There are many combinatorial optimization problems such as traveling salesman problem, quadratic-assignment problem, flow shop scheduling, that are computationally intractable. Tabu search based simulated annealing is a stochastic search algorithm that is widely used to solve combinatorial optimization problems. Due to excessive run time, there is a strong demand for a parallel version that can be applied to any problem with minimal modifications. Existing advanced and/or parallel versions of tabu search algorithms are specific to the problem at hand. This leads to a drawback of optimization only for that particular problem. In this work, we propose a parallel version of tabu search based SA on the Graphics Processing Unit (GPU) platform. We propose two variants of the algorithm based on where the tabu list is stored (global vs. local). In the first version, the list is stored in the global shared memory such that all threads can access this list. Multiple random walks in solution space are carried out. Each walk avoids the moves made in rest of the walks due to their access to global tabu list at the expense of more time. In the second version, the list is stored at the block level and is shared by only the block threads. Groups of random walks are performed in parallel and a walk in a group avoids the moves made by the rest of the walks within that group due to their access to shared local tabu list. This version is better than the first version in terms of execution time. On the other hand, the first version finds the global optima more often. We present experimental results for six difficult optimization functions with known global optima. Compared to the CPU implementation with similar workload, the proposed GPU versions are faster by approximately three orders of magnitude and often find the global optima.

Includes bibliographical references. / 1 v. (various foliations) : ill. (some col) ; 30 cm. / Title page, contents and abstract only. The complete thesis in print form is available from the University Library. / Thesis (Ph.D.)--University of Adelaide, Dept. of Community Medicine, 1994

There are over one thousand closed ‘small’ landfills in Ontario, each with differing circumstances and potential problems. This project proposes a method of addressing such dormant sites in situ, based upon a case study in Hamilton. Of the four closed landfills within Hamilton city limits, three of them lie in the low lands of the Red Hill Creek Valley. Perched at the source of the Red Hill Creek, the Upper Ottawa Street Landfill introduces unspoken toxins into the ecosystem of the entire valley. As the storm water catchment for the escarpment watershed, the creek serves a critical role in the recreational green belt which divides Hamilton and Stoney Creek. The source of this creek must be celebrated, not fenced off from public access due to landfill hazards. This proposal investigates beyond material recovery, into the possibilities of resource, ecosystem, and community recovery. Landfill mining, material sorting, and power generation through incineration are employed to reduce landfill volume. As the landfill is consumed, a new landscape is constructed, providing improved flood-prevention at the creek and a sanitary lined landfill for those materials remaining on site. Creek, forest, and field habitats are restored on site without the threat of contamination from landfill contents. The public can safely view the landfill mining operations from an elevated walkway, having new experiences with every visit. As the boundaries of the closed landfill are stripped away, the source of the Red Hill Creek and the new recreational parkland are made publicly accessible. Using this design as a reference, the equipment and operations designed for this site can be developed into a province-wide proposal.

Acute demyelination in children may be a monophasic illness or the sentinel attack of multiple sclerosis (MS) – a chronic inflammatory neurodegenerative demyelinating disease. MS risk is largely determined during childhood and vitamin D may protect against MS. The primary objective of this thesis was to evaluate vitamin D status in children presenting with acute demyelinating syndromes (ADS) as a potential contributor to MS outcome. The LIAISON “25 OH Vitamin D TOTAL” assay was validated to assess the biomarker of vitamin D status – serum 25-hydroxyvitamin D (25(OH)D) concentrations. Consecutive patients (<16 y) were enrolled at presentation with ADS and prospectively evaluated at 23 Canadian centres. MS was defined by a second clinical demyelinating event or by MRI evidence of new lesions over time. Cox proportional hazards regression models assessed risk of MS outcome as a function of serum 25(OH)D tertiles, accounting for factors associated with either MS risk or vitamin D status – age, sex, season, and HLA-DRB1*15 status. Of 211 children with 25(OH)D measured in sera obtained a median of 9 days from onset (interquartile range, 5 – 17 d; maximum 36 days), 20% (n = 41) were diagnosed with MS after 3.7 mos. (3.1 – 7.3 mos.). Risk of MS was lower in children with 25(OH)D levels in the highest tertile (≥ 74 nmol/L) at ADS versus those in the lowest tertile (<50 nmol/L) (HR 0.41; 95% CI 0.18 to 0.97, adjusted model). Children with higher circulating 25(OH)D concentrations at ADS have a lower risk of MS. Further evidence for a role of vitamin D insufficiency during childhood and adolescence contributing to MS risk comes from three MS patients with suboptimally managed pseudo-vitamin D deficiency rickets. Finally, a sun exposure questionnaire was validated in the latter part of this thesis for use in future research into determinants of vitamin D status and their association with risk of MS.

The nervous systems of animals evolved to exert dynamic control of behavior in response to the needs of the animal and changing signals from the environment. To understand the mechanisms of dynamic control, we need a means of predicting how individual neural and body elements will interact to produce the performance of the entire system. We have developed a neuromechanical application named AnimatLab that addresses this problem through simulation. A computational model of a body and nervous system can be constructed from simple components and situated in a virtual world for testing. Simulations and live experiments were used to investigate questions about locust jumping. The neural circuitry and biomechanics of kicking in locusts have been extensively studied. It has been hypothesized that the same neural circuit and biomechanics governed both behaviors, but this hypothesis was not testable with current technology. We built a neuromechanical model to test this and to gain a better understanding of the role of the semi-lunar process (SLP) in jump dynamics. The SLP are bands of cuticle that store energy for use during jumping. The results of the model were compared to a variety of published data and were similar. The SLP significantly increased jump distance, power, total energy, and duration of the jump impulse. Locust can jump precisely to a target, but also exhibit tumbling. We proposed two mechanisms for controlling tumbling during the jump. The first was that locusts adjust the pitch of their body prior to the jump to move the center of mass closer to the thrust vector. The second was that contraction of the abdominal muscles during the jump produced torques that countered the torque due to thrust. There was a strong correlation relating increased pitch and takeoff angle. In simulations there was an optimal pitch-takeoff combination that minimized tumbling that was similar to the live data. The direction and magnitude of tumbling could be controlled by adjusting abdominal tension. Tumbling also influenced jump elevation. Neuromechanical simulation addressed problems that would be difficult to examine using traditional physiological approaches. It is a powerful tool for understanding the neural basis of behavior.

Acute demyelination in children may be a monophasic illness or the sentinel attack of multiple sclerosis (MS) – a chronic inflammatory neurodegenerative demyelinating disease. MS risk is largely determined during childhood and vitamin D may protect against MS. The primary objective of this thesis was to evaluate vitamin D status in children presenting with acute demyelinating syndromes (ADS) as a potential contributor to MS outcome. The LIAISON “25 OH Vitamin D TOTAL” assay was validated to assess the biomarker of vitamin D status – serum 25-hydroxyvitamin D (25(OH)D) concentrations. Consecutive patients (<16 y) were enrolled at presentation with ADS and prospectively evaluated at 23 Canadian centres. MS was defined by a second clinical demyelinating event or by MRI evidence of new lesions over time. Cox proportional hazards regression models assessed risk of MS outcome as a function of serum 25(OH)D tertiles, accounting for factors associated with either MS risk or vitamin D status – age, sex, season, and HLA-DRB1*15 status. Of 211 children with 25(OH)D measured in sera obtained a median of 9 days from onset (interquartile range, 5 – 17 d; maximum 36 days), 20% (n = 41) were diagnosed with MS after 3.7 mos. (3.1 – 7.3 mos.). Risk of MS was lower in children with 25(OH)D levels in the highest tertile (≥ 74 nmol/L) at ADS versus those in the lowest tertile (<50 nmol/L) (HR 0.41; 95% CI 0.18 to 0.97, adjusted model). Children with higher circulating 25(OH)D concentrations at ADS have a lower risk of MS. Further evidence for a role of vitamin D insufficiency during childhood and adolescence contributing to MS risk comes from three MS patients with suboptimally managed pseudo-vitamin D deficiency rickets. Finally, a sun exposure questionnaire was validated in the latter part of this thesis for use in future research into determinants of vitamin D status and their association with risk of MS.

Electricity prices in competitive markets are extremely volatile with salient features such as mean-reversion and jumps and spikes. Modeling electricity spot prices is essential for asset and project valuation as well as risk management. I introduce the mean-reversion feature into a classical variance gamma model to model the electricity price dynamics as a mean-reverting variance gamma (MRVG) process. Derivative pricing formulae are derived through transform analysis and model parameters are estimated by the generalized method of moments and the Markov Chain Monte Carlo method. A real option approach is proposed to value a tolling contract incorporating operational characteristics of the generation asset and contractual constraints. Two simulation-based methods are proposed to solve the valuation problem. The effects of different electricity price assumptions on the valuation of tolling contracts are examined. Based on the valuation model, I also propose a heuristic scheme for hedging tolling contracts and demonstrate the validity of the hedging scheme through numerical examples. Autoregressive Conditional Heteroscedasticity (ARCH) and Generalized ARCH (GARCH) models are widely used to model price volatility in financial markets. Considering a GARCH model with heavy-tailed innovations for electricity price, I characterize the limiting distribution of a Value-at-Risk (VaR) estimator of the conditional electricity price distribution, which corresponds to the extremal quantile of the conditional distribution of the GARCH price process. I propose two methods, the normal approximation method and the data tilting method, for constructing confidence intervals for the conditional VaR estimator and assess their accuracies by simulation studies. The proposed approach is applied to electricity spot price data taken from the Pennsylvania-New Jersey-Maryland market to obtain confidence intervals of the empirically estimated Value-at-Risk of electricity prices. Several directions that deserve further investigation are pointed out for future research.

Analytical and numerical models are developed to investigate the effect of orthotropy on the stress distribution in variable thickness solid and annular rotating disks. The plastic treatment is based on Hill&rsquo / s quadratic yield criterion, total deformation theory, and Swift&rsquo / s hardening law. The elastic-plastic stress distributions, residual stresses and radial displacement distributions are obtained after having analysed the cases of rotating solid disk, annular disk with rigid inclusion, annular disk subjected to either internal or external pressure. Thermal loading is also considered for the annular disk with rigid inclusion. Effects of different values of elastic and plastic orthotropy parameters are investigated. It is observed that the elastic orthotropy significantly affects the residual stresses in disks. The most remarkable effect of the plastic orthotropy is observed on the disk with rigid inclusion.

<p>The fluid flow in the mold during up-hill teeming is of great importance for the quality of the cast ingot and therefore the quality of the final steel products. At the early stage of the filling of an up-hill teeming mold, liquid steel enters, with high velocity, from the runner into the mold and the turbulence on the meniscus could lead to entrainment of mold flux. The entrained mold flux might subsequently end up as defects in the final product. It is therefore very important to get a mild and stable inlet flow in the entrance region of the mold. It has been acknowledged recently that swirling motion induced using a helix shaped swirl blade, in the submerged entry nozzle is remarkably effective to control the fluid flow pattern in both the slab and billet type continuous casting molds. This result in increased productivity and quality of the produced steel. Due to the result with continuous casting there is reason to investigate the swirling effect for up-hill teeming, a casting method with similar problem with turbulence.</p><p>With this thesis we will study the effect of swirling flow generated through a swirl blade inserted into the entry nozzle, as a new method of reducing the deformation of the rising surface and the unevenness of the flow during filling of the up-hill teeming mold. The swirling blade has two features: (1) to generate a swirling flow in the entrance nozzle and (2) to suppress the uneven flow, generated/developed after flowing through the elbow. The effect of the use of a helix shaped swirl blade was studied using both numerical calculations and physical modelling. Water modelling was used to assert the effect of the swirling blade on rectifying of tangential and axial velocities in the filling tube for the up-hill teeming and also to verify the results from the numerical calculations. The effect of swirl in combination with diverged nozzle was also investigated in a similar way, i. e. with water model trials and numerical calculations.</p>

This report examines the cultural and economic growth of the Texas Hill Country resulting from the construction of the Highland Lake chain. It compares the current political and social climate of the region with the historical past, and offers solutions to avert an inevitable collision between a rising population and the constraints of limited water resources. / text