFC and NASA/LeRC, has developed a magnetically suspended flywheel for energy storage applications. The system is referred to as an Open Core Composite Flywheel (OCCF) energy storage system. Successful application of flywheel energy storage requires integration of several technologies, viz. bearings, rotor design, motor/generator, power conditioning, and system control. In this paper we present a parameter design method which has been developed for analyzing the linear SISO model of the magnetic bearing controller for the OCCF. The objective of this continued research is to principally analyze the magnetic bearing system for nonlinear effects in order to increase the region of stability, as determined by high speed and large air gap control. This is achieved by four tasks: (1) physical modeling, design, prototyping, and testing of an improved magnetically suspended flywheel energy storage system, (2) identification of problems that limit performance and their corresponding solutions, (3) development of a design methodology for magnetic bearings, and (4) design of an optimal controller for future high speed applications. Both nonlinear SISO and MIMO models of the magnetic system were built to study limit cycle oscillations and power amplifier saturation phenomenon observed in experiments. The nonlinear models include the inductance of EM coils, the power amplifier saturation, and the physical limitation of the flywheel movement as discussed earlier. The control program EASY5 is used to study the nonlinear SISO and MIMO models. Our results have shown that the characteristics and frequency responses of the magnetic bearing system obtained from modeling are comparable to those obtained experimentally. Although magnetic saturation is shown in the bearings, there are good correlations between the theoretical model and experimental data. Both simulation and experiment confirm large variations of the magnetic bearing characteristics due to air gap growth. Therefore, the gap growth effect should be considered in the magnetic bearing system design. Additionally, the magnetic bearing control system will be compared to other design methods using not only parameter design but H-infinity optimal control and mu synthesis.
Previous efforts determined a set of calibrated, optimal model parameter values for Reynolds-averaged Navier–Stokes (RANS) simulations of a compressible jet in crossflow (JIC) using a $k–ε$ turbulence model. These parameters were derived by comparing simulation results to particle image velocimetry (PIV) data of a complementary JIC experiment under a limited set of flow conditions. Here, a $k–ε$ model using both nominal and calibrated parameters is validated against PIV data acquired from a much wider variety of JIC cases, including a realistic flight vehicle. The results from the simulations using the calibrated model parameters showed considerable improvements over those using the nominal values, even for cases that were not used in the calibration procedure that defined the optimal parameters. This improvement is demonstrated using a number of quality metrics that test the spatial alignment of the jet core, the magnitudes of multiple flow variables, and the location and strengths of vortices in the counter-rotating vortex cores on the PIV planes. These results suggest that the calibrated parameters have applicability well outside the specific flow case used in defining them and that with the right model parameters, RANS solutions for the JIC can be improved significantly over those obtained from the nominal model.
Statistical estimates for random search parameters in optimization of control system
Land-atmosphere (L-A) interactions play a critical role in determining the diurnal evolution of both planetary boundary layer (PBL) and land surface temperature and moisture budgets, as well as controlling feedbacks with clouds and precipitation that lead to the persistence of dry and wet regimes. Recent efforts to quantify the strength of L-A coupling in prediction models have produced diagnostics that integrate across both the land and PBL components of the system. In this study, we examine the impact of improved specification of land surface states, anomalies, and fluxes on coupled WRF forecasts during the summers of extreme dry (2006) and wet (2007) conditions in the U.S. Southern Great Plains. The improved land initialization and surface flux parameterizations are obtained through the use of a new optimization and uncertainty module in NASA's Land Information System (LIS-OPT), whereby parameter sets are calibrated in the Noah land surface model and classified according to the land cover and soil type mapping of the observations and the full domain. The impact of the calibrated parameters on the a) spin up of land surface states used as initial conditions, and b) heat and moisture fluxes of the coupled (LIS-WRF) simulations are then assessed in terms of ambient weather, PBL budgets, and precipitation along with L-A coupling diagnostics. In addition, the sensitivity of this approach to the period of calibration (dry, wet, normal) is investigated. Finally, tradeoffs of computational tractability and scientific validity (e.g.,. relating to the representation of the spatial dependence of parameters) and the feasibility of calibrating to multiple observational datasets are also discussed.
Density functional tight binding (DFTB) models for f-element species are challenging to parametrize owing to the large number of adjustable parameters. The explicit optimization of the terms entering the semiempirical DFTB Hamiltonian related to f orbitals is crucial to generating a reliable parametrization for f-block elements, because they play import roles in bonding interactions. However, since the number of parameters grows quadratically with the number of orbitals, the computational cost for parameter optimization is much more expensive for the f-elements than for the main group elements. In this work we present a set of efficient approaches for mitigating the hurdle imposed by the large size of the parameter space. A novel group-by-orbital correction functions for two-center bond integrals was developed. With this approach the number of parameters is reduced, and it grows linearly with the number of elements, maintaining the accuracy and the number of parameters, in the case of f elements, by more than 40%. The parameter optimization step was accelerated by means of the mini-batch BFGS method. This method allows parameter optimizations with much larger training sets than other single batch methods. A stochastic optimizer was employed that helped overcome shallow local minima in the objective function. The proposed algorithm was used to parametrize the DFTB Hamiltonian for the Th–O system, which was subsequently applied to the study of ThO 2 nanoparticles. The training set consisted of 6322 unique structures, which is barely feasible with conventional optimization methods. The optimized parameter set, LANL-ThO, displays good agreement with DFT-calculated properties such as energies, forces, and structures for both clusters and bulk ThO 2 . Benefiting from the fewer number of parameters and lower computational costs for objective function evaluations, this new approach shows its potential applications in DFTB parametrization for elements with high angular momentum, which present a challenge to conventional methods.
Methods and systems for real-time engine control optimization are provided. A value of an engine performance variable is determined, a value of a first operating condition and a value of a second operating condition of a vehicle engine are detected, and initial values for a first engine control parameter and a second engine control parameter are determined based on the detected first operating condition and the detected second operating condition. The initial values for the first engine control parameter and the second engine control parameter are adjusted based on the determined value of the engine performance variable to cause the engine performance variable to approach a target engine performance variable. In order to cause the engine performance variable to approach the target engine performance variable, adjusting the initial value for the first engine control parameter necessitates a corresponding adjustment of the initial value for the second engine control parameter.
The propagation of an intense (a{sub 0}=3), short-pulse (L{approx}{lambda}{sub p}) laser through a homogeneous plasma has been investigated. Using two-dimensional simulations for a{sub 0}=3, the pulse-length and spot-size at three different plasma densities were optimized in order to get a better quality beam in laser wakefield accelerator. The study reveals that with increasing pulse-length the acceleration increases, but after a certain pulse-length (L>0.23{lambda}{sub p}) the emittance blows-up unacceptably. For spot-sizes less than that given by k{sub p0}r{sub s}=2{radical}(a{sub 0}), trapping is poor or nonexistent, and the optimal spot-size is larger. The deviation of the optimal spot-size from this formula increases as the density decreases. The efficacy of these two-dimensional simulations has been validated by running three-dimensional simulations at the highest density. It has been shown that good quality GeV-class beams can be obtained at plasma densities of {approx}10{sup 18} cm{sup -3}. The quality of the beam can be substantially improved by selecting only the high-energy peak; in this fashion an energy-spread of better than 1% and a current in tens of kA can be achieved, which are important for applications such as free-electron lasers.
In storage rings with slow resonant extraction one of the main goals of optimization is the extraction inefficiency, or the fraction of beam lost due to hitting the septum plane. A simple analytical model based on a perturbation theoretic approach is built to help determine optimal parameters. It also helps to evaluate the design performance in time and in a range of machine parameters.