To connect different electrical, network and data devices with the minimum cost and shortest path, is a complex job. In huge buildings, where the devices are placed at different locations on different floors and only some specific routes are available to pass the cables and buses, the shortest path search becomes more complex. The aim of this thesis project is, to develop an application which indentifies the best path to connect all objects or devices by following the specific routes.To address the above issue we adopted three algorithms Greedy Algorithm, Simulated Annealing and Exhaustive search and analyzed their results. The given problem is similar to Travelling Salesman Problem. Exhaustive search is a best algorithm to solve this problem as it checks each and every possibility and give the accurate result but it is an impractical solution because of huge time consumption. If no. of objects increased from 12 it takes hours to search the shortest path. Simulated annealing is emerged with some promising results with lower time cost. As of probabilistic nature, Simulated annealing could be non optimal but it gives a near optimal solution in a reasonable duration. Greedy algorithm is not a good choice for this problem. So, simulated annealing is proved best algorithm for this problem. The project has been implemented in C-language which takes input and store output in an Excel Workbook

Telecommunications companies (telcos) and Internet Service Providers (ISPs) monitor the traffic passing through their networks for the purposes of network evaluation and planning for future growth. Most monitoring techniques currently use a form of packet sampling. However, exhaustive monitoring is a preferable solution because it ensures accurate traffic characterization and also allows encoding operations, such as compression and encryption, to be performed. To overcome the very high computational cost of exhaustive monitoring and encoding of data, this thesis suggests exploiting parallelism. By utilizing a parallel cluster in conjunction with load balancing techniques, a simulation is created to distribute the load across the parallel processors. It is shown that a very scalable system, capable of supporting a fairly high data rate can potentially be designed and implemented. A complete system is then implemented in the form of a transparent Ethernet bridge, ensuring that the system can be deployed into a network without any change to the network. The system focuses its encoding efforts on obtaining the maximum compression rate and, to that end, utilizes the concept of streams, which attempts to separate data packets into individual flows that are correlated and whose redundancy can be removed through compression. Experiments show that compression rates are favourable and confirms good throughput rates and high scalability.

One of the petroleum industry?s goals is to reduce the environmental impact of oil and gas operations in environmentally sensitive areas. To achieve this, a number of Environmentally Friendly Drilling (EFD) technologies have been developed to varying degrees. For example, the use of an elevated platform as an alternative to the gravel pad is less intrusive and leads to a more environmentally friendly approach to drilling operations. Elevated drilling platforms will require the use of piles. Another alternative to the gravel pad is the use of composite mats. Since the demand of low impact technologies for drill site construction has rapidly increased, the parametric study for the feasibility of using pile foundations and composite mats is conducted in this research. Even though a number of EFD technologies have already been developed to varying degrees, few have been integrated into a field demonstrable drilling system (i.e., combination of technologies) compatible with ecologically sensitive areas. In general, it is difficult to select the best combination of EFD technologies for a given site because there are many possible combinations and many different evaluation criteria. The proposed technology evaluation method is based on a systems analysis that can be used for integrating current and new EFD technologies into an optimal EFD system. An optimization scheme is suggested based on a combination of multi-attribute utility theory and exhaustively enumerating all possible technology combinations to provide a quantitative rationale and suggest the best set of systems according to a set of criteria, with the relative importance of the different criteria defined by the decision-maker. In this research, the sensitivity of the optimal solution to the weight factors and the effects of the uncertainty of input scores are also discussed using a case study. An application of the proposed approach is described by conducting a case study in Green Lake at McFaddin, TX. The main purpose of this case study is to test the proposed technology evaluation protocol in a real site and then to refine the protocol. This research describes the results of the case study which provided a more logical and comprehensive approach that maximized the economic and environmental goals of both the landowner and the oil company leaseholder.

In this thesis, we propose and evaluate quasi-optimal algorithms for solving the code-constellation optimization problem in M-ary CDMA system. The M-ary CDMA system is a new CDMA architecture. The more spreading codes used in each user, and the higher bandwidth efficiency can achieve with more bits packed in each symbol. We use a code, which we refer to as ¡§mapping code¡¨, to help form a multidimensional spherical code-constellation. The M codewords of the mapping code correspond one-to-one to the M points on the code-constellation. Thus, the code-constellation optimization problem is a combinatorial optimization problem. We present that an exhaustive search (ES) algorithm would have compute and check all possible subset, and then this problem becomes a NP-hard. Based on the exhaustive search algorithm, we propose symmetric points search (SPS) algorithm to reduce computation complexity, but it is not optimal algorithm. In addition, we propose a quasi-optimal algorithm, namely Manhattan distance search (MDS) algorithm. Numerical results and comparisons are provided to illustrate that the computation complexity of the Manhattan distance search algorithm increases linearly with dimension of code-constellation and its performance is better than others.

No presente trabalho propõe-se empregar algoritmos genéticos em associação com o design construtal para a otimização de geometrias em problemas de transferência de calor. O objetivo principal de todos os estudos deste trabalho é minimizar a máxima temperatura que ocorre no domínio computacional. Investigou-se, inicialmente, uma cavidade isotérmica em forma de Y inserida em um sólido retangular com geração de calor uniforme a uma taxa volumétrica constante, onde foi feita uma comparação e validação do algoritmo genético frente à busca exaustiva para poucos graus de liberdade. Após, foi feita uma otimização usando somente algoritmos genéticos considerando todos os quatro graus de liberdade do problema e diferentes valores para suas restrições geométricas. O estudo seguinte foi feito considerando a mesma geometria anteriormente discutida, porém considerou-se as paredes da cavidade Y com uma condição de contorno convectiva. Da mesma forma anterior, foi feita uma validação do algoritmo genético frente à busca exaustiva e na sequência uma otimização de todos os quatro graus de liberdade e diferentes valores do parâmetro convectivo a, empregando somente algoritmos genéticos. No terceiro caso, estudou-se um caminho assimétrico em forma de V de um material de alta condutividade. A geometria tem sua base recebendo um fluxo de calor constante e o remove através das extremidades de dois braços ligados a um sumidouro de calor. Otimizou-se a forma pelo método exaustivo considerando quatro graus de liberdade e uma restrição constante . Após, usou-se algoritmos genéticos para otimizar a geometria considerando os mesmos graus de liberdade e diferentes valores para a restrição de ocupação do material condutivo. Similarmente ao caso da cavidade convectiva em forma de Y, por fim, estudou-se a otimização geométrica de um corpo cilíndrico onde cavidades convectivas retangulares com dois pares de braços são inseridas. Realizaram-se otimizações de até sete graus de liberdade e também se estudou a influência de um parâmetro convectivo e das frações de ocupação das áreas do corpo e braços da cavidade. Deste estudo, concluiu-se que quanto maior o número de cavidades, menores são as máximas temperaturas que ocorrem no domínio. Destaca-se, também, a dependência do parâmetro convectivo, que influenciou na forma da melhor geometria encontrada. Para todos os estudos feitos, os resultados mostraram que a busca por meio de algoritmos genéticos levou a uma redução significativa do número de simulações necessárias para obter a geometria ótima com resultados concordantes aos obtidos com busca exaustiva. Além disso, foi possível estender o estudo para problemas com mais graus de liberdade, restrições e propriedades térmicas. Conclui-se que o melhor design é altamente dependente dos graus de liberdade e restrições, este sendo alcançado de acordo com o princípio construtal da ótima distribuição das imperfeições. / In this work, we propose employing genetic algorithms in association with constructal design for geometry optimization in heat transfer problems. The main objective of all studies is to minimize the maximum temperature that occurs in the computational domain. It was investigated initially an isothermal Y-shaped cavity intruded into a rectangular solid conducting wall with heat generation uniformly at a volumetric rate, where a comparison and validation of genetic algorithm against exhaustive search for few degrees of freedom was made. Then, an optimization is performed by means of genetic algorithms considering all four degrees of freedom of the problem and different values for geometric constraints. The following study has been done considering the same geometry as previously discussed, but it is considered the walls of the Y-cavity with a convective boundary condition. Thus, a dimensionless heat transfer parameter to study (a) was added. Similarly, foregoing study, a genetic algorithm validation was performed comparing to the exhaustive search. After, all four degrees of freedom and different values of a parameter only using genetic algorithms were optimized. In the next investigation, an asymmetric V-shaped pathway of high conductivity material was studied. This geometry receives a constant heat transfer rate in its base and removes it by the end of the two branches that are in touch with the heat sink. The shape was optimized by exhaustive approach considering four degree of freedom and a constraint. After, we used genetic algorithms to optimize the geometry considering the same degrees of freedom and different values for the restriction. Finally, similar to the case of the Y-shaped convective cavity, rectangular convective cavities with two pairs of arms inserted into a cylindrical solid body were optimized. Optimizations of up to seven degrees of freedom were performed and the influence of the convective parameter and of the area fractions of the body and arms of the cavity, were also investigated. From this study, it was concluded that the higher the number of cavities, the lower the maximum temperatures occurring in the domain. Also, the dependence of the convective parameter, influenced in the form of the best geometry, is highlighted. For all studies carried out, the results showed that the search using genetic algorithms led to a significant reduction of the number of simulations required to obtain the optimal geometry. Moreover, it was possible to extend the study where it was considered other degrees of freedom, constraints and thermal properties. We concluded that the best design is highly dependent of degrees of freedom and constraints, and this has been achieved according to the constructal principle of optimal distribution of imperfections.

Este trabalho propõe uma metodologia para enumerar soluções, que indiquem a barra e a compensação de potência reativa necessária para o sistema elétrico sob análise, que atendam aos requisitos avaliados pela função objetivo e as restrições. Nessa alocação de compensação ótima de potência reativa, obtemos as melhores barras e configurações de potências e tecnologias de dispositivos de compensação, minimizando as perdas totais de potência ativa da rede. Em redes fracas com conversores de frequência (por exemplo, para conexão de fontes renováveis, ou interligações utilizando conversores HVDC), esta metodologia proposta busca a melhor relação de curto-circuito trifásico (SCR) no ponto de conexão do conversor de frequência, melhorando a conexão da barra de interesse. O método busca soluções para alocar um único dispositivo de compensação, e soluções alocando simultaneamente dois dispositivos. A metodologia proposta baseia-se na enumeração exaustiva das soluções, e o estudo de caso nos sistemas de 14 e 30 barras do IEEE mostrou a aplicabilidade e funcionalidade da metodologia proposta. / This work proposes a methodology to enumerate solutions, which indicate the bar and the reactive power compensation required for the electrical system under analysis, that meet the requirements evaluated by the objective function and the constraints. In this allocation of optimal compensation of reactive power, we obtain the optimal bars and technologies of compensation devices, minimizing the total losses of active power of the network. In weak networks with frequency converters (e.g. for connection of renewable sources, or interconnections using HVDC converters), the proposed methodology seeks the best threephase short-circuit (SCR) relation at the connection point, improving the connection of the new generation. The method looks for solutions to allocate a single compensation device, and solutions to allocate two devices simultaneously. The proposed methodology is based on the exhaustive enumeration of the solutions. A case study carried out in the IEEE 14 and 30 bus systems shows the applicability and performance of the proposed methodology.

No presente trabalho propõe-se empregar algoritmos genéticos em associação com o design construtal para a otimização de geometrias em problemas de transferência de calor. O objetivo principal de todos os estudos deste trabalho é minimizar a máxima temperatura que ocorre no domínio computacional. Investigou-se, inicialmente, uma cavidade isotérmica em forma de Y inserida em um sólido retangular com geração de calor uniforme a uma taxa volumétrica constante, onde foi feita uma comparação e validação do algoritmo genético frente à busca exaustiva para poucos graus de liberdade. Após, foi feita uma otimização usando somente algoritmos genéticos considerando todos os quatro graus de liberdade do problema e diferentes valores para suas restrições geométricas. O estudo seguinte foi feito considerando a mesma geometria anteriormente discutida, porém considerou-se as paredes da cavidade Y com uma condição de contorno convectiva. Da mesma forma anterior, foi feita uma validação do algoritmo genético frente à busca exaustiva e na sequência uma otimização de todos os quatro graus de liberdade e diferentes valores do parâmetro convectivo a, empregando somente algoritmos genéticos. No terceiro caso, estudou-se um caminho assimétrico em forma de V de um material de alta condutividade. A geometria tem sua base recebendo um fluxo de calor constante e o remove através das extremidades de dois braços ligados a um sumidouro de calor. Otimizou-se a forma pelo método exaustivo considerando quatro graus de liberdade e uma restrição constante . Após, usou-se algoritmos genéticos para otimizar a geometria considerando os mesmos graus de liberdade e diferentes valores para a restrição de ocupação do material condutivo. Similarmente ao caso da cavidade convectiva em forma de Y, por fim, estudou-se a otimização geométrica de um corpo cilíndrico onde cavidades convectivas retangulares com dois pares de braços são inseridas. Realizaram-se otimizações de até sete graus de liberdade e também se estudou a influência de um parâmetro convectivo e das frações de ocupação das áreas do corpo e braços da cavidade. Deste estudo, concluiu-se que quanto maior o número de cavidades, menores são as máximas temperaturas que ocorrem no domínio. Destaca-se, também, a dependência do parâmetro convectivo, que influenciou na forma da melhor geometria encontrada. Para todos os estudos feitos, os resultados mostraram que a busca por meio de algoritmos genéticos levou a uma redução significativa do número de simulações necessárias para obter a geometria ótima com resultados concordantes aos obtidos com busca exaustiva. Além disso, foi possível estender o estudo para problemas com mais graus de liberdade, restrições e propriedades térmicas. Conclui-se que o melhor design é altamente dependente dos graus de liberdade e restrições, este sendo alcançado de acordo com o princípio construtal da ótima distribuição das imperfeições. / In this work, we propose employing genetic algorithms in association with constructal design for geometry optimization in heat transfer problems. The main objective of all studies is to minimize the maximum temperature that occurs in the computational domain. It was investigated initially an isothermal Y-shaped cavity intruded into a rectangular solid conducting wall with heat generation uniformly at a volumetric rate, where a comparison and validation of genetic algorithm against exhaustive search for few degrees of freedom was made. Then, an optimization is performed by means of genetic algorithms considering all four degrees of freedom of the problem and different values for geometric constraints. The following study has been done considering the same geometry as previously discussed, but it is considered the walls of the Y-cavity with a convective boundary condition. Thus, a dimensionless heat transfer parameter to study (a) was added. Similarly, foregoing study, a genetic algorithm validation was performed comparing to the exhaustive search. After, all four degrees of freedom and different values of a parameter only using genetic algorithms were optimized. In the next investigation, an asymmetric V-shaped pathway of high conductivity material was studied. This geometry receives a constant heat transfer rate in its base and removes it by the end of the two branches that are in touch with the heat sink. The shape was optimized by exhaustive approach considering four degree of freedom and a constraint. After, we used genetic algorithms to optimize the geometry considering the same degrees of freedom and different values for the restriction. Finally, similar to the case of the Y-shaped convective cavity, rectangular convective cavities with two pairs of arms inserted into a cylindrical solid body were optimized. Optimizations of up to seven degrees of freedom were performed and the influence of the convective parameter and of the area fractions of the body and arms of the cavity, were also investigated. From this study, it was concluded that the higher the number of cavities, the lower the maximum temperatures occurring in the domain. Also, the dependence of the convective parameter, influenced in the form of the best geometry, is highlighted. For all studies carried out, the results showed that the search using genetic algorithms led to a significant reduction of the number of simulations required to obtain the optimal geometry. Moreover, it was possible to extend the study where it was considered other degrees of freedom, constraints and thermal properties. We concluded that the best design is highly dependent of degrees of freedom and constraints, and this has been achieved according to the constructal principle of optimal distribution of imperfections.

Este trabalho propõe uma metodologia para enumerar soluções, que indiquem a barra e a compensação de potência reativa necessária para o sistema elétrico sob análise, que atendam aos requisitos avaliados pela função objetivo e as restrições. Nessa alocação de compensação ótima de potência reativa, obtemos as melhores barras e configurações de potências e tecnologias de dispositivos de compensação, minimizando as perdas totais de potência ativa da rede. Em redes fracas com conversores de frequência (por exemplo, para conexão de fontes renováveis, ou interligações utilizando conversores HVDC), esta metodologia proposta busca a melhor relação de curto-circuito trifásico (SCR) no ponto de conexão do conversor de frequência, melhorando a conexão da barra de interesse. O método busca soluções para alocar um único dispositivo de compensação, e soluções alocando simultaneamente dois dispositivos. A metodologia proposta baseia-se na enumeração exaustiva das soluções, e o estudo de caso nos sistemas de 14 e 30 barras do IEEE mostrou a aplicabilidade e funcionalidade da metodologia proposta. / This work proposes a methodology to enumerate solutions, which indicate the bar and the reactive power compensation required for the electrical system under analysis, that meet the requirements evaluated by the objective function and the constraints. In this allocation of optimal compensation of reactive power, we obtain the optimal bars and technologies of compensation devices, minimizing the total losses of active power of the network. In weak networks with frequency converters (e.g. for connection of renewable sources, or interconnections using HVDC converters), the proposed methodology seeks the best threephase short-circuit (SCR) relation at the connection point, improving the connection of the new generation. The method looks for solutions to allocate a single compensation device, and solutions to allocate two devices simultaneously. The proposed methodology is based on the exhaustive enumeration of the solutions. A case study carried out in the IEEE 14 and 30 bus systems shows the applicability and performance of the proposed methodology.

<div>By Machine Learning in the Open World, we are trying to build models that can be used in a more realistic setting where there could always be something "unknown" happening. Beyond the traditional machine learning tasks such as classification and segmentation where all classes are predefined, we are dealing with the challenges from newly emerged classes, irrelevant classes, outliers, and class imbalance.</div><div>At the beginning, we focus on the Non-Exhaustive Learning (NEL) problem from a statistical aspect. By NEL, we assume that our training classes are non-exhaustive, where the testing data could contain unknown classes. And we aim to build models that could simultaneously perform classification and class discovery. We proposed a non-parametric Bayesian model that learns some hyper-parameters from both training and discovered classes (which is empty at the beginning), then infer the label partitioning under the guidance of the learned hyper-parameters, and repeat the above procedure until convergence.</div><div>After obtaining good results on applications with plain and low dimensional data such flow-cytometry and some benchmark datasets, we move forward to Non-Exhaustive Feature Learning (NEFL). For NEFL, we extend our work with deep learning techniques to learn representations on datasets with complex structural and spatial correlations. We proposed a metric learning approach to learn a feature space with good discrimination on both training classes and generalize well on unknown classes. Then we developed some variants of this metric learning algorithm to deal with outliers and irrelevant classes. We applied our final model to applications such as open world image classification, image segmentation, and SRS hyperspectral image segmentation and obtained promising results.</div><div>Finally, we did some explorations with Out of Distribution detection (OOD) to detect irrelevant sample and outliers to complete the story.</div>

No description available.