Over the past decade, fuzzy systems have gained remarkable acceptance in many fields including control and automation, pattern recognition, medical diagnosis and forecasting. The fuzzy system application has also been accepted as a promising approach to dealing with uncertainty in real estate valuation analysis. This is mainly due to the necessity of coping with a large number of qualitative and quantitative variables that affect the value of a real property. The appraisers use a great deal of judgment to identify both the characteristics that contribute to property values and the relationships among these characteristics in order to derive estimates of market values. This thesis uses the two widely-used fuzzy rule-based systems / namely the Mamdani and Takagi- Sugeno-Kang (TSK) type fuzzy models in an attempt to examine the main determinants of office rents in Istanbul Central Business District (CBD). The input variables of the fuzzy rule-based systems (FRBS) comprise: i) physical attributes of office spaces and office buildings, ii) lease contract terms, and iii) tenants&rsquo / perception of the office rent determinants, tenants&rsquo / location of residence, tenants&rsquo / transportation modes, etc and as the output the system proposes the office property&rsquo / s rental price. Obtaining office rent determinants is a significant issue for both practitioners and academics. While,practitioners use them directly in demand and sensitivity analyses, academics are more interested in the relative significance of these variables and their effect on the variation in office rent to forecast market behavior. Our data set includes a detailed survey of 500 office spaces located in Istanbul CBD. We have carried out two Mamdani-type FRBS and two TSK-type FRBS for the office space and office building data sets. In these FRBS analyses, firstly the so-called representative office spaces are determined, then the average office space rents are estimated. Finally, the spatial variation in the average office rents across the CBD sub-districts, along with the Office space rent variations with respect to different clusters, like number of workers, number of floors and so on, have been analyzed. We believe that presenting the spatial variation in office rents will make a noteworthy contribution both to the real estate investors and appraisers interested in Istanbul office market.

As the complexity of a spatial phenomenon increases, traditional modeling becomes impractical. Alternatively, data-driven modeling, which is based on the analysis of data characterizing the phenomena, can be used. In this thesis, the generation of understandable and reliable spatial models using observational data is addressed. An interpretability oriented data-driven fuzzy modeling approach is proposed. The methodology is based on construction of fuzzy models from data, tuning and fuzzy model simplification. Mamdani type fuzzy models with triangular membership functions are considered. Fuzzy models are constructed using fuzzy clustering algorithms and simulated annealing metaheuristic is adapted for the tuning step. To obtain compact and interpretable fuzzy models a simplification methodology is proposed. Simplification methodology reduced the number of fuzzy sets for each variable and simplified the rule base. Prototype software is developed and mean annual precipitation data of Turkey is examined as case study to assess the results of the approach in terms of both precision and interpretability. In the first step of the approach, in which fuzzy models are constructed from data, &quot / Fuzzy Clustering and Data Analysis Toolbox&quot / , which is developed for use with MATLAB, is used. For the other steps, the optimization of obtained fuzzy models from data using adapted simulated annealing algorithm step and the generation of compact and interpretable fuzzy models by simplification algorithm step, developed prototype software is used. If the accuracy is the primary objective then the proposed approach can produce more accurate solutions for training data than geographically weighted regression method. The minimum training error value produced by the proposed approach is 74.82 mm while the error obtained by geographically weighted regression method is 106.78 mm. The minimum error value on test data is 202.93 mm. An understandable fuzzy model for annual precipitation is generated only with 12 membership functions and 8 fuzzy rules. Furthermore, more interpretable fuzzy models are obtained when Gath-Geva fuzzy clustering algorithms are used during fuzzy model construction.

Conselho Nacional de Desenvolvimento CientÃfico e TecnolÃgico / Este trabalho propÃe uma metodologia para estudar o comportamento das concentraÃÃes da Demanda BioquÃmica de OxigÃnio e do OxigÃnio Dissolvido em um rio sujeito a lanÃamentos de efluentes. O estudo à baseado na transformaÃÃo do Modelo MatemÃtico de Streeter-Phelps em um modelo de natureza fuzzy, onde as concentraÃÃes de DBO e de OD sÃo calculadas na forma de funÃÃes de pertinÃncia. Desta forma, à possÃvel incorporar incertezas no modelo, o que permite desenvolver uma metodologia para a determinaÃÃo do risco de um corpo hÃdrico nÃo atender as condiÃÃes de uso previsto por norma, quando recebe uma carga poluente, proveniente de uma concessÃo de outorga de lanÃamento de efluentes. A pesquisa usa um programa computacional, desenvolvido para este trabalho, para calcular, a partir das equaÃÃes do modelo, as concentraÃÃes de DBO e de OD nas suas formas fuzzy. TambÃm foi desenvolvida uma sub-rotina que permite que sejam calculados o risco e a confiabilidade para o rio em questÃo que venha a receber lanÃamentos de efluentes. Os resultados mostraram que esta metodologia fuzzy à uma alternativa para ser considerada nas questÃes pertinentes à gestÃo integrada dos recursos hÃdricos. / This research proposes a methodology to study the behavior of the Biochemical Oxygen Demand and of the Dissolved Oxygen Concentrations, in a river, subject to effluent discharges. The study is based on the transformation of the Streeter-Phelps Mathematical Model to a Streeter-Phelps Fuzzy Model, where the concentrations of BDO and of DO are calculated as membership functions. In such way, it is possible to incorporate uncertainties in the model, so that it allows developing a methodology for the determination risk and the reliability of a body of water do not assist the use conditions established by norm, when it receives a load pollutant, originating from a concession of grants of effluent discharges. The research uses a computational program, developed for this research, to make calculations, starting from the model equations, the BOD and the DO concentrations, in the membership functions. It was also developed a subroutine that allows that the risk and the reliability could be calculated for the river, subject that receives effluents discharges. The results showed that this fuzzy methodology is an alternative to be considered in the questions related with Integrated Water Resources Management.

Ce travail de thèse traite de l’analyse de la stabilité et la stabilisation des systèmes non-linéaires représentés par des modèles T-S. L’objectif est de réduire le conservatisme des conditions de stabilité, obtenue par la méthode directe de Lyapunov, et écrites, dans la mesure du possible, sous forme de LMIs. Dans ce cadre, deux contributions principales ont été apportées. Tout d’abord, nous avons proposé de nouvelles conditions de synthèse non-quadratique de lois de commande, strictement LMIs et sans restriction d’ordre, pour les modèles T-S via des FLICs. En effet, dans ce contexte, les résultats de la littérature ne sont valables que pour les modèles T-S d’ordre inférieur ou égal à 2. Afin de lever cette restriction, les conditions ont été obtenues grâce à la démonstration d’une propriété de dualité. Ensuite, peu de travaux traitant de la spécification des performances en boucle fermée, de nouvelles conditions LMIs (quadratiques et non-quadratiques) ont été proposées via le concept de D-stabilité. Dans un premier temps, la synthèse de lois de commande PDC et non-PDC D-stabilisantes a été proposée pour les modèles T-S nominaux. Ensuite, ces résultats ont été étendus au cas des modèles T-S incertains. De plus, nous avons mis en évidence, au travers d’un exemple de D-stabilisation en attitude d’un modèle de drone quadrirotor, que les modèles T-S incertains pouvaient être avantageusement considérés lorsque les non-linéarités d’un modèle non-linéaire dépendent à la fois de l’état et de l’entrée. / This work deals with the stability analysis and the stabilisation of nonlinear systems represented by T-S models.The goal is to reduce the conservatism of the stability conditions, obtained through the direct Lyapunov methodand written, when it is possible, as LMIs. In this framework, two main contributions has been proposed. First ofall, we have proposed some new conditions based on FLICs, strictly LMIs and without any order restrictions, forthe non-quadratic design of control laws devoted to stabilize T-S models. Indeed, in this non-quadratic context,the existing works are only available for 2nd order T-S models. In order to unlock this restriction, the proposed conditions have been obtained based on the proof of a dual property. Then, starting from the fact that few worksdeals with the closed-loop performances specification, some new LMI conditions (quadratic and non-quadratic)have been proposed via the D-stability concept. As a first step, D-stabilizing PDC and non-PDC controller designhas been considered for nominal T-S models. Then, these results have been extended to uncertain T-S models.Moreover, it has been highlighted, from an example of the attitude D-stabilization of a quadrotor model, that wecan make use of uncertain T-S models to cope with nonlinear models involving nonlinearities depending on bothstate and input variables.

The diploma thesis deals with the application of the theory of fuzzy logic in the evaluation of client translation commissions for a foreign language text. This fuzzy model is used for better selection of orders and faster allocation of human resources for specific orders. The fuzzy model is composed of multi-valued decision-making criteria that are essential for the company. The model is processed in MS Excel using VBA and MathWorks MATLAB.

Gas Metal Arc Welding (GMAW) is one of the main methods used to join structural members. One of the largest challenges involved in production of welds is ensuring the quality of the weld. One of the main factors attributing to weld quality is penetration depth. Automatic control of the welding process requires non-contact, non-destructive sensors that can operate in the presence of high temperatures and electrical noise found in the welding environment. Inspection using laser generation and electromagnetic acoustic transducer (EMAT) reception of ultrasound was found to satisfy these conditions. Using this technique, the time of flight of the ultrasonic wave is measured and used to calculate penetration depth. Previous works have shown that penetration depth measurement performance is drastically reduced when performed during welding. This work seeks to realize in-process penetration depth measurement by compensating for errors caused by elevated temperature. Neuro-fuzzy models are developed that predict penetration depth based on in-process time of flight measurements and the welding process input. Two scenarios are considered in which destructive penetration depth measurements are or are not available for model training. Results show the two scenarios are successful. When destructive measurements are unavailable, model error is comparable to that of offline ultrasonic measurements. When destructive measurements are available, measurement error is reduced by 50% compared to offline ultrasonic measurements. The two models can be effectively applied to permit in-process penetration depth measurements for the purpose of real-time monitoring and control. This will reduce material, production time, and labor costs and increase the quality of welded parts.

Portfolio optimisation has a number of constraints resulting from some practical matters and regulations. The closed-form mathematical solution of portfolio optimisation problems usually cannot include these constraints. Exhaustive search to reach the exact solution can take prohibitive amount of computational time. Portfolio optimisation models are also usually impaired by the estimation error problem caused by lack of ability to predict the future accurately. A number of Multi-Objective Genetic Algorithms are proposed to solve the problem with two objectives subject to cardinality constraints, floor constraints and round-lot constraints. Fuzzy logic is incorporated into the Vector Evaluated Genetic Algorithm (VEGA) to but solutions tend to cluster around a few points. Strength Pareto Evolutionary Algorithm 2 (SPEA2) gives solutions which are evenly distributed portfolio along the effective front while MOGA is more time efficient. An Evolutionary Artificial Neural Network (EANN) is proposed. It automatically evolves the ANN's initial values and structures hidden nodes and layers. The EANN gives a better performance in stock return forecasts in comparison with those of Ordinary Least Square Estimation and of Back Propagation and Elman Recurrent ANNs. Adaptation algorithms for selecting a pair of forecasting models, which are based on fuzzy logic-like rules, are proposed to select best models given an economic scenario. Their predictive performances are better than those of the comparing forecasting models. MOGA and SPEA2 are modified to include a third objective to handle model risk and are evaluated and tested for their performances. The result shows that they perform better than those without the third objective.

Orientadores: Wagner Caradori do Amaral, Ricardo Jose Grabrielli Barreto Campello / Dissertação (mestrado) - Universidade Estadual de Campinas, Faculdade de Engenharia Eletrica e de Computação / Made available in DSpace on 2018-08-09T14:19:44Z (GMT). No. of bitstreams: 1 Machado_JeremiasBarbosa_M.pdf: 6477617 bytes, checksum: 3f0c4fec476306e8cc05a7940894b0a0 (MD5) Previous issue date: 2007 / Resumo: o interesse na utilização de algoritmos de controle sofisticados cresce no meio industrial devido à necessidade de melhor qualidade dos produtos produzidos. Uma abordagem que vem ganhando destaque é a utilização de sistemas de controle não-linear que modelam os sistemas por meio de multimodelos lineares. Neste contexto, este trabalho apresenta a modelagem e controle de sistemas não-lineares através de controladores preditivos não-lineares que utilizam multimodelos lineares. Os controladores preditivos baseados em modelos (MBPC - Model Based Predictive Controllers) são controladores cuja principal característica é a utilização de um modelo na determinação de um conjunto de previsões de saída, e a lei de controle é calculada em função destas previsões minimizando-se uma função de custo. O desempenho deste controlador depende da qualidade do modelo utilizado para predição dos sinais de saída. A proposta do trabalho é modelar as não-linearidades do processo sob controle através de modelos fuzzy Takagi-Sugeno - TS com funções de base ortonormal - FBO nos conseqüentes das regras. As FBO's apresentam diversas características conceituais e estruturais de interesse na elaboração dos modelos utilizados nos controladores preditivos, como a ausência de realimentação de saída, o que evita a propagação de erro, além de outras que serão discutidas ao longo deste trabalho. Os parâmetros de um modelo fuzzy TS a serem determinados são os antecedentes das regras, com suas funções de pertinência, e as funções nos conseqüentes das regras, que neste trabalho dar-se-ão de forma automática, sendo os antecedentes das regras obtidos através de agrupamento fuzzy (fuzzy clustering) das amostras de entrada e saída. Para esta tarefa será utilizado o algoritmo de GustafsonKessel. A fim de determinar o número de grupos que irão compor o modelo e, por conseqüência, defil)ir o número de regras e modelos locais, utilizar-se-ão critérios que avaliam a qualidade dos agrupamentos juzzy, como Fuzzy Silhouette, Fuzzy Hipervolume, Average Partition Density e Average Within-Cluster Distance, sendo proposta a combinação dos resultados obtidos em cada um dos critérios. O controle é feito de forma que, para cada modelo local, presente no modelo fuzzy TS-FBO, tem-se um controlador atuando sobre este. As ações de controle locais são combinadas conforme a ativação de cada regra do respectivo modelo local, e a ação de controle global resultante dessa combinação é aplicada ao processo a ser controlado. A abordagem proposta apresenta vantagens estruturais na modelagem e controle de processos nãolineares, quando comparado a outras metodologias de modelagem (como modelos polinomiais NARMAX) e controle, uma vez que esta abordagem é composta de uma estrutura simples com modelos locais lineares (ou afins) formados por FBO's. Para ilustrar o que foi desenvolvido, são apresentadas, no final destes trabalho, implementações na modelagem e controle de processos não-lineares / Abstract: The use of advanced control strategies has been increased in the last years due to the needs of more accurate quality on products. An approach that seems attractive on control and modeling of the nonlinear processes is the use of multiple linear models. In this context, this work presents an altemative approach for modeling and controlling nonlinear processes through nonlinear predictive control (NMBPC) using multi-models. The main characteristic of the Model Based Predictive Controllers is the use of a model for the determination ofthe output predictions. The controllaw is derived based on these output predictions, minimizing a specified cost function. Its performance is directly related to the quality of the model predictor. Therefore, in this work, the process is modeling through Takagi-Sugeno- TS fuzzy models with orthonormal base functions - OBF - on the mIes consequents. OBF' s models present several conceptual and structural characteristics of interest on the elaboration of models predictors, such as, absence of output recursion and feedback of prediction errors, often leading to superior performances over long-range horizon predictions and natural decoupling between multiple outputs; there is no need for previous knowledge about the relevant past terms of the system signals; the representation of a stable system is assuredly stable; tolerance to unmodeled dynamics; ability to deal with time delays. The antecedents ofthe TS fuzzy models are obtained through fuzzy c1ustering ofthe input and output measures. The algorithm of Gustafson-Kessel is used to perform this task. In order to determine the number ofthe local models, clustering validity criteria such as Fuzzy Silhouette, Fuzzy Hipervolume, Average Partition Density e Average Within-Cluster Distance are used. A predictive controller is derived for local model and the global controllaw is obtained by combining each local control law, using the degree of activation of every mIe of the respective local model. The proposed approach presents structural advantages in the modeling and controlling nonlinear process, when compared to other modeling (like polynomial models-NARMAX) and controlling strategies, as this approach is constituted of a simple structure with linear local models using OBF' s. The performance of the proposed strategies is illustrated using some simulated examples / Mestrado / Automação / Mestre em Engenharia Elétrica

This diploma thesis deals with the design of fuzzy models to support decision making for selecting the most suitable suppliers for PSL, a.s. Describes methods and procedures for modeling in MS Excel and MATLAB. The goal is to create a decision-making system that will evaluate suppliers to optimize the choice of the most suitable supplier based on the requirements of the selected company.

This master´s thesis “The Localization of Foreign Direct Investment in the region of Middle Europe“ is focused on localization of foreign direct investment in the region of Middle Europe and possiblities of making easier localization decision. The first part of the thesis summarizes teoretical knowledge of foreign direct investment, their difinitions, typology and effects on economy. The second part is focused on the analysis of localization factors of strategic services sector. On the analysis basis, in the final part is created the model, which can be used for easier localization decision making between economies of Middle Europe. Afterwards the localization of a foreign direct investment is projected and analysed.