SEARCH · Search NASA

Results for “Entropy minimization”

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.

SourceResearch connectionFrom yearThrough year

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 19 records

All sources NASA NTRS DOE OSTI

NASA NTRS1979

Feature selection via entropy minimization: An example using LANDSAT satellite data

The author has identified the following significant results. The minimum entropy model may provide several useful advantages over traditional techniques for processing LANDSAT data. Total computer time to conduct a complete pattern recognition process is reduced. Subjective (transformed image), as well as statistically derived information is made available to the analyst/user much earlier in the analysis process. A rapid feedback loop in which numerous training set combinations can be tested for difference and representativeness is available. Additional tests of LANDSAT data processing using the minimum entropy model are clearly justified.

Zandonella, A.↗

NASA NTRS1990

Euler analysis comparison with LDV data for an advanced counter-rotation propfan at cruise

A fine mesh Euler solution of the F4/A4 unducted fan (UDF) model flowfield is compared with laser Doppler velocimeter (LDV) data taken in the NASA Lewis 8- by 6-Foot Supersonic Wind Tunnel. The comparison is made primarily at one axial plane downstream of the front rotor where the LDV particle lag errors are reduced. The agreement between measured and predicted velocities in this axial plane is good. The results show that a dense mesh is needed in the centerbody stagnation region to minimize entropy generation that weakens the aft row passage shock. The predicted radial location of the tip vortex downstream of the front rotor agrees well with the experimental results but the strength is overpredicted. With 40 points per chord line, the integrated performance quantities are nearly converged, but more points are needed to resolve passage shocks and flow field details.

Miller, Christopher J.↗

NASA NTRS1990

Euler analysis comparison with LDV data for an advanced counter-rotation propfan at cruise

Miller, Christopher J.↗

NASA NTRS1996

Algorithm For A Self-Growing Neural Network

CID3 algorithm simulates self-growing neural network. Constructs decision trees equivalent to hidden layers of neural network. Based on ID3 algorithm, which dynamically generates decision tree while minimizing entropy of information. CID3 algorithm generates feedforward neural network by use of either crisp or fuzzy measure of entropy.

Cios, Krzysztof J.↗

NASA NTRS1993

Multi-Interval Discretization of Continuous-Valued Attributes for Classification Learning

Since most real-world applications of classification learning involve continuous-valued attributes, properly addressing the discretization process is an important problem. This paper addresses the use of the entropy minimization heuristic for discretizing the range of a continuous-valued attribute into multiple intervals.

discretization continuous-valued attributes classi↗

NASA NTRS1989

An entropy method for induced drag minimization

A fundamentally new approach to the aircraft minimum induced drag problem is presented. The method, a 'viscous lifting line', is based on the minimum entropy production principle and does not require the planar wake assumption. An approximate, closed form solution is obtained for several wing configurations including a comparison of wing extension, winglets, and in-plane wing sweep, with and without a constraint on wing-root bending moment. Like the classical lifting-line theory, this theory predicts that induced drag is proportional to the square of the lift coefficient and inversely proportioinal to the wing aspect ratio. Unlike the classical theory, it predicts that induced drag is Reynolds number dependent and that the optimum spanwise circulation distribution is non-elliptic.

Greene, George C.↗

NASA NTRS1984

Discrimination of coherent features in turbulent boundary layers by the entropy method

Entropy in information theory is defined as the expected or mean value of the measure of the amount of self-information contained in the ith point of a distribution series x sub i, based on its probability of occurrence p(x sub i). If p(x sub i) is the probability of the ith state of the system in probability space, then the entropy, E(X) = - sigma p(x sub i) logp (x sub i), is a measure of the disorder in the system. Based on this concept, a method was devised which sought to minimize the entropy in a time series in order to construct the signature of the most coherent motions. The constrained minimization was performed using a Lagrange multiplier approach which resulted in the solution of a simultaneous set of non-linear coupled equations to obtain the coherent time series. The application of the method to space-time data taken by a rake of sensors in the near-wall region of a turbulent boundary layer was presented. The results yielded coherent velocity motions made up of locally decelerated or accelerated fluid having a streamwise scale of approximately 100 nu/u(tau), which is in qualitative agreement with the results from other less objective discrimination methods.

Corke, T. C.↗

NASA NTRS2021

An Information Theory Approach to Physical Domain Discovery

The project of physics discovery is often equivalent to finding the most concise description of a physical system. The description with optimum predictive capability for a dataset generated by a physical system is one that minimizes both predictive error on the dataset and the complexity of the description. The discovery of the governing physics of a system can therefore be viewed as a mathematical optimization problem. We outline here a method to optimize the description of arbitrarily complex physical systems by minimizing the entropy of the description of the system. The Recursive Domain Partitioning (RDP) procedure finds the optimum partitioning of each physical domain into subdomains, and the optimum predictive function within each subdomain. Penalty functions are introduced to limit the complexity of the predictive function within each domain. Examples are shown in 1D and 2D. In 1D, the technique effectively discovers the elastic and plastic regions within a stress-strain curve generated by simulations of amorphous carbon material, while in 2D the technique discovers the free-flow region and the inertially-obstructed flow region in the simulation of fluid flow across a plate.

Daniel Shea↗

NASA NTRS1990

A Boltzmann machine for the organization of intelligent machines

A three-tier structure consisting of organization, coordination, and execution levels forms the architecture of an intelligent machine using the principle of increasing precision with decreasing intelligence from a hierarchically intelligent control. This system has been formulated as a probabilistic model, where uncertainty and imprecision can be expressed in terms of entropies. The optimal strategy for decision planning and task execution can be found by minimizing the total entropy in the system. The focus is on the design of the organization level as a Boltzmann machine. Since this level is responsible for planning the actions of the machine, the Boltzmann machine is reformulated to use entropy as the cost function to be minimized. Simulated annealing, expanding subinterval random search, and the genetic algorithm are presented as search techniques to efficiently find the desired action sequence and illustrated with numerical examples.

Moed, Michael C.↗

NASA NTRS1987

Analytical design of intelligent machines

The problem of designing 'intelligent machines' to operate in uncertain environments with minimum supervision or interaction with a human operator is examined. The structure of an 'intelligent machine' is defined to be the structure of a Hierarchically Intelligent Control System, composed of three levels hierarchically ordered according to the principle of 'increasing precision with decreasing intelligence', namely: the organizational level, performing general information processing tasks in association with a long-term memory; the coordination level, dealing with specific information processing tasks with a short-term memory; and the control level, which performs the execution of various tasks through hardware using feedback control methods. The behavior of such a machine may be managed by controls with special considerations and its 'intelligence' is directly related to the derivation of a compatible measure that associates the intelligence of the higher levels with the concept of entropy, which is a sufficient analytic measure that unifies the treatment of all the levels of an 'intelligent machine' as the mathematical problem of finding the right sequence of internal decisions and controls for a system structured in the order of intelligence and inverse order of precision such that it minimizes its total entropy. A case study on the automatic maintenance of a nuclear plant illustrates the proposed approach.

Saridis, George N.↗

NASA NTRS1984

Optimation of cooled shields in insulations

A method to optimize the location, temperature, and heat dissipation rate of each cooled shield inside an insulation layer was developed. The method is based on the minimization of the entropy production rate which is proportional to the heat leak across the insulation. It is shown that the maximum number of shields to be used in most practical applications is three. However, cooled shields are useful only at low values of the overall, cold wall to hot wall absolute temperature ratio. The performance of the insulation system is relatively insensitive to deviations from the optimum values of the temperature and location of the cooling shields. Design curves for rapid estimates of the locations and temperatures of cooling shields in various types of insulations, and an equation for calculating the cooling loads for the shields are presented.

Chato, J. C.↗

NASA NTRS2000

Progress in the Development of a Class of Efficient Low Dissipative High Order Shock-capturing Methods

In a series of papers, Olsson (1994, 1995), Olsson & Oliger (1994), Strand (1994), Gerritsen Olsson (1996), Yee et al. (1999a,b, 2000) and Sandham & Yee (2000), the issue of nonlinear stability of the compressible Euler and Navier-Stokes Equations, including physical boundaries, and the corresponding development of the discrete analogue of nonlinear stable high order schemes, including boundary schemes, were developed, extended and evaluated for various fluid flows. High order here refers to spatial schemes that are essentially fourth-order or higher away from shock and shear regions. The objective of this paper is to give an overview of the progress of the low dissipative high order shock-capturing schemes proposed by Yee et al. (1999a,b, 2000). This class of schemes consists of simple non-dissipative high order compact or non-compact central spatial differencings and adaptive nonlinear numerical dissipation operators to minimize the use of numerical dissipation. The amount of numerical dissipation is further minimized by applying the scheme to the entropy splitting form of the inviscid flux derivatives, and by rewriting the viscous terms to minimize odd-even decoupling before the application of the central scheme (Sandham & Yee). The efficiency and accuracy of these scheme are compared with spectral, TVD and fifth- order WENO schemes. A new approach of Sjogreen & Yee (2000) utilizing non-orthogonal multi-resolution wavelet basis functions as sensors to dynamically determine the appropriate amount of numerical dissipation to be added to the non-dissipative high order spatial scheme at each grid point will be discussed. Numerical experiments of long time integration of smooth flows, shock-turbulence interactions, direct numerical simulations of a 3-D compressible turbulent plane channel flow, and various mixing layer problems indicate that these schemes are especially suitable for practical complex problems in nonlinear aeroacoustics, rotorcraft dynamics, direct numerical simulation or large eddy simulation of compressible turbulent flows at various speeds including high-speed shock-turbulence interactions, and general long time wave propagation problems. These schemes, including entropy splitting, have also been extended to freestream preserving schemes on curvilinear moving grids for a thermally perfect gas (Vinokur & Yee 2000).

Yee, H. C.↗

NASA NTRS1972

Sonic-boom minimization.

There have been many attempts to reduce or eliminate the sonic boom. Such attempts fall into two categories: (1) aerodynamic minimization and (2) exotic configurations. In the first category changes in the entropy and the Bernoulli constant are neglected and equivalent body shapes required to minimize the overpressure, the shock pressure rise and the impulse are deduced. These results include the beneficial effects of atmospheric stratification. In the second category, the effective length of the aircraft is increased or its base area decreased by modifying the Bernoulli constant a significant fraction of the flow past the aircraft. A figure of merit is introduced which makes it possible to judge the effectiveness of the latter schemes.

Seebass, R.↗

NASA NTRS2000

Extension of Efficient Low Dissipative High Order Schemes for 3-D Curvilinear Moving Grids

The efficient low dissipative high order schemes proposed by Yee et al. is formulated for 3-D curvilinear moving grids. These schemes consists of a high order base schemes combined with nonlinear characteristic filters. The amount of numerical dissipation is minimized by applying the schemes to the entropy splitting form of the inviscid flux derivatives. The analysis is given for a thermally perfect gas. The main difficulty in the extension of higher order schemes that were formulated in Cartesian coordinates to curvilinear moving grids is the higher order transformed metric evaluations. The higher order numerical evaluation of the transformed metric terms to insure freestream preservation is done in a coordinate invariant manner. The formulation is an improvement over existing formulation of high order scheme in curvilinear moving grids.

Vinokur, Marcel↗

NASA NTRS1990

Test images for the maximum entropy image restoration method

One of the major activities of any experimentalist is data analysis and reduction. In solar physics, remote observations are made of the sun in a variety of wavelengths and circumstances. In no case is the data collected free from the influence of the design and operation of the data gathering instrument as well as the ever present problem of noise. The presence of significant noise invalidates the simple inversion procedure regardless of the range of known correlation functions. The Maximum Entropy Method (MEM) attempts to perform this inversion by making minimal assumptions about the data. To provide a means of testing the MEM and characterizing its sensitivity to noise, choice of point spread function, type of data, etc., one would like to have test images of known characteristics that can represent the type of data being analyzed. A means of reconstructing these images is presented.

Mackey, James E.↗

NASA NTRS2016

Temperature-Staged Thermal Energy Storage Enabling Low Thermal Exergy Loss Reflux Boiling in Full Spectrum Solar Systems

Hybrid full spectrum solar systems (FSSS) designed to capture and convert the full solar wavelength spectrum use hybrid solar photovoltaic/thermodynamic cycles that require low thermal exergy loss systems capable of transferring high thermal energy rates and fluxes with very low temperature differentials and losses. One approach to achieving this capability are high-heat-flux reflux boiling systems that take advantage of high heat transfer boiling and condensation mechanisms. Advanced solar systems are also intermittent by their nature and their electrical generation is often out-of-phase with electric utility power demand, and their required power system cycling reduces efficiency, performance (dispatch ability), lifetime, and reliability. High temperature thermal energy storage (TES) at 300-600°C enables these reflux boiling systems to simultaneously store thermal energy internally to increase the energy dispatch ability of the associated solar system, as this can increase the power generation profile by several hours (up to 6-10 hours) per day. Many TES phase change materials (PCM’s) exist including KNO3, NaNO3, LiBr/KBr, MgCl2/NaCl/KCl, Zn/Mg, and CuCl/NaCl, which have various operating melting points and different latent heats of fusion. Common, cost effective TES PCM's are FeCl2/NaCl/KCl mixtures, whose phase change temperature can be varied and controlled by simple composition adjustments. This paper presents and discusses unique "temperature-staged" thermal energy storage configurations using these TES materials and analysis of such systems integrated into high-heat-flux reflux boiling systems. In this specific application, the TES materials are designed to operate at staged temperatures surrounding an operating design point near 350°C, while providing 18 kW of source heat transfer to operate a thermoacoustic power system during off-sun conditions (e.g., temporary cloud conditions, after sun-down). This work discusses relevant configurations, and critical thermal and entropy models of the TES configurations, which show the inherent minimization of thermal exergy during critical heat transfers within the configurations and systems envisioned.

Hendricks, Terry J.↗

NASA NTRS2001

Entropy Splitting for High Order Numerical Simulation of Vortex Sound at Low Mach Numbers

A method of minimizing numerical errors, and improving nonlinear stability and accuracy associated with low Mach number computational aeroacoustics (CAA) is proposed. The method consists of two levels. From the governing equation level, we condition the Euler equations in two steps. The first step is to split the inviscid flux derivatives into a conservative and a non-conservative portion that satisfies a so called generalized energy estimate. This involves the symmetrization of the Euler equations via a transformation of variables that are functions of the physical entropy. Owing to the large disparity of acoustic and stagnation quantities in low Mach number aeroacoustics, the second step is to reformulate the split Euler equations in perturbation form with the new unknowns as the small changes of the conservative variables with respect to their large stagnation values. From the numerical scheme level, a stable sixth-order central interior scheme with a third-order boundary schemes that satisfies the discrete analogue of the integration-by-parts procedure used in the continuous energy estimate (summation-by-parts property) is employed.

Mueller, B.↗

NASA NTRS1989

A physical model for earthquakes. III - Thermodynamical approach and its relation to nonclassical theories

An earthquake theory based on fundamental thermodynamical considerations was developed using two principles. The first principle is that physical systems, including fault systems, tend toward a state of minimal Helmholtz free energy, which corresponds to static equilibrium. The second postulate is that, if the system is not at minimal free energy, thermodynamic forces arise which drive the system back toward a minimum. According to this theory, the propagation of a fracture, or the formation of a slip on a frictional surface, are manifestations of the physics which attempt to return the system to minimal free energy. It is shown that the process of return to minimal free energy by the system corresponds to the production of entropy, thereby rendering the process irreversible.

Rundle, John B.↗