This thesis consists of five papers. Paper [1] analyzes one possible way of replacing dynamic Pigouvian taxes by a static approximation of such taxes from the point of view of social accounting. The idea is to approximate a Pigouvian emission tax by using the instantaneous marginal willingness to pay to reduce the stock of pollution. If this approximation is close enough to the correct Pigouvian tax it will be useful for at least two reasons: (i) it brings the economy close to the socially optimal solution; and (ii) it provides information relevant for social accounting by closely approximating the value of additions to the stock of pollution. Paper [2] analyzes the welfare effects of an agreement between countries to slightly increase their emission taxes. The results indicate that such an agreement need not necessarily increase the global welfare level, even if each individual country has set its prereform emission tax to be lower than the marginal social cost of pollution. Paper [3] provides an economic framework for analyzing the global warming problem, emphasizing the use of forests as a means of carbon sequestration. We explore the difference between the decentralized economy and the socially optimal resource allocation, and discuss the appropriate tax system required to implement the first best optimum. Paper [4] incorporates the uncertainty involved in the production of nuclear energy into a dynamic general equilibrium growth model. We compare the resource allocation in the decentralized economy with the socially optimal resource allocation and design the dynamic Pigouvian taxes that make the decentralized economy reproduce the socially optimal resource allocation. Paper [5] treats externalities from nuclear power in a dynamic differential game framework involving two countries, which differ with regard to their nuclear technology. The model is solved numerically, where one country is considered relatively safe and the other relatively less safe. / <p>Härtill 5 delarbeten.</p> / digitalisering@umu

Un cadre d'optimisation multi-Objectif pour les pales d'hélicoptère est développé en utilisant des modèles de simulation de haute fidélité. Des fonctions objectives caractérisant la performance de l'hélicoptère en vol stationnaire et d'avancement sont retenues. Deux solveurs sont utilisés afin de prédire la performance du rotor: le code général HOST et le solveur CFD elsA. Le premier axe de recherche porte sur la caractérisation de la précision des méthodes de prédiction. L'influence de la prise en compte de la souplesse de la pale, du trim rotor et/ou de l'utilisation de modélisations aérodynamiques simplifiées est caractérisée pour chaque cas de vol en utilisant des mesures en soufflerie. Un cadre numérique adapté à l'optimisation est développé. Le deuxième axe de recherche porte sur des techniques d'optimisation multi-Objectif de pales en vol stationnaire et d'avancement. Deux algorithmes novateurs basés sur la compétition (Jeux de Nash) et la coopération (Descente à Gradients Multiples) sont présentés comme des alternatives aux approches traditionnelles pour traiter le problème multi-Objectif. Afin de réduire le coût des simulations, un cadre de simulation basé sur des métamodèles est développé y compris une stratégie multi-Fidélité pour prédire la performance du rotor en vol d'avancement. Ces techniques sont appliquées à un cas réaliste de rotor, en utilisant des simulations CFD trimmées avec pale souple pour le cas du vol d'avancement et des simulations CFD avec pale rigide en vol stationnaire. Les résultats sont ensuite analysés, démontrant le potentiel de ces techniques pour l'obtention de conceptions réalistes représentant des bons compromis entre les objectifs. / This work addresses the development of a multi-Objective optimization framework for helicopter rotor blades using high-Fidelity simulation models. In particular, objective functions corresponding to hover and forward flight are considered. Two solvers are used to predict the rotor performance: the comprehensive rotor code HOST and the Computational Fluid Dynamics (CFD) solver elsA. The first research axis of this work is the characterization of the accuracy of each available prediction method. The influence of considering the blade elasticity, the rotor trim and/or simplified aerodynamics is characterized for each flight case using wind-Tunnel data. As a result, a numerical framework adapted to the optimization is developed. The second part of this work concerns the formulation and development of techniques adapted to the multi-Objective optimization of rotor blades in hover and in forward flight. Innovative algorithms based on competition (Nash Games) and cooperation (Multi-Gradient Descent) are presented as alternatives to traditional multi-Objective approaches. In order to reduce the simulation costs, a surrogate-Based framework is developed, including a multi-Fidelity strategy to predict the rotor performance in forward flight. These techniques are finally applied to a realistic rotor, considering trimmed elastic CFD computations in the forward flight case and rigid blade CFD computations in the hover case. The results are subsequently analyzed, demonstrating the potential of these techniques to obtain realistic designs realizing interesting trade-Offs.

京都大学 / 新制・課程博士 / 博士(情報学) / 甲第24741号 / 情博第829号 / 新制||情||139(附属図書館) / 京都大学大学院情報学研究科数理工学専攻 / (主査)教授 山下 信雄, 教授 太田 快人, 教授 永持 仁 / 学位規則第4条第1項該当 / Doctor of Informatics / Kyoto University / DFAM

Game theory is the study of tactical interactions involving conflicts and cooperations among multiple decision makers called players with applications in diverse disciplines such as economics, biology, management, communication networks, electric power systems and control. This dissertation studies a statistical differential game problem where finite N players optimize their system performance by shaping the distribution of their cost function through cost cumulants. This research integrates game theory with statistical optimal control theory and considers a statistical Nash non-cooperative nonzero-sum game for a nonlinear dynamic system with nonquadratic cost functions. The objective of the statistical Nash game is to find the equilibrium solution where no player has the incentive to deviate once other players maintain their equilibrium strategy. The necessary condition for the existence of the Nash equilibrium solution is given for the m-th cumulant cost optimization using the Hamilton-Jacobi-Bellman (HJB) equations. In addition, the sufficient condition which is the verification theorem for the existence of Nash equilibrium solution is given for the m-th cumulant cost optimization using the Hamilton-Jacobi-Bellman (HJB) equations. However, solving the HJB equations even for relatively low dimensional game problem is not trivial, we propose to use neural network approximate method to find the solution of the HJB partial differential equations for the statistical game problem. Convergence proof of the neural network approximate method solution to exact solution is given. In addition, numerical examples are provided for the statistical game to demonstrate the applicability of the proposed theoretical developments. / Electrical and Computer Engineering

<p>This thesis consists of a summary and four papers. The first two papers address political economy and indus-trial organization aspects of agricultural policy, and the last two international aspects of environmental policy.</p><p>Paper [I] explains Common Agricultural Policy (CAP) subsidies to farmers by the influence of farmer interest-groups with an EU-wide membership. The analysis is based on panel-data for fifteen commodities over the period 1986-2003. Because the CAP is set as an overall EU policy, effective lobbying presents a collective ac-tion problem to the farmers in the EU as a whole. Indicators of lobbying, which are based on this perception, are found to explain part of the variation in agricultural support.</p><p>In Paper [II], the Bresnahan-Lau framework is used to analyze whether policy reforms, i.e. the two-price sys-tem (an input quota, 1986-1991) and a general deregulation of dairy policy (1991-1994) had any market power effects on the Swedish butter market. The results show that the null hypothesis of no market power cannot be rejected, for any of the specific policy reforms, at any reasonable significance level.</p><p>Paper [III] concerns the welfare consequences of environmental policy cooperation. It is assumed that coun-tries finance their public expenditures by using distortionary taxes, and that they differ with respect to compe-tition in the labor market. It is shown how the welfare effect of an increase in the expenditures on abatement depends on changes in the environmental damage, employment and work hours. The welfare effect is also related to the strategic interaction among the countries in the prereform equilibrium.</p><p>In Paper [IV] environmental policy in an economic federation, where each national government faces a mixed tax problem, is addressed. It is assumed that the federal government sets emission targets, which are imple-mented at the national level. It is also assumed that the economic federation is decentralized. The results high-light a strategic role of income and commodity taxation, i.e. each country uses its policy instruments, at least in part, to influence the emission target.</p>

This thesis consists of a summary and four papers. The first two papers address political economy and indus-trial organization aspects of agricultural policy, and the last two international aspects of environmental policy. Paper [I] explains Common Agricultural Policy (CAP) subsidies to farmers by the influence of farmer interest-groups with an EU-wide membership. The analysis is based on panel-data for fifteen commodities over the period 1986-2003. Because the CAP is set as an overall EU policy, effective lobbying presents a collective ac-tion problem to the farmers in the EU as a whole. Indicators of lobbying, which are based on this perception, are found to explain part of the variation in agricultural support. In Paper [II], the Bresnahan-Lau framework is used to analyze whether policy reforms, i.e. the two-price sys-tem (an input quota, 1986-1991) and a general deregulation of dairy policy (1991-1994) had any market power effects on the Swedish butter market. The results show that the null hypothesis of no market power cannot be rejected, for any of the specific policy reforms, at any reasonable significance level. Paper [III] concerns the welfare consequences of environmental policy cooperation. It is assumed that coun-tries finance their public expenditures by using distortionary taxes, and that they differ with respect to compe-tition in the labor market. It is shown how the welfare effect of an increase in the expenditures on abatement depends on changes in the environmental damage, employment and work hours. The welfare effect is also related to the strategic interaction among the countries in the prereform equilibrium. In Paper [IV] environmental policy in an economic federation, where each national government faces a mixed tax problem, is addressed. It is assumed that the federal government sets emission targets, which are imple-mented at the national level. It is also assumed that the economic federation is decentralized. The results high-light a strategic role of income and commodity taxation, i.e. each country uses its policy instruments, at least in part, to influence the emission target.