Thesis (M. P. A.)--Texas State University-San Marcos, 2009. / "Fall 2009." Includes bibliographical references (leaves 104-109).

Thesis (M.A.)--Ohio University, November, 2009. / Release of full electronic text on OhioLINK has been delayed until September 1, 2014. Title from PDF t.p. Includes bibliographical references.

The unprecedented demand for performance in the latest technologies will ultimately require changes in the way we design cache. Emerging high density embedded memories such as Spin-Transfer Torque Random Access Memory (STTRAM) have emerged as a possible candidate for universal memory due to its high speed, low power, non-volatility, and low cost. Although attractive, STTRAM is susceptible to contactless tampering through malicious exposure to magnetic field with the intention to steal or modify the bitcell content. In this thesis, we explore various attack techniques on STTRAM and then propose a novel array-based sensor to detect the polarity and magnitude of such attacks and then propose two design techniques to mitigate the attack. With our research, we have been able to successfully implement and accurately detect an attack while providing sufficient compensation window (few ns to ~100 us) to enable proactive protection measures. Finally, we show that variable-strength ECC can adapt correction capability to tolerate failures with various strength of an attack.

Countries around the world face social, economic, and ecological damage from escalating natural disasters caused by climate change. In an effort to curtail climate change impacts, local and regional governments are beginning to employ green house gas (GHG) mitigation strategies to reduce their carbon footprint. These strategies work to eliminate a range of GHG emissions from entering the atmosphere. Apart from carbon dioxide (CO₂), the most prevalent GHG is methane. In terms of global warming, methane is approximately 21 times more harmful to the atmosphere than CO₂. Natural gas systems, coal mining, manure management, rice cultivation, wastewater treatment, and landfills all contribute to methane generation. According to the US Environmental Protection Agency's 2011 US GHG inventory, landfills generate 1.5% of total GHG emissions in carbon dioxide equivalents. Recognizing the global impacts of its policies and operations, municipalities are working to reduce their GHG emissions. Coalitions like the C40 Cities Climate Leadership Group were created to specifically address GHG reductions, which will result in a 248 million MT reduction in GHGs released to the atmosphere by 2020. Guided by existing literature, this Master's Report calculates methane generation and transport to determine the effectiveness of applying two methane mitigation alternatives--passive methane oxidation biocovers (PMOBs) and landfill gas to energy technologies (LFGTE)--at an inactive landfill site to reduce GHG emissions. LFGTE generates energy for direct use such as space heating or industrial processes or for electricity generation. Cost-saving strategies abound for landfills which utilize LFGTE. PMOBs optimize the landfill surface soil cover environment to promote microbial growth of bacteria, called methanotrophs, which convert methane into carbon dioxide. When employed, these mitigation alternatives are designed to significantly reduce methane emissions from landfills. The EPA has developed a computer modeling program (LANDGEM) to aid in the calculation of landfill gas generation. A hypothetical case study of a one million ton landfill was created and modeled for methane generation over a 35 year period. With methane generation rates calculated, assessment of potential LFGTE was performed and methane oxidation rate calculations were made to determine the impact of a PMOB and LFGTE on net GHG emissions at the landfill. The overall GHG reductions with these engineering controls were two-thirds of the level a landfill without controls would emit. These results indicate that implementing methane mitigation steps at closed landfills throughout the world would yield significant reductions in GHG emissions. / text

No description available.

Market power mitigation has been an integral part of wholesale electricity markets since deregulation. In wholesale electricity markets, different regions in the US take different approaches to regulating market power. While the exercise of market power has received considerable attention in the literature, the issue of market power mitigation has attracted scant attention. In the first chapter, I examine the market power mitigation rules used in New York ISO (Independent System Operator) and California ISO (CAISO) with respect to day-ahead and real-time energy markets. I test whether markups associated with New York in-city generators would be lower with an alternative approach to mitigation, the CAISO approach. Results indicate the difference in markups between these two mitigation rules is driven by the shape of residual demand curves for suppliers. Analysis of residual demand curves faced by New York in-city suppliers show similar markups under both mitigation rules when no one supplier is necessary to meet the demand (i.e., when no supplier is pivotal). However, when some supplier is crucial for the market to clear, the mitigation rule adopted by the NYISO consistently leads to higher markups than would the CAISO rule. This result suggest that market power episodes in New York is confined to periods where some supplier is pivotal. As a result, I find that applying the CAISOs' mitigation rules to the New York market could lower wholesale electricity prices by 18%. The second chapter of my dissertation focuses on supply function equilibrium. In power markets, suppliers submit offer curves in auctions, indicating their willingness to supply at different price levels. Although firms are allowed to submit different offer curves for different time periods, surprisingly many firms stick to a single offer curve for the entire day. This essentially means that firms are submitting a single offer curve for multiple demand realizations. A suitable framework to analyze such oligopolistic competition between power market suppliers is supply function equilibrium models. Using detailed bidding data, I develop equilibrium in supply functions by restricting supplier offers to a class of supply functions. By collating equilibrium supply functions corresponding to different realizations of demand, I obtain a single optimal supply function for the entire day. Then I compare the resulting supply function with actual day-ahead offers in New York. In addition to supply function equilibrium, I also develop a conservative bidding approach in which each firm assumes that rivals bid at marginal costs. Results show that the supply functions derived from equilibrium bidding model in this paper is not consistent with actual bidding in New York. This result is mainly driven by the class of supply functions used in this study to generate the equilibrium. Further, actual offers do not resemble offers generated by the conservative bidding algorithm.

Reptile populations are suffering substantial global losses and roads are identified as one of the leading threats to their persistence. Currently, efforts to mitigate this threat are being implemented with various levels of success. I studied the effectiveness of exclusion structures (i.e., fencing) at preventing reptiles from gaining access to the road, and reducing road mortality. I also examined if population connectivity structures (i.e., ecopassages) were effective at reducing habitat and population fragmentation and allowing individuals to access habitats, resources, and mates on both sides of a major road (4 lane highway). I found that the fence was ineffective at preventing reptiles from gaining access to the road; however, reptiles were observed using the ecopassages to cross the road. Behavioural trials testing painted turtles’ (Chrysemys picta) willingness to use an ecopassage demonstrated that refusal was twice more likely than use of an ecopassage. I also examined the potential for roads to pose a physiological threat to roadside populations of reptiles by examining corticosterone (CORT), a stress hormone linked to negative health effects in cases of elevated levels over the long-term. To assess if individuals living near a major road had higher CORT levels than individuals from a less impacted population, I developed a novel means of measuring CORT from painted turtle claws in partnership with Dr. Gabriela Mastromonaco (Toronto Zoo). With long-term CORT levels considered as a proxy for chronic physiological stress, I did not find evidence that populations near roads had altered stress levels. However, this seminal study will provide the framework for further examination of more species, including species-at-risk, and a better understanding of effects of anthropogenic environments on wildlife health. As road ecologists strive to expand our understanding of the threats roads pose to reptiles, it is important that this field spans multiple disciplines, so that we can both understand the direct and indirect threats that roads cause and develop effective mitigation that preserves biodiversity within our anthropogenic landscape.

This thesis proposes a classification system of risk mitigation strategies based on literature search and industry interviews. Following that, a list of generic properties was generated to describe individual strategies. In parallel, populating the properties of a large number of identified strategies was attempted. The practical implications are discussed mainly focusing on knowledge management for risk mitigation strategies.

The safety and security of liquefied natural gas (LNG) facilities has prompted the need for continued study of LNG mitigation systems. Water spray systems are widely recognized as an effective measure for dispersing LNG vapor clouds. Currently, there are no engineering guidelines available for water curtain applications in the LNG industry due to a lack of understanding of the complex interactions between the LNG vapor cloud and water droplets. This research applies computational fluid dynamics (CFD) modeling to investigate the forced dispersion of LNG vapor using upward-oriented full-cone spray nozzles. A Eulerian-Lagrangian approach was applied to simulate the energy and momentum exchange between the continuous (gas flow) and discrete (droplets) phases. Discussed are the physical parameters that are essential inputs to the CFD simulation of the water spray-LNG system. The experimental data collected from the Mary Kay O’Connor Process Safety Center’s outdoor LNG spill work in March 2009 at the Brayton Fire Training Field were used to calibrate the physical parameters. The physical mechanisms of the water spray application were investigated using LNG forced dispersion modeling. The effects of momentum imparting from the droplets to the air- vapor mixture, thermal transfer between the two phases (droplet/vapor) and effects of various levels of air entrainment rates on the behavior of the LNG vapors are evaluated. Lastly, the key parametric dependences of the design elements for an effective water curtain system are investigated. The effects of different droplet sizes, droplet temperatures, nozzle cone angles, and installation configurations of water spray applications on LNG vapor behavior are analyzed. This work aims to investigate the complex interaction of the water droplet-LNG vapor system, which will serve in developing guidelines and establishing engineering criteria for a site-specific LNG mitigation system. Finally, the potentials of applying CFD modeling in providing guidance for setting up the design criteria for an effective forced mitigation system as an integrated safety element for LNG facilities are discussed.

No description available.