Currently, data size that we have to deal with is growing bigger and bigger. This fact implies that the computing time and computing power for dealing with the data is demanded. A way to circumvent the difficulty is as follows: Divide the data into several small blocks and then process these small blocks by several computers. Therefore, we need a tool for the decomposition-coordinated procedure. Alternating direction method of multipliers (ADMM) is a powerful algorithm for the mentioned purpose and has widely used in distributed optimizations. With ADMM algorithm, a big global optimization problem can be decomposed into several small local optimization problems. ADMM algorithm has been used in several recent distributed systems such as cloud systems and distributed antenna systems. In this thesis, we aim to apply the ADMM in a distributed antenna system. For the uplink setting, we develop a distributed demodulation algorithm, where multiple base stations collaborate with each other for data detection. On the other hand, for the downlink setting, we develop a distributed beamforming design algorithm, where multiple base stations collaborate to form a beamforming for mitigating the inter-cell interference. Finally, simulations are conducted to verify the efficiency of our designs.

This thesis presents a study of reduced-order models based on proper orthogonal decomposition applied to non-isothermal transport phenomena in °uidized beds. A numerical °ow solver called Multiphase Flow with Interphase eXchanges (MFIX) was used to generate a database of solution snapshots for proper orthogonal decomposi- tion (POD). Using POD, time independent basis functions were extracted from the data and the governing equations of the numerical solver were projected onto the basis functions to generate reduced-order models. A reduced-order model was constructed that simulates multi-phase isothermal and non-isothermal °ow. In the reduced-order models (ROMs) the large number of partial di®erential equations were replaced by a much smaller number of ordinary di®erential equations. These reduced-order models were applied to two reference cases, a time extrapolation case and a time-dependent period boundary condition case. Three additional acceleration techniques were devel- oped to further improve computational e±ciency of the POD based ROM: 1) Database splitting, 2) Freezing the matrix of the linear system and 3) Time step adjustment. Detailed numerical analysis of both the full-order model, MFIX and the POD-based ROM, including estimating the number of operations and the CPU time per iteration, was performed as part of this study. The results of this investigation show that the reduced-order models are capable of producing qualitatively accurate results with less than 5% error with a two-order of magnitude reduction of computational costs.

To better study frequency related effects such as attenuation and tuning, we developed a frequency dependent seismic reflection analysis. Comprehensive tests on full waveform synthetics and observations from the Teal South ocean bottom seismic (OBS) data set confirmed that normal moveout (NMO) stretch could distort both frequency and amplitude information severely in shallow events and far offset traces. In synthetic tests, our algorithm recovered amplitude and frequency information ac-curately. This simple but robust target oriented NMO stretch correction scheme can be used on top of an existing seismic processing flow for further analyses. By combining the NMO stretch correction, spectral decomposition, and crossplots of am-plitude versus offset (AVO) attributes, we tested the frequency dependent workflow over Teal south and Ursa field data sets for improved reservoir characterization. As expected from NMO stretch characteristics, low frequencies have been less affected while mid and high frequency ranges were affected considerably. In seismic attribute analysis, the AVO crossplots from spectrally decomposed prestack data confirmed the improved accuracy and effectiveness of our workflow in mid and high frequency regions. To overcome poor spectral decomposition results due to low signal to noise ratio (S/N) in the Teal South application, we also implemented a substack scheme that stacks adjacent traces to increase S/N ratio while reducing the amount of data to process and increasing the accuracy of the spectral decomposition step. Synthetic tests verified the effectiveness of this additional step. An application to the Ursa, Gulf of Mexico, deep water data set showed significant improvement in high frequency data while correcting biased low frequency information.

This research project examines variations in inequality in individual earned incomes across U.S. metropolitan areas. The main analysis includes thirteen explanatory variables from three major perspectives - the political economy perspective, the demand-side perspective and the labor force supply-side perspective. In addition, I applied path models to explain causalities between some independent variables and used the decomposition of the Theil index to show the between-group effects. The results indicate that most demand-side and supply-side factors significantly contribute to variances in metropolitan income inequalities, while the impact of political economic factors are very limited. The paper is organized in the following manner: Chapter I is the introduction; Chapter II reviews literature focusing on the level of earning inequality and its predictors; Chapter III describes data and measures of variables; Chapter IV introduces statistical methods (including OLS regression model, path analysis, and decomposition of the Theil index); Chapter V presents the results of OLS regression model and its explanations; Chapter VI explains path analysis and decomposition analysis and their results; and finally, Chapter VII discusses the current research project and its implications for future studies.

This thesis presents advances in reduced-order modeling based on proper orthogonal decomposition (POD) for single and two-phase flows. Reduced-order models (ROMs) are generated for two-phase gas-solid flows. A multiphase numerical flow solver, MFIX, is used to generate a database of solution snapshots for proper orthogonal decomposition. Time-independent basis functions are extracted using POD from the data, and the governing equations of the MFIX are projected onto the basis functions to generate the multiphase POD-based ROMs. Reduced-order models are constructed to simulate multiphase two-dimensional non-isothermal flow and isothermal flow particle kinetics and three-dimensional isothermal flow. These reduced-order models are applied to three reference cases. The results of this investigation show that the two-dimensional reduced-order models are capable of producing qualitatively accurate results with less than 5 percent error with at least an order of magnitude reduction of computational costs. The three-dimensional ROM shows improvements in computational costs. This thesis also presents an algorithm based on mathematical morphology used to extract discontinuities present in quasi-steady and unsteady flows for POD basis augmentation. Both MFIX and a Reynolds Average Navier-Stokes (RANS) flow solver, UNS3D, are used to generate solution databases for feature extraction. The algorithm is applied to bubbling uidized beds, transonic airfoils, and turbomachinery seals. The results of this investigation show that all of the important features are extracted without loss in accuracy.

Uranium-Molybdenum (U-Mo) alloys are of interest to the nuclear engineering community for their potential use as reactor fuel. The addition of molybdenum serves to stabilize the gamma phase of uranium, as well as increasing the melting point of the fuel. Thermal properties of U-Mo alloys have not been fully characterized, especially within the area of partial phase decomposition of the gamma phase of the alloy. Additional data was acquired through this research to expand the characterization data set for U-Mo alloys. The U-Mo alloys used for this research were acquired from the Idaho National Laboratory and consisted of three alloys of nominal 7, 10, and 13 percent molybdenum by weight. The sample pins were formed by vacuum induction melt casting. Once the three sample pins were fabricated and sent to the Fuel Cycle and Materials Laboratory at Texas A&M University, the pins were homogenized and sectioned for heat treatment. Several heat treatments were performed on the samples to induce varying degrees of phase decomposition, and the samples were subsequently sectioned for phase verification and thermal analysis. An Electron Probe Microanalyzer with wavelength dispersive spectroscopy was used to observe the phases in the samples as well as to characterize each phase. The density of each sample was determined using Archimedes method. Finally, a light flash analyzer was used to determine thermal diffusivity of the samples up to 300 degrees C as well as to estimate the thermal conductivity. For U-10Mo, thermal diffusivity increased with increasing phase decomposition from gamma to alpha +U2Mo while U-7Mo saw a flattening of the thermal diffusivity curve with increased phase decomposition.

Gas turbine engines are the main propulsion source for world wide aviation and are also used for power generation. Even though they rely mainly on fossil fuel and emit climate active gasses, their importance is not likely to decrease in the future. But more efficient ways of using finite resources and hence reducing emissions have to be found. Thus, the interest to improve engine efficiency is growing. Considering the efficiency of the underlying thermodynamic cycle, an increase can be achieved by raising the turbine inlet temperature or compression ratio. Due to the complex nature of the underlying flow physics, however, the aero-thermal processes are still not fully understood. For this reason, one needs to perform research at high spatial and temporal resolution, in turn creating the need for effective means of postprocessing the large amounts of data. This dissertation addresses both sides of the problem - using high-scale, high resolution simulations as well as effective post processing techniques. As an example for the latter, a temporal highly resolved data set from wall pressure measurements of a transonic compressor stage is analyzed using proper orthogonal decomposition. The underlying experiments were performed by collaborators at Technical University Darmstadt. To decompose signals into optimal orthogonal basis functions based on temporal correlations including temperature, a formal mathematical framework is developed. A method to rank the reduced order representations with respect to heat transfer effectiveness is presented. To test both methods, a Reynolds-averaged Navier-Stokes (RANS) simulation and large eddy simulation (LES) are performed on turbulent heat transfer in a square duct with one single row of pin fins. While the LES results show closer agreement to experiments, both simulations unveil flow parts that do not contribute to heat transfer augmentation and can be considered wasteful. From the most effective mode, a wall contour for the same domain is derived and applied. In the wall contoured domain, energy in wasteful modes decreased, heat transfer increased and the temperature fluctuations at the wall decreased. Another stagnating boundary layer flow is examined in a direct numerical simulation of a first stage stator vane. Elevated levels of free stream turbulence and integral length scale are generated to simulate the features of combustor exit flow. The horseshoe vortex dynamics cause an increase in endwall heat transfer upstream of the vane. The link between energy optimal orthogonal basis functions and flow structures is examined using this data and the reduced order heat transfer analysis shows high energy modes with comparatively low impact on turbulent heat transfer. The analysis further shows that there are multiple horseshoe vortices that oscillate upstream of the blade, vanish, regenerate and can also merge. There is a punctual correlation of intense vortex dynamics and peaks in the orthogonal temperature basis function. For all data considered, the link between the energy optimal orthogonal basis functions and flow structures is neither guaranteed to exist nor straightforward to establish. The orthogonal expansion locks onto flow parts with high fluctuating kinetic energy - which might or might not be the ones that are looked for. The heat transfer ranking eliminates this problem and is valid independently of how certain basis functions are interpreted.

Implementing information system within an enterprise is supposed to integrate the business vision and goal sufficiently. Failing to do so shall waste a lot of information system resources. In this study, we utilize an Architectural Integrated Business Process (AIBP) approach that makes use of software architecture on both business process and information system. This approach refers business process when doing analysis and design of information system. AIBP approach is likely favored by corporate executives, thenceforth help gain the most profit for the information system within an enterprise. AIBP approach uses software architecture to describe business process, and information system as well. There are two major factors in software architecture. The first one is the structure and the second one is the behavior. Structure is used to describe the elements of a system, while behavior is used to describe the interaction among those elements. Both structure and behavior will be formed in the same hierarchy, also called the architecture hierarchy. Once an architectural integrated business process is specified, the following implementation of information system becomes an easy task. The architecture hierarchy will guide the decomposition of information system correctly. AIBP also offers great amount of knowledge on the development of software components.

第23回名古屋大学年代測定総合研究センターシンポジウム平成22(2010)年度報告

none