The effect of controlled atmosphere (CA) storage, and antioxidant treatment on polyphenoloxidase (EC 1. 1418.1:PPO), peroxidase (EC11.11.1.7:POD), and superoxide dismutase (EC 1.15.1.1:S0D) activities and superficial scald and soft scald development in 'Virginiagold','Stayman', and 'Rome' apples (Malus domestica Borkh.) was investigated. 'Virginiagold' apples treated postharvest with an aqueous solution of diphenylamine (DPA) + 1,2, dihydro -6- ethoxy -2,2,4-trimethylquinoline (ethoxyquin) and stored in CA exhibited lower soft scald incidence and higher firmness than comparable fruits stored at O°C in air storage. Fruit firmness and titratable acids (TA) decreased continuously, while ethylene increased in storage. Titratable acids and firmness were positively correlated, while TA and ethylene and firmness and ethylene were negatively correlated. 'Stayman' and 'Rome' apples analyzed during storage showed increase in levels of λ±-famesene and its oxidation products conjugated trienes. During storage fruits showed an increase in the activities of polyphenoloxidase (PPO), peroxidase (POD), and superoxide dismutase (SOD), which reached a maximum when scald symptoms in 'Stayman' were observed. Our results indicate that 'Stayman' apples exhibited scald and 'Rome' did not, because 'Stayman' apples had a ten fold higher PPO activity than 'Rome.' / Master of Science

The traditional Euclidean approach to problem solving in AI has always designed representations for a domain and then spent considerable effort on the methods of efficiently searching the representation in order to extract the desired information. We feel that the emphasis in problem solving should be on the automated construction of the knowledge representation and not on the searching of the representation. This thesis proposes and implements an alternative approach: that of analogical representation. Analogical representation differs from the Euclidean methodology in that it creates a representation for the data from which the acquisition of information is done by simple 'observation.' It is not our goal to propose a system that reduces the NP-hard problem of temporal reasoning to a lower complexity. Our approach simply minimizes the number of times that we must pay the exponential expense. Furthermore, the representation can encode uncertainty and unknownness in an efficient manner. This allows for 'intelligent' creation of a representation and removes the 'mindless' mechanical search techniques from information retrieval, placing the computational effort where it should be: on representation construction. / Master of Science

The strength and stiffness characteristics of longspan composite open-web steel joists are studied. Eight full scale single span composite joists were tested to failure to evaluate their performance in both the elastic and inelastic ranges. The test specimens' spans, depths, slab depth, and amount of shear connection was varied. Analytical methods are presented to predict both the ultimate strength and the stiffness of the composite joist. A non-linear finite element model is presented to predict the behavior of the composite joist in the elastic and inelastic regions. / Master of Science

T’ao Yüan-ming (365-427 A.D.) remains China's greatest medieval poet. While no adequate biographical sources exist, the poet’s own work presents the most intimate picture of T’ao Yüan-ming himself, in addition to reflecting various aspects of his historical and cultural era. This essay explores T’ao Yüan-ming’s character as reflected in his verse. When poised against a broader historical and cultural background, T’ao Yüan-ming’s poetry goes furthest in revealing the poet as an individual as well as a product of his times. / Master of Arts

Network data restructuring is an essential function in Geographic Information Systems (GIS) when adapted to transportation. Implementing effective data restructuring models in GIS allows users to collect and maintain data in the format with which they are most familiar while allowing others to utilize it in a format they require. Aggregation and disaggregation of network data facilitates storage, display and plotting times. There could however be important adverse effects. The effects of generalizing attribute data for aggregation/disaggregation is being researched here. Milepoint referenced data in road inventory files provides valuable network information for transportation research. Individual records in these files represent variable length sections of roads. A new record is created each time a highway attribute changes. Consequently, a segment of road between two intersections may be represented by several records in a road inventory file. Further, all attributes in these records are associated with both directions of travel along a road. Many transportation analysis models require networks to be represented by a node-link structure where nodes symbolize an intersection of two or more roads. Further, if a road is two directional, it is represented by two links each of which has its own set of attributes. To utilize road inventory data in these analysis models, network information has to be converted into an intersection/link format. This process involves aggregating· and disaggregating attribute data to represent longer and shorter road segments and also disaggregating data into bi-directional information. This thesis describes data conversion efforts needed to produce intersectional-link network representations. The development of a microcomputer model for data conversion is detailed and application issues and model sensitivities are addressed. / Master of Science

Steel is a material with a high tensile shear strength and is relatively inexpensive to produce. Further, it is widely used in industry and its use in adhesive bonding applications is increasing. Adhesively bonded structures offer several advantages over mechanical fastening such as reductions in cost and weight and also provide uniform stress distribution. The goal of this research is to study the effect of four metal alkoxides on the adhesive bonding of steel. The topography and chemical composition of gritblasted, alkoxide primed surfaces were determined using primarily scanning electron microscopy and x-ray photoelectron spectroscopy. Gritblasted steel was primed with the metal alkoxide solutions and then bonded with both polyethersulfone (thermoplastic) and FM 300U, a structural epoxy (thermoset). Bond durabilities were studied using wedge specimens. The two aluminum alkoxide primed steel surfaces generally showed slower crack propagation than titanium and silicon alkoxide primed surfaces. Bond strength information was obtained from lap shear joints that were primed with aluminum tri-sec-butoxide, titanium (IV) butoxide and control samples where no primer was used. Statistical analysis showed that there were significant differences in bond strengths, with the aluminum alkoxide primed surfaces showing higher breaking strengths. Further, the effect of applied stress and temperature on bond strength was determined. Lap shear joints that were primed and unprimed and bonded with PES and FM 300U were loaded to 10-40%, depending on the ultimate breaking strength and placed in a stress durability tester (immersion in deionized water at 80°C). A significant decrease in breaking strength was observed between the stress durability samples and the lap shear joints tested at room temperature. The failure surfaces of wedge, lap shear and stress durability samples were analyzed using scanning electron microscopy, energy dispersive x-ray analysis, x-ray photoelectron spectroscopy and Auger electron spectroscopy. / Master of Science

Experiments were performed with single, sonic, helium jets at downstream angles of 15° and 30° relative to the free stream to determine their mixing, penetration and total pressure loss when injected into a supersonic air cross flow. From this information, the performance of these jets as fuel injectors in a supersonic combustion ramjet (scramjet) combustion chamber was estimated. Both injection angle jets were made flush to the wind tunnel wall. The jets were injected into a Mach 3 free stream with a total pressure of 6.5 atm, a total temperature of 283 K and a Reynolds Number of 52.5x10⁶ /m. The flow field of each injection angle was documented at jet expansion ratios of one and five. Spark schlieren and nanoshadowgraph methods were used to visualize each flowfield. At axial stations 20, 40, and 90 jet diameters downstream of each jet, continuous vertical profiles of flow quantities were made. Profiles were taken at seven lateral stations including the jet centerline at each axial station. Spacing between the lateral stations was one jet diameter. This data yielded profiles of helium concentration, Mach number, static temperature, static pressure, density, flow speed, mass flux, total pressure, and total temperature. The different injection schemes were then compared on the basis of helium mass fraction decay, the distance required to reach the stochiometric H₂-air concentration and total pressure loss. For all cases except the 15° jet with an expansion ratio of one, large eddies were observed to penetrate into the free stream. These eddies were believed to significantly enhance large scale mixing. The jet cores of the underexpanded jets had bifurcated 20 jet diameters downstream of the injection point, but had re-united by the 40 diameter station. Wandering of the jet core about the geometric centerline was observed for all cases. The decay rates increased rapidly with the jet to free stream dynamic pressure ratio until about 1.5 where the decay rate leveled off. This indicated that there was no significant increase in mixing from increasing the dynamic pressure ratio of the present jets past 1.5. The decay rate of the present 30°, matched pressure case was about 16 percent greater than that of a normal jet at similar dynamic pressure and expansion ratios. These results were reflected in the distances required to reach the stochiometric H₂-air concentration. The 15° jet with an expansion ratio of one had the lowest total pressure loss. It was concluded that injection at low downstream angle shows promise for application to scramjet fuel injection and merits further study. / Master of Science

A series of experiments was performed on groups of samples made from materials currently used to manufacture tooling for extruders to determine if the ASTM G65 dry sand rubber wheel abrasion test could be used as an accelerated test to evaluate candidate materials. Samples were tested in the heat treated condition and after surface modification by plasma ion nitriding. The range of materials tested included medium and high alloy steels and steel bonded carbide composites. The abrasives used were AFS 50/70 test sand and Dresser Glasgrain crushed fused silica. Evaluation of test wear scars and wear debris from the tests using AFS 50/70 showed that delamination was the primary wear mechanism for the composite materials, with some ploughing and microcutting, while ploughing and microcutting were the primary mechanisms in the wear of the steels. Evaluation of parts made from a composite material which were removed from service indicated that matrix erosion was the primary wear mechanism. Tests with Glasgrain fused silica as the abrasive yielded wear scars with the same morphology as the parts returned from service, but the very poor flow characteristics of this material caused inconsistency in the supply of this abrasive to the contact region. Interrupted tests showed that the wear rate was constant for the steels in the non-nitrided condition. After nitriding, the wear rate increased with test duration. The nitriding was found to act as a barrier coating providing an initial period of very low wear until the nitride layer is broached. The wear rate then increases to approximate the wear rate of the non-nitrided samples. It was found that the friction force alters the location of the maximum normal force, shifting the point of greatest contact force toward the entry end of the wear scar. / Master of Science

The line-of-sight angle (LOSA) was investigated for a simple tracking task. The distance from the eyes of the subject to the image was varied through the use of optical lenses. The subject was seated in a chair with 5 different back angles: 90 degrees, 111.6 degrees, 125 degrees, 158.5 degrees, and 177.1 degrees. No significant difference in the selection of LOSA was found between image distances of .5 meters, 1 meter, 2 meters, and infinity. The subjects' selection of LOSA was affected by the change in seat inclination angles. / Master of Science

The simulation of an algorithm for the active control of combustion noise generated by a turbulent flame produced by an open burner is developed in this thesis. The simulation includes (1) an autoregressive model of real combustion noise, (2) a feedback loop based on the "observer" method, (3) a model of the transfer function between the acoustic driver and the sensor through the flame, and (4) a method to take into account the time-delay due the calculation of the algorithm. A practical implementation of the control strategy is also proposed. An attenuation of up to 40 dB is obtained in the 0 - 3000 Hz band, decreasing with the time-delay required for creation of the feedback signal. The influence of the order of the autoregressive model is studied, and it is shown that better results are obtained by increasing the order. The choices for the location of the activator and for the type of sensor are investigated and discussed. Further analytical research on the method of computing the feedback signal in the presence of time-delay is identified. / Master of Science