Query-Driven Analysis and Visualization for Large-Scale Scientific Dataset using Geometry Summarization and Bitmap Indexing

Wei, Tzu-Hsuan The Ohio State University ProQuest Dissertations & 2017 해외박사(DDOD)

소속기관이 구독 중이 아닌 경우 오후 4시부터 익일 오전 9시까지 원문보기가 가능합니다.

The computational power of modern supercomputers grows rapidly, and it facilitates scientists to produce high-resolution datasets when simulating physical or weather models, which generate extreme scale data with multiple variables most of the t. First, we focus on the problem of identifying salient features and evaluating selected features for creating a data summarization. To analyze a volumetric dataset, displaying isosurface is typically used to reveal the locations of values that th.

Essays on Technology Adoption, Demographics, and Development

Lai, Ting-Wei Washington University in St. Louis 2015 해외박사(DDOD)

소속기관이 구독 중이 아닌 경우 오후 4시부터 익일 오전 9시까지 원문보기가 가능합니다.

This dissertation is to connect empirical findings with grounded theoretical analysis on two economic issues. One of the studies investigates industrial productivity by fitting in a theoretical model with quantitative methods. In addition, I explore how a demographic policy in China brings forth a profound impact in all aspects of the fast-growing economy. The first chapter, ''Casual Labor, Uncertainty, and Technology Adoption in Agriculture," examines why both the technology adoption rate and labor productivity in agriculture are low in the context of developing countries. A two-stage model is built to explain how the availability of casual (non-permanent) labor ex-post, in the presence of uncertainty may affect agents' ex-ante technology choices. A higher degree of uncertainty induces the agents to choose traditional production technology that relies heavily on the labor input instead of using any modern intermediate inputs. By calibrating the model to fit the micro data in Tanzania, I show that this proposed framework can be used to account for two targets of interest: low aggregate labor productivity and the low technology adoption rate. Counterfactual exercises suggest that the severity of uncertainty before the harvest stage and the abundance of casual labor are the potential drivers for the two targets to be explained. The second chapter, ''Growth in a Patrilocal Economy: Female Schooling, Household Savings, and China's One-Child Policy," is co-authored Wei-Cheng Chen. We develop a model of parental education decision to analyze how a population control policy affects saving and schooling in a patrilocal society, where sons are responsible to support aged parents more than daughters. Parent's investment in education depends on the degree of parental altruism and the need for old-age security. A tightened population control policy makes parental altruism more important relative to the security motive and shortens gender gap in education. We also take another crucial intergenerational incentive for daughter's education into account, since lower fertility promotes female labor market participation and increases the value of female education. Our model explains why the Chinese economy under the ''One-Child Policy'' exhibits a rapid growth of relative female schooling. Moreover, this chapter also articulates the relationship between household savings and demographic changes based on a general equilibrium analysis, which has been discussed extensively in recent years to explain the China's saving puzzle.

The Functional Role of Receptor-Interacting Protein 140 (RIP140) in Innate Immunity and Metabolic Syndrome

Lin, Yi-Wei ProQuest Dissertations & Theses University of Minn 2016 해외박사(DDOD)

소속기관이 구독 중이 아닌 경우 오후 4시부터 익일 오전 9시까지 원문보기가 가능합니다.

Metabolic diseases, such as type II diabetes (T2DM), atherosclerosis and other cardiovascular diseases, are prevalent and are important health issues in the modern world. T2DM contributes to the development of various metabolic diseases. Atherosclerosis is one of the major causes leading to multiple cardiovascular diseases. In order to develop therapeutic strategies, understanding the mechanisms of these metabolic diseases is crucial. It is known that the immune system is highly involved in initiation and progression of metabolic diseases. Macrophages are one of the major leukocytes in innate immunity. Macrophages have two major polarized phenotypes: classical/pro-inflammatory (M1) and alternative/anti-inflammatory (M2). It is widely accepted that M1-M2 switch in macrophage population is essential in disease progression or damage recovery; however, the detailed mechanism of macrophage phenotype switch has not been fully elucidated. In addition, the effect of altering the macrophage phenotype on treating metabolic diseases remains uncertain. Receptor-interacting protein 140 (RIP140) is a co-regulator of numerous nuclear receptors and transcription factors. RIP140 is expressed in various cell types including adipocytes, liver, muscle, heart, neurons, and cells in the monocyte--macrophage lineage. Studies showed that RIP140 expression is positively associated with the progression of metabolic disorders such as obesity, insulin resistance, and glucose intolerance. In addition, studies indicate that RIP140 acts as a co-activator of NFkappaB to promote macrophage M1 activation and pro-inflammatory responses. My studies further build on this knowledge to uncover the role of RIP140 in the metabolic diseases. First, it was found that RIP140 elevates cholesterol content in macrophages by reducing expression of ABC transporters, which are responsible for cholesterol efflux. The elevated cytosolic cholesterol induces foam cell formation and further enhances progression of atherosclerosis. This study indicated that reducing RIP140 levels effectively ameliorates high-cholesterol diet-induced atherosclerosis. Second, my study found that reducing RIP140 in macrophages leads to macrophage M2 polarization, resulting in adipose tissue remodeling to brown/beige adipose tissue. This further ameliorates high fat diet-induced T2DM associated metabolic disorders. Moreover, later studies address how RIP140 mediates macrophage M2 activation and M1/M2 switch by its cytosolic function in a wound healing animal model. Final study is to identify a beneficial taxonomic repertoire from macrophage specific RIP140 knockdown (M&phis;RIP140KD) mice. Fecal microbiota transplantation (FMT) from HFD-fed M&phis;RIP140KD to wild type (WT) mice acquired the benefits from donors, which is resistant to development of HFD-induced metabolic diseases. Taken together, this thesis studies elucidate novel functions of RIP140 in polarization and inflammatory responses in macrophages, and identify the benefits of reducing RIP140 expression in macrophages. These findings contribute to our understanding of the relationship between immune and metabolic systems as well as provide a therapeutic target of resolving inflammation and preventing/improving metabolic profiles in T2DM, and atherosclerosis.

Enhancing the prediction accuracy of regression trees: Linear splits and variable selection

Chen, Chien-Wei The University of Wisconsin - Madison 2008 해외박사(DDOD)

소속기관이 구독 중이 아닌 경우 오후 4시부터 익일 오전 9시까지 원문보기가 가능합니다.

We extend univariate splits to linear splits for regression trees and to two response variables. The linear split algorithm, LIST, improves the prediction accuracy in simulation and empirical results although it takes more computing time than other algorithms. The algorithm proposed for handling two response variables, RT2D, performs well when the dependent variables are highly correlated. We also explore the performance of bagging algorithms with all variables and with only one-third of them when searching for the split variables. A new algorithm which combines bagging with stepwise regression and regression tree, VAST, is proposed in order to select the important variables for regression problems. This algorithm solves the situation of having of more variables than observations. In simulation results, its performance is better than Random Forest and GUIDE on average. Three different methods proposed to handle missing values are also implemented in this algorithm. None of these methods is dominant in all situations. Another variable selection algorithm based on bagging QUEST also shows efficiency in selecting the important variables for classification problems.

Nonlinear optical microscopy for the invisible: Vibrational imaging of small molecules in live cells and electronic imaging of fluorophores into the ultra deep

Wei, Lu Columbia University 2015 해외박사(DDOD)

소속기관이 구독 중이 아닌 경우 오후 4시부터 익일 오전 9시까지 원문보기가 가능합니다.

Nonlinear optical microscopy (NOM) has become increasingly popular in biomedical research in recent years with the developments of laser sources, contrast mechanisms, novel probes and etc. One of the advantages of NOM over the linear counterpart is the ability to image deep into scattering tissues or even on the whole animals. This is due to the adoption of near-infrared excitation that is of less scattering than visible excitation, and the intrinsic optical sectioning capability minimizing the excitation background beyond focal volume. Such an advantage is particularly prominent in two-photon fluorescence microscopy compared to one-photon fluorescence microscopy. In addition, NOM may provide extra molecular information (e.g. second harmonic generation and third harmonic generation) or stronger signal (e.g. stimulated Raman scattering and coherent anti-Stokes Raman scattering compared to spontaneous Raman scattering), because of the nonlinear interaction between strong optical fields and molecules. However, the merits of NOM are not yet fully exploited to tackle important questions in biomedical research. This thesis contributes to the developments of NOM in two aspects that correspond to two fundamental problems in biomedical imaging: first, how to non invasively image small functional biomolecules in live biological systems (Chapters 1-4); second, how to extend the optical imaging depth inside scattering tissues (Chapters 5-6). The ability to non-perturbatively image vital small biomolecules is crucial for understanding the complex functions of biological systems. However, it has proven to be highly challenging with the prevailing method of fluorescence microscopy. Because it requires the utilization of large-size fluorophore tagging (e.g., the Green Fluorescent Protein tagging) that would severely perturb the natural functions of small bio-molecules. Hence, we devise and construct a nonlinear Raman imaging platform, with the coupling of the emerging stimulated Raman scattering (SRS) microscopy and tiny vibrational tags, which provides superb sensitivity, specificity and biocompatibility for imaging small biomolecules (Chapters 1-4). Chapter 1 outlines the theoretical background for Raman scattering. Chapter 2 describes the instrumentation for SRS microscopy, followed with an overview of recent technical developments. Chapter 3 depicts the coupling of SRS microscopy with small alkyne tags (C≡C) to sensitively and specifically image a broad spectrum of small and functionally vital biomolecules (i.e. nucleic acids, amino acids, choline, fatty acids and small molecule drugs) in live cells, tissues and animals. Chapter 4 reports the combination of SRS microscopy with small carbon-deuterium (C-D) bonds to probe the complex and dynamic protein metabolism, including protein synthesis, degradation and trafficking, with subcellular resolution through metabolic labeling. It is to my belief that the coupling of SRS microscopy with alkyne or C-D tags will be readily applied in answering key biological questions in the near future. The remaining chapters of this thesis (Chapters 5-6) present the super-nonlinear fluorescence microscopy (SNFM) techniques for extending the optical imaging depth into scattering tissues. Unlike SRS microscopy that is an emerging technique, multiphoton microscopy (mainly referred as two-photon fluorescence microscopy), has matured over 20 years with its setup scheme and biological applications. Although it offers the deepest penetration in the optical microscopy, it still poses a fundamental depth limit set by the signal-to-background ratio when imaging into scattering tissues. Three SNFM techniques are proposed to extend such a depth limit: unlike the conventional multiphoton microscopy whose nonlinearity stems from virtual-states mediated simultaneous interactions between the incident photons and the molecules, the high-order nonlinearity of the SNFM techniques that we have conceived is generated through real-state mediated population-transfer kinetics. In particular, Chapter 5 demonstrates the multiphoton activation and imaging (MPAI) microscopy, which adopts a new class of fluorophores, the photoactivatable fluorophores, to significantly extend the fundamental imaging depth limit. Chapter 6 theoretically and analytically depicts two additional SNFM techniques of stimulated emission reduced fluorescence (SERF) microscopy and focal saturation microscopy. Both MPAI and focal saturation microscopies exhibit a fourth order power dependence, which is effectively a four-photon process. SERF presents a third order power dependence for a three-photon process.

Numerische und statistische Verfahren zur kosmologischen Simulation der groβra@umigen Galaxienverteilung unter Verwendung der Lagrangeschen Sto@rungsrechnung

Weiβ, Arno Gerrit 1995 국내박사

Study on the Deduction and Changes of the Decorative Art of Dunhuang Murals

Chen Wei 경기대학교 대학원 2021 국내박사

Decoration accompanies the formation of cultures, with different regional and cultural characteristics. These characteristics repeatedly merge and form a melodic rhythm. Dunhuang Murals are an organism of multiple aesthetic factors in the development of derivative meaning, which is demonstrative, they are repeatedly reconstruct in a complex fusion of decorative elements. The decorative forms in Dunhuang murals evolved through this recurring blend of representation. The first step was to summarize the data collected and organize two stages of research into the decorative modeling of Dunhuang murals: The first stage is from Northern Wei to Sui Dynasty, and the second stage is from Sui Dynasty to Tang Dynasty. A preliminary classification of the categories of decorative elements characteristic of these two phases into periods and patterns of change in their stylistic characteristics. In addition, we will compare the development curves of the foreign culture and the local culture and summarize the characteristics and progressive links between the two periods. In the first phase of the Northern Wei and Sui Dynasties, decorative elements of foreign and local cultures were juxtaposed, and the boundaries between the two began to blur within the picture, shifting to a composite; In the second phase of the Sui-Tang period, decorative elements were superimposed on metaphors of similarity, and decorative forms began to be reconstructed. Secondly, on the basis of the induction of the first stage, the second stage is compared with the first stage from the point of view of the continuity of decoration, and the characteristics of the decorative stylistic language developed in the second stage are explained deconstructively. For example, the blurring of boundaries between decorative forms to Deduction and Changes, structural refinement to reconstruction, Deduction and Changes forms to symbolic, and Sophistication and simplicity coexist. The above study is divided into five parts in this paper. The first part describes the purpose and methodology of the study and explains the proper nouns that appear in the text. The second part describes the internal and external cultural factors that influenced Dunhuang murals, based on the region of Dunhuang. We will discuss the historical background and the period of artistic development of the decorative arts in the Dunhuang region, mainly in the Gandhara and Western regions, among the foreign cultures. In the local culture of the Dunhuang region, the relationship between the socio-historical background and the artistic development of Dunhuang mural, mainly in the two stages of development studied in this paper, is explained. The third and fourth parts are progressive in form. It is a study of the decorative and stylistic characteristics of the two stages of Dunhuang mural. The third part, based on the specific cases identified in the data collection, will show that the major foreign cultures that influenced Dunhuang murals, such as the Gandhara culture and the Quici culture, are an extension of the roots of foreign cultures. To juxtapose and study the metaphors and characteristics of decorative elements in the above cultures and the metaphors and characteristics of decorative forms in local cultures of the Han period. We will compare and analyze the metaphorical and decorative styling of the caves with specific examples of Dunhuang caves from the five periods of the first phase of the Northern Liang, Northern Wei, Western Wei, and Northern Zhou dynasties. The decorative features of the decorative elements at the 'nano-different' stage are contrasted with the structural organization of the elements. According to the decorative features summarized in the third part, the fourth part analyzes and explains the features formed by the continuity of the repeatedly reconstruct of decorative shapes under the influence of the complex overall style of the Tang dynasty and with the perspective of 'integration'. The re-presentation of decorative forms, using existing art theories for formal aesthetic analysis, and concluding that the continuity of Dunhuang murals in decorative forms is characterized by demonstrative, and the development of self-discipline and heteronomy. A summary of the decorative modeling characteristics presented in the two periods of Dunhuang mural painting according to the fifth part. Key words : Dunhuang Murals, Decorative Styling, Deduction and Changes, Repeatedly reconstruct, Demonstrative. 장식은 문화의 형성에 따라 서로 다른 지역적·문화적 특징을 가지게 되었다. 이러한 특징은 반복적으로 융합돼 멜로디와 같은 리듬을 형성해 냈다. 특히 敦煌壁畵는 넓은 의미에 있어 여러 가지 심미적인 요소로 구성된 유기체이며 표현성도 가지고 있다. 또한 이것은 장식 요소 간에 서로융합하여 복합적으로 재구성 된다. 돈황벽화의 장식 조형은 표현성의 반복적인 융합 가운데 발전해 나가고 있다. 먼저 돈황벽화 장식 조형의 시기를 설정하고 그 특징을 논술하였다. 첫째, 시기는 北魏에서 隨代, 隋에서 唐代까지 설정하였다. 이 시기 구분에서 특징을 가진 장식 요소를 시간과 조형적 특징 그리고 변천의 규칙에 따라 일차적으로 분류한다. 외래 문화와 현지 토착 문화 이 두 가지 맥락의 발전 양상을 비교해 보았으며, 시기별 비교분석을 통하여 특징과 연관성을 요약하였다. 즉, 北魏에서 隨代까지는 외래 문화 및 현지 토착문화 문화 이 두 가지 맥락의 발전 양상을 비교해 보았으며, 시기별 비교분석을 통하여 특징과 연관성을 요약하였다.즉, 北魏에서 隨代까지는 외래 문화 및 현지 토착 문화와 유사한 은유를 지닌 장식 요소가 나란히 존재한다. 외래 문화와 현지 벽화에 있는 장식 조형이 매우 모호해져 복합으로 전환되기 시작했다. 隨에서 唐代는 유사한 성질을 지닌 장식 요소들이 중첩과 같은 변화의 과정을 통하여 장식 조형이 재구성되기 시작했다. 둘째, 시기별 특징과 연관성을 바탕으로 장식의 연속성 관점에 입각하여 비교하였다. 아울러 두 시기에서 형성된 장식 조형 언어의 특징 구조를 해체하여 설명하였다. 예를 들어, 장식 조형 사이의 경계는 모호함에서 衍義로 발전 되고, 구조는 추출에서 재구축으로, 衍義의 형태는 象徵으로 변형되며, 巧拙함이 서로 어우러져 轉變되어지는 것 들이다. 셋째, 진화의 관점에서 돈황벽화의 장식 요소의 조형적 특징 변화에 나타난 연속성 규율을 찾았다. 본 논문은 내부 및 외부 요인의 비교 분석 방법을 채택하였다. 외래문화와 현지 토착문화의 융합으로 형성된 은유와 그 조형적 특징을 분석함으로써 돈황벽화의 衍義 발전은 선형적이며 복합적인 특징이 있음을 밝혔다. 이 논문은 총 5장으로 나뉜다. 제1장은 연구의 목적과 방법을 논술하고 논문에 나오는 주제어를 해석하였다. 제2장은 돈황의 지역적 특성을 바탕으로 돈황벽화에 영향을 미치는 내·외부적 문화 요소를 언급하였다. 돈황 지역의 외래 문화에서 역사적 배경과 예술 발전은 주로 간다라의 특성과 서부 지역에 나타난 장식 예술을 기반으로 서술하였으며, 현지 토착문화는 두 시기에서 나타난 사회적, 역사적 배경과 예술적 발전의 관계를 기반으로 하였다. 제3장은 돈황벽화의 두 시기에서 나타난 장식 조형 특징에 관한 연구이다. 자료 수집에서 찾은 구체적인 사례를 근거로 하였다. 외래문화를 추적함에 있어 간다라 문화와 龜茲문화가 돈황벽화에 어떠한 영향을 미치는지 살펴보았다. 외래문화에 있는 장식 요소의 은유와 그 조형 특징을 漢代와 비교 논술하였다. 또한 돈황동굴 첫 번째 단계에 속한 5 時期인 北凉, 北魏, 西魏, 北周 중 돈황동굴의 구체적인 사례를 은유와 장식 조형으로 비교 분석하고. 그리고 納異’의 단계에서 나타나는 장식 요소의 특징과 구조 형식을 비교하였다. 제4장은 제3장에서 요약한 장식 특징을 바탕으로 唐代에 번영한 풍격의 영향을 받아 융합의 관점에서 장식 조형이 연속적이고 복합적으로 재구축 되어지는 특성을 해석하였다. 장식 조형의 재현은 기존의 예술 이론으로 심미를 분석 하였다. 장식 조형에서 돈황벽화의 연속성은 표현성의 특징과 자율과 타율의 발전적 특징을 가지고 있음을 확인하였다. 제5장은 두 시기에 걸친 돈황벽화의 장식 조형 특징을 정리하였다.

Submodular Optimization and Data Processing

Wei, Kai University of Washington ProQuest Dissertations & 2016 해외박사(DDOD)

소속기관이 구독 중이 아닌 경우 오후 4시부터 익일 오전 9시까지 원문보기가 가능합니다.

Data sets are large and and are getting larger. Two common paradigms -- data summarization and data partitioning, are often used to handle the big data. Data summarization aims at identifying a small sized subset of the data that attains the maximum utility or information, while the goal of data partitioning is to split the data across multiple compute nodes so that the data block residing on each node becomes manageable. In this dissertation, we investigate how to apply submodularity to these two data processing paradigms. In the first part of this thesis, we study the connection of submodularity to the data summarization paradigm. First we show that data summarization subsumes a number of applications, including acoustic data subset selection for training speech recognizers [Wei et al., 2014], genomics assay panel selection [Wei et al., 2016], batch active learning [Wei et al., 2015], image summarization [Tschiatschek et al., 2014], document summarization [Lin and Bilmes, 2012], feature subset selection [Liu et al., 2013], and etc. Among these tasks, we perform case studies on the former three applications. We show how to apply the appropriate submodular set functions to model the utility for these tasks, and formulate the correspond- ing data summarization task as a constrained submodular maximization, which admits an efficient greedy heuristic for optimization [Nemhauser et al., 1978]. To better model the util- ity function for an underlying data summarization task, we also propose a novel "interactive" setting for learning mixtures of submodular functions. For such interactive learning setting, we propose an algorithmic framework and show that it is effective for both the acoustic data selection and the image summarization tasks. While the simple greedy heuristic al- ready efficiently and near-optimally solves the constrained submodular maximization, data summarization tasks may still be computationally challenging for large-scale scenarios. To this end, we introduce a novel multistage algorithmic framework called MultiGreed, to significantly scale the greedy algorithm to even larger problem instances. We theoretically show that MultGreed performs very closely to the greedy algorithm and also empirically demonstrate the significant speedup of MultGreed over the standard greedy algorithm on a number of real-world data summarization tasks. In the second part of this thesis, we connect submodularity to data partitioning. We first propose two novel submodular data partitioning problems that we collectively call Submodular Partitioning. To solve the submodular partitioning, we propose several novel algorithmic frameworks (including greedy, majorization-minimization, minorization-maximization, and relaxation algorithms) that not only scale to large datasets but that also achieve theoretical approximation guarantees comparable to the state-of-the-art. We show that submodular partitioning subsumes a number of machine learning applications, including load balancing for parallel systems, intelligent data partitioning for parallel training of statistical models, and unsupervised image segmentation. We perform a case study on the last application. For this case, we demonstrate the appropriate choice of submodular utility model and the corresponding submodular partitioning formulation. Empirical evidences suggest that the submodular partitioning framework is effective for the intelligent data partitioning task.

Solid State Dye-sensitized Solar Cells with Improved Conductivity and Recombination Based on Polymer Electrolyte : 고분자 전해질의 향상된 전도도와 재결합 성능을 응용한 고체형 연료감응 태양전지

Wei Wei 한양대학교 2016 국내박사

In recent years, dye-sensitized solar cell (DSC) has been rapidly developed to reach its efficiency from 7% (by O’Regan and Grätzel in1991) to 14.3 % at 1 sun conditions. It has a trend to take place of one of the important traditional energy conversion systems. This dissertation comprises 7 chapters, where Chapter 1 is an introduction on renewable energy technologies to solve various problems associated with fossil fuels. The photovoltaic device and valuable market developments are taken into great description. In addition, the DSCs operating principle and tandem DSCs are simply introduced. In Chapter 2, it is discussed the characteristics of four major components and interfaces such as dye, semiconductor, electrolyte and counter electrode in DSCs. Furthermore the concept of electrochemical impedance spectroscopy (EIS) is explained. In Chapter 3, poly(3,4-ethylenedioxythiophene) (PEDOT) doped with different imidazolium iodides such as DMII, MPII, and BMII showed the improved performance mostly due to the enhancement of Jsc, when compared with Pt or the pristine PEDOT counter electrode. In particular, the NF PEDOT CE doped with 0.1 M DMII has the maximum efficiency of 9.05 %, higher than the traditional Pt counter electrode. Electrochemical techniques such as Tafel, CV, and EIS were used to analyze and understand the increased performance in addition to the Raman spectroscopy. In Chapter 4, mesoporous TiO2 surface engineering is a field to be still advanced toward highly efficient DSCs utilizing solid-state polymer electrolytes. Herein, a facile and stepwise co-sensitization method is applied for ruthenium C106 dye and organic D131 dye. After the C106 dyes are adsorbed on the TiO2 surface, D131 dyes are subsequently co-adsorbed onto empty sites of the mesoporous TiO2 surface without desorption of the C106 dyes. And their quantity is controlled by varying the concentration of the D131 dye solution. At the optimal concentration of the D131 dye solution, a remarkable energy conversion efficiency of 8.4 % is attained under 1 sun illumination conditions, as a result of a noticeable rise in photocurrent density (Jsc). According to the photovoltaic characterizations, both the light harvesting and the charge collection efficiencies prepared by the co-sensitization process is greatly superior to those of individual C106 or D131 dyes. In Chapter 5, the concentration of PVdF-HFP in polymer-gel electrolyte was optimized to be 5 wt%. In addition, various additives such as tBP, LiI, GuNCS were added into the 5 wt% PVdF-HFP based polymer gel electrolyte, and the cell performance was improved to a large extent to the overall energy conversion efficiency of 7.05 % under 1 sun conditions due to the increases in both Voc of 0.7 V and Jsc of 14.00 mA/cm2. In conclusion, it is realized that the overall energy conversion efficiency is promoted up to 30 % by optimizing material and kinetic properties of each component. Therefore, it needs to emphasize the importance of the understanding of structural and kinetic fundamentals on DSC. In this context, the efficiency of solid state DSCs utilizing solid polymer electrolyte would be expected to be increased to a large extent soon. Keyword: dye sensitized solar cells, renewable energies, NF PEDOT, imidazolium iodide, co-sensitizers, C106, D131, PVdF-HFP, tBP, LiI, GuNCS. List of Figure 1 List of Schemes 6 List of Table 7 Nomenclature 9 Abstract (Korean) 13 Chapter 1. Introduction 15 1.1. Background 15 1.2. Evolution of photovoltaic and market value 18 1.3. Dye sensitized solar cells 19 1.4. References 25 Chapter 2. Characterictics of 4 Major Components and Interfaces 27 2.1. Dyes 27 2.2. Nanostructured Metal Oxide Electrodes 28 2.3. Counter electrode 30 2.4. Electrolyte 30 2.5. Characterization of interface and electron transport 33 2.6. References 39 Chapter 3. Polymer Counter Electrode for DSCs Employing Solid Polymer Electrolyte 42 3.1. Introduction 42 3.2.Results & Discussion 49 3.3. Conclusions 60 3.4. References 61 Chapter 4. Co-sensitization of Ruthenium and Organic Dyes 63 4.1 Introduction 63 4.2 Results & Discussion 67 4.3. Conclusions 73 4.4. References 74 Chapter 5. Additive Effects on PVdF-HFP Polymer Electrolyte 76 5.1. Introduction 76 5.2. Results & Discussion 78 5.3. Conclusions 88 5.4. References 89 Chapter 6. Summary 91 Chapter 7. Future prospects 93 7.1. Results & Discussion 93 7.2. References 98 Abstract (English) 99

A Study on Optimal Capacities of a New Cruise Line with Quadratic Operating Cost using Stackelberg Game Model

Wei Wei 한국해양대학교 대학원 2017 국내석사

The cruise industry has grown rapidly since the 1990s and today represents one of the healthiest sectors in the shipping industry, with newer and bigger vessel on order. With the capacity of cruise companies increasing steadily, competition in the market has also become fiercer in several regions. This research study, aims at investigating a market strategy which smaller cruise lines could adopt in order to gain market shares and compete with bigger players. This thesis develops and applies a Stackelberg Game Model to find out what factors are important for a new cruise line penetrating into a particular market. Using the concept of Nash equilibrium, the optimal capacity of the follower, and then the total capacity and the price of the cruise market are computed. Numerical examples with artificial market data are used to derive the most significant managerial implications that could help boosting the share of cruise market for the new entrants (followers). They show how the new entrants (followers) get a suitable way to have a larger market share. Findings of thesis study could be useful for a beginning small cruise line to develop the marketing strategies to increase its market share after entering the cruise market. In addition, this study also provides a method for controlling the variable cost of the follower. KEYWORDS: Nash Equilibrium, Stackelberg Game Model, Marketing Strategies, Cruise Line, Optimal Capacity