Showing papers by "Georgia Institute of Technology published in 1979"

Social processes and proofs of theorems and programs

Abstract: It is argued that formal verifications of programs, no matter how obtained, will not play the same key role in the development of computer science and software engineering as proofs do in mathematics. Furthermore the absence of continuity, the inevitability of change, and the complexity of specification of significantly many real programs make the formal verification process difficult to justify and manage. It is felt that ease of formal verification should not dominate program language design.

The processing of hexagonally sampled two-dimensional signals

Abstract: Two-dimensional signals are normally processed as rectangularly sampled arrays; i.e., they are periodically sampled in each of two orthogonal independent variables. Another form of periodic sampling, hexagonal sampling, offers substantial savings in machine storage and arithmetic computations for many signal processing operations. In this paper, methods for the processing of two-dimensional signals which have been sampled as two-dimensional hexagonal arrays are presented. Included are methods for signal representation, linear system implementation, frequency response calculation, DFT calculation, filter design, and filter implementation. These algorithms bear strong resemblances to the corresponding results for rectangular arrays; however, there are also many important differences. Some comparisons between the two methods for representing planar data will also be presented.

A modification of the Hynes method for estimating secondary production with particular significance for multivoltine populations

Abstract: The accepted procedure for de termining production of multivoltine invertebrates by use of the Hynes method is to multiply the Hynes value by the number of generations per year. For aquatic insects, if pupal, adult, or egg stages comprise a significant portion of total generation time, this procedure will underestimate production. For crustaceans, if reproduction occurs before attaining the final size class, the procedure, using generation time, will overestimate production. It is necessary to multiply the Hynes value by 365/CPI, where CPI is the cohort production interval (in days) from hatching to the attainment of the largest aquatic size class.

Stochastic behavior of a quantum pendulum under a periodic perturbation

Abstract: This paper discusses a numerical technique for computing the quantum solutions of a driver pendulum governed by the Hamiltonian $$H = (p_\theta ^2 /2m\ell ^2 ) - [m\ell ^2 \omega _o ^2 \cos \theta ] \delta _p (t/T) ,$$ where pe is angular momentum, ϑ is angular displacement, m is pendulum mass, l is pendulum length, ωO2 = g/l is the small displacement natural frequency, and where δp (t/T) is a periodic delta function of period T. The virtue of this rather singular Hamiltonian system is that both its classical and quantum equations of motion can be reduced to mappings which can be iterated numerically and that, under suitable circumstances, the motion for this system can be wildly chaotic. Indeed, the classical version of this model is known to exhibit certain types of stochastic behavior, and we here seek to verify that similar behavior occurs in the quantum description. In particular, we present evidence that the quantum motion can yield a linear (diffusive-like) growth of average pendulum energy with time and an angular momentum probability distribution which is a time-dependent Gaussian just as does the classical motion. However, there are several surprising distinctions between the classical and quantum motions which are discussed herein.

A Rheological Model for Elastohydrodynamic Contacts Based on Primary Laboratory Data

Abstract: A shear rheological model based on primary laboratory data is proposed for concentrated contact lubrication. The model is a Maxwell model modified with a limiting shear stress. Three material properties are required: Low shear stress viscosity, limiting elastic shear modulus, and the limiting shear stress the material can withstand. All three are functions of temperature and pressure. In applying the model to EHD contacts the predicted response possesses the characteristics expected from several experiments reported in the literature and, in one specific case where direct comparison could be made, good numerical agreement is shown.

Resilient Properties of Subgrade Soils

Abstract: Resilient properties of 50 typical Illinois fine-grained soils were evaluated. Soil resilient properties can be related to soil characteristics (texture, plasticity, organic carbon content, AASHTO group index). Current soil classification procedures do not group fine-grained soils into groups with distinctive resilient properties. Moisture-density conditions and degree of saturation significantly influence the resilient properties of fine-grained soils. Average resilient properties for various soil classification groups were determined. Regression equations were developed for estimating resilient properties based on soil characteristics and degree of saturation.

Shear strength measurements of lubricants at high pressure

Abstract: Measurements of lubricant shear rheological behavior in the amorphous solid region and near the liquid-solid transition are reported on three lubricants under pressure. Elastic, plastic and viscous behavior was observed. The maximum yield shear stress (limiting shear stress) is a function of temperature and pressure and is believed to be the property which determines the maximum traction in elastohydrodynamic contacts such as traction drives.

A Comparison of the Internal Validity of Alternative Parameter Estimation Methods in Decompositional Multiattribute Preference Models

Abstract: Since 1971, interest in the use of decompositional multiattribute preference models in marketing has been increasing. The applications have varied in terms of the type of data used, behavior predic...

Holographic grating formation in photorefractive crystals with arbitrary electron transport lengths

Abstract: The time development of optically induced refractive‐index hologram gratings in photrefractive electro‐optic crystals such as lithium niobate and bismuth silicon oxide is analyzed allowing arbitrary electron‐transport lengths and nonsinusoidal grating shapes. In this analysis, diffusion and drift in applied and space‐charge fields as well as the bulk photovoltaic effect (when applicable) are included. Results for the photoinduced space‐charge field are presented for a range of realistic experimental conditions. It is shown that the spatial phase of the space‐charge field and therefore the beam coupling is strongly dependent on the electron‐transport lengths, whereas the magnitude of the field and therefore the diffraction efficiency is not dependent. It is shown that for paraelectric materials (such as BSO) the results of short‐ and long‐transport‐length analyses must converge for long recording times (saturation).

Statistical synthesis of power system functional load models

Abstract: Functional load models characterize the load demand portion of a component load when the influence factors are weather, and human use patterns, etc. Physically based functional load models for individual components are proposed. From these the functional load model of a large number of similar components is obtained by statistical aggregation.

Fatigue Crack initiation and microcrack growth in 2024-T4 and 2124-T4 aluminum alloys

Abstract: A metallograph was mounted directly on a closed-loop electrohydraulic testing unit and initiation of fatigue cracks was directly observed on polished notches at magnifications up to 800 times in aluminum alloys 2024 and 2124 in the T-4 condition. The latter is a high purity version of 2024 and contains considerably fewer constituent particles. At high stresses on the notch surface the fatigue cracks initiated on coarse slip lines in both alloys. At low stresses almost all of the cracks in 2024 initiated in the matrix adjacent to constituent particles. In 2124 at low stresses 50 pct of the cracks initiated near constituent particles and 50 pct in the matrix not near constituent particles. The probability that a constituent particle in 2024 initiates a fatigue crack falls off very rapidly as the particle size decreases below 6 μm. Growth of microcracks is impeded by grain boundaries.

Internal radiation therapy of hepatic cancer.

Abstract: Established cancer in the liver can, in selected patients who have a good arterial circulation in these tumors, be effectively treated by intrahepatic artery radioactive yttrium-90 resin microspheres. Even in unselected patients treated in the last five years by the author, 17 of 25 patients treated have had good objective regression of cancers, improvement of symptoms and prolongation of life. Treatment is relatively simple and associated with few side effects. For adjuvant therapy of colon cancer having positive nodes (Dukes C), internal radiation therapy of the liver is best done with Phosphorus-32 Colloid passed through the circulation of the gut to be effectively and homogeneously trapped by the Kupffer cells of the liver. Four such patients have been subjected to a pilot study--three of the four are doing well without significant side effects and no evidence of liver cancer after two years. When the fourth died of brain metastases, he too had less liver cancer than would be expected.

Some relationships between lagrangian and surrogate duality in integer programming

Abstract: Lagrangian dual approaches have been employed successfully in a number of integer programming situations to provide bounds for branch-and-bound procedures. This paper investigates some relationship between bounds obtained from lagrangian duals and those derived from the lesser known, but theoretically more powerful surrogate duals. A generalization of Geoffrion's integrality property, some complementary slackness relationships between optimal solutions, and some empirical results are presented and used to argue for the relative value of surrogate duals in integer programming. These and other results are then shown to lead naturally to a two-phase algorithm which optimizes first the computationally easier lagrangian dual and then the surrogate dual.

Role of iron reduction in the conversion of smectite to illite in bentonites in the disturbed belt, montana

Abstract: Abstraet--Cretaceous bentonites were collected in outcrop from the Sweetgrass Arch and the Disturbed Belt in Montana. The mixed-layer illite-smectite (I/S) components of the bentonites from the Sweetgrass Arch have from 0 to 25% illite layers and no detectable structural Fe 2+, whereas the samples from the Disturbed Belt have from about 25 to 90% illite layers, and all contain Fe 2+. A positive correlation (r = 0.89) exists between the percentage of structural iron that is Fe z+ and the amount of fixed interlayer K in the US. The higher percentage of illite layers in the samples from the Disturbed Belt is attributed to reactions related to elevated temperatures caused by burial beneath thrust sheets. The increase in Fe2+/Fe 3+ with increasing percentages of illite layers is tentatively attributed to a redox reaction involving the oxidation of organic matter. Although there is no statistical evidence for an increase in octahedral charge with an increase in illite layers when all the samples are considered together, iron reduction may have contributed as much as 10 to 30% of the increase in total structural-charge that occurred in any given sample during metamorphism. The remaining structural charge increase can be attributed to the substitution of Al 3+ for Si 4§ in the tetrahedral sites.

Line Plume and Ocean Outfall Dispersion

Abstract: The three-dimensional flowfield created by a line plume of finite length in a steady unstratified current was investigated experimentally and theoretically. The results are applicable to ocean sewer outfall design. Even though the diffuser length may be much greater than the water depth, the flowfield cannot be considered to be two dimensional. The most important dynamic parameter is a Froude number, F, given by the ratio of the current speed cubed to the buoyancy flux discharged per unit diffuser length. Different mixing regimes can occur depending on the value of F and the plume can be attached to the lower boundary. Dilution depends on current speed and direction only for F>0.1. For F>0.1, dilution increases with current speed and diffusers perpendicular to the current will result in greater dilution than if parallel. The surface wastefield spreads rapidly in a parallel current, showing that Y or similarly shaped diffusers are not essential to produce a widely dispersed wastefield. The buoyant spreading of the surface wastefield is a complex phenomenon, having growth laws that change with distance downstream.

A time-space tradeoff for sorting on non-oblivious machines

Abstract: A model of computation is introduced which permits the analysis of both the time and space requirements of non-oblivious programs. Using this model, it is demonstrated that any algorithm for sorting n inputs which is based on comparisons of individual inputs requires time-space product proportional to n2. Uniform and non-uniform sorting algorithms are presented which show that this lower bound is nearly tight.

Interannual and Month-to-Month Variations of Wind Speed

Abstract: Spatial cross correlations and interannual and month-to-month variations of monthly mean wind speed were studied at 40 sites throughout the United States. Sites were selected on the basis of availability of 10 or more years of data from a fixed anemometer location and height and a climatological mean speed of 5 m s−1 (11 mph) or higher. Spatial cross correlations of monthly deviations from climatic means were found to be about 0.5 for sites separated less than 200 km, with annual mean wind deviations from the climatic mean correlated with coefficient value 0.32 for similarly separated sites. Applications for using nearby “climatological” site wind speed data to adjust short-term “candidate” site data are examined with several methods. The best results show only minimal improvement in estimating the long-term annual mean over that obtained from one year of on-site data. These results indicate that, for candidate wind energy site evaluation, on-site data must be relied on more than originally consi...

The Generic Nature of Optimality Conditions in Nonlinear Programming

Abstract: First-and second-order optimality conditions are treated for parameterized classes of nonlinear programming problems in Rn. Under certain assumptions, it is shown that almost all problems in the class are such that every local minimizer satisfies the strong form of the optimality conditions.

Coupling of optical and mechanical properties in crystalline polymers

Abstract: The modules and birefringence of oriented polymers are known to couple through the structural and intrinsic properties of the polymer. Recent developments in both optical and mechanical theories have allowed this coupling to be expanded so that it includes predictions of modulus and refractive indices of uniaxially-oriented polymer films at various angles to the draw direction. The validity of these theories has been tested in this study by predicting the angular dependence of the optical and mechanical properties of seven uniaxiallyoriented films of isotactic polypropylene. The present study also provides intrinsic material properties associated with the crystalline and noncrystalline components of the polymer, and consequently allows the identification of the structural parameters that control the properties of ultra-high oriented isotactic polypropylene.

A model relating low cycle fatigue properties and microstructure to fatigue crack propagation rates

Abstract: — The fatigue crack propagation rate is related to the cyclic flow stress and the cyclic ductility of materials by means of a ductility exhaustion mechanism. The da/dN vs. ΔK relationship is derived from the assumption that an elemental crack extension occurs when the sum of the fractional fatigue life of an element at various cycles equals unity. A comparison of predicted crack growth curves with those obtained experimentally for representative alloys shows good agreement. The threshold ΔK behavior has been explained in terms of the existence of a threshold plastic strain amplitude.

An exact branch‐and‐bound procedure for the quadratic‐assignment problem

Abstract: The quadratic-assignment problem is a difficult combinatorial problem which still remains unsolved. In this study, an exact branch-and-bound procedure, which is able to produce optimal solutions for problems with twelve facilities or less, is developed. The method incorporates the concept of stepped fathoming to reduce the effort expended in searching the decision trees. Computational experience with the procedure is presented.

Conveyor Theory: A Survey

Abstract: This paper surveys the research which has been published in the area of conveyor theory. Deterministic and probabilistic approaches, as well as descriptive and normative approaches to modeling conveyor systems are discussed. Areas for further study are suggested.

Stochastic theory of multistate diffusion in perfect and defective systems. I. Mathematical formalism

Abstract: A stochastic theory of multistate transport on ideal and defective lattices is presented. A continuous-time random-walk formalism with the inclusion of internal states is used in the derivation of a matrix (whose dimensions are the number of internal states) probability propagator. The propagator describing motion on an ideal lattice is modified owing to the presence of a periodic arrangement of defects. The expression for the modified probability propagator greatly simplifies in the long-time (diffusion) limit. In this limit the presence of defects renormalizes the ideal-lattice propagator through the inclusion of a self-energy-type term which depends upon the concentration of defect sites and the differences in transition rates of the propagating species associated with these sites and ideal ones. The formalism enables the study of complex diffusion mechanisms, as illustrated in the following paper, and allows the calculation of observables such as positional moments, diffusion coefficients, and occupation probability of states.