[pt] A simulação da estrutura eletrônica de átomos e moléculas, desde do início da década de 90, tem se mostrado uma ferramenta imprescindível para o desenvolvimento de áreas estratégicas, ainda emergentes, mas fundamentais, como por exemplo, a área de Nanotecnologia. No entanto, esse tipo de simulação, ainda hoje é de grande complexidade e exige alto poder computacional. Dessa forma, torna-se fundamental a criação de métodos de simulação mais precisos e computacionalmente menos custosos. Este trabalho utiliza Algoritmos Evolutivos e métodos de Inteligência Computacional junto de algumas ferramentas desenvolvidas e estudadas tradicionalmente pela Matemática Aplicada em cálculos de estrutura eletrônica. Em particular, são construídas novas formas de aproximação de soluções para equação de Schrodinger, que contemplem os requisitos físicos necessários. Essas soluções serão nomeadas de funções de ondas evolucionárias, que neste trabalho serão tratadas como pontos em um espaço de Hilbert formado pelo fecho em L (2) da interseção das funções definidas na semi-reta, contínuas e de derivadas contínuas, simétricas ou antissimétricas com relação à permutação de suas coordenadas. São demonstrados alguns resultados, requisitos para utilização de Algoritmos Evolucionários e Séries de Fourier Generalizadas, baseadas em polinômios de Laguerre modificados e Wavelets. Esta pesquisa é desenvolvida inicialmente para sistemas de dois elétrons, e mais tarde é estendida para sistemas mais complexos, a fim de criar uma abordagem alternativa as tradicionais. / [en] The simulation of the electronic structure of atoms and molecules has been shown to be, from the beginning of 90 s, an indispensable tool for the development of strategic areas, that are still emergent, but fundamental, like nanotechnology. However, this type of simulation is still of great complexity today and demands high computational power. Thus, the creation of more precise and less costly methods becomes fundamental. With the elaboration of this research, the intention is to create alternatives basis that can be used into the traditional methods of simulation of electronic structure, such as the Hartree-Fock method, GVB, among others. This essay intends to investigate part of the mathematical tools used in the calculations of electronic structure, in order to create disruptive approaches, related to the precision or velocity of the obtainment of relevant results. The new methods are based in Computational Intelligence and concepts of Functional Analysis like Wavelets.

[pt] A abstração matemática de um processo físico é essencial em problemas de engenharia, pois muitas vezes pode ser impraticável ou impossível realizar experimentos no sistema real. Além disso, modelos matemáticos são mais flexíveis que protótipos físicos, permitindo um rápido refinamento dos projetos do sistema para otimizar várias medidas de desempenho. As aplicações dos modelos podem ser divididas em quatro partes, a saber: projeto, estimativa, controle e monitoramento. Algumas aplicações específicas são i) simulações, ii) soft sensors, iii) avaliação de desempenho, iv) controle estatístico de qualidade e v) detecção e diagnóstico de falhas. Este trabalho visa então: i) desenvolver diferentes classes de modelos capazes de simular com precisão a variável de saída de um sistema, ii) avaliar a eficiência dos algoritmos de otimização utilizados na tarefa de estimação de parâmetros, iii) avaliar qual modelo de atrito é o mais adequado para descrever esse fenômeno em um sistema de posicionamento. Os resultados mostraram que o atrito no sistema de posicionamento apresenta comportamento não linear e assimétrico, já que alguns termos dos modelos de atrito relacionados às velocidades positiva e negativa são significativamente diferentes um do outro. O resultado final do processo de otimização que usou um algoritmo de busca local foi altamente dependente das condições iniciais e do número de parâmetros estimados, o que elevou o erro de simulação. Entretanto, melhores estimativas da variável de saída foram alcançadas quando essa abordagem foi combinada com outros modelos de diferentes classes. Através dessa última abordagem o erro relativo foi reduzido em mais de 20 porcento. As simulações realizadas com os parâmetros estimados pelos algoritmos evolucionários foram mais acuradas, eles foram capazes de reduzir o erro relativo em quase 30 porcento quando comparados com o algoritmo de busca local. Considerando o segundo estudo de caso, o otimizador baseado em árvores de decisão se mostrou igualmente eficaz se comparado aos algoritmos evolucionários. O erro relativo das simulações usando os parâmetros estimados por esses algoritmos foi inferior a 8 porcento. Além disso, a forma do atrito reconstruído na segunda junta do manipulador robótico através dos parâmetros estimados pelos algoritmos está de acordo com o esperado. / [en] The mathematical abstraction of a physical process is essential in engineering problems, as it can often be impractical or impossible to perform experiments on the real system. Besides, mathematical models are more flexible than physical prototypes, allowing for quick refinement of system designs to optimize various performance measures. The applications of the models can be divided into four parts, namely: design, estimation, control and monitoring. Some specific applications are i) simulations, ii) soft sensors, iii) performance evaluation, iv) statistical quality control and, v) fault detection and diagnosis. This work aims to: i) develop different classes of models capable of accurately simulating the output variable of a system, ii) evaluate the efficiency of optimization algorithms used in the parameter estimation task, iii) assess which friction model is the most appropriate to describe this phenomenon in a positioning system. The results showed that the friction in the positioning system presents a nonlinear and asymmetric behavior since some terms of the friction models related to the positive and negative velocities are significantly different from each other. The final result of the optimization process that used a local search algorithm was highly dependent on the initial conditions and the number of estimated parameters, which increased the simulation error. However, better estimates of the output variable were achieved when this approach was combined with other models of different classes. Through this last approach, the relative error was reduced by more than 20 percent. The simulations performed with the parameters estimated by the evolutionary algorithms were more accurate, they were able to reduce the relative error by almost 30 percent when compared with the local search algorithm. Considering the second case study, the decision tree-based optimizer proved to be equally effective compared to evolutionary algorithms. The relative error of the simulations using the parameters estimated by these algorithms was less than 8 percent. Besides, the shape of the friction reconstructed in the second joint of the robotic manipulator through the parameters estimated by the algorithms is in accordance with the expected.

Les solutions basées sur le logiciel/matériel jouent un rôle important dans le domaine de l'automobile. Il est de plus en plus fréquent que l’implémentation de certaines fonctions jusqu’ici réalisées par des composants mécaniques, se fasse dans les véhicules d’aujourd’hui par des composants électroniques embarquant du logiciel. Cette tendance conduit à un grand nombre de fonctions implémentées comme un ensemble de composants logiciels déployés sur unités de commande électronique (ECU). Par conséquent, la quantité de code embarqué dans les automobiles est estimée à des dizaines de giga-octets et le nombre d’ECU de l’ordre de la centaine. Les pratiques actuelles de développement deviennent donc inefficaces et sont en cours d’évolution. L'objectif de cette thèse est de contribuer aux efforts actuels qui consistent à introduire l’utilisation de l'Ingénierie Dirigée par les Modèles dans la conception d’architectures automobiles basées sur le logiciel/matériel. Une première série de contributions de cette thèse porte sur la proposition de techniques pour soutenir les activités décrites dans la méthodologie automobile établie par le langage EAST-ADL2 et le standard AUTOSAR dont l’objectif principal est l'intégration de l'architecture logicielle avec la plate-forme matérielle. Bien que de nombreux travaux sur la synthèse d’architectures existent, cette thèse met en exergue les principaux défauts les empêchant de pleinement supporter la méthodologie EAST-ADL2/AUTOSAR et propose de nouvelles techniques aidant à surmonter les déficiences actuelles. Une deuxième série de contributions concerne les approches de modélisation. L'utilisation de langages de modélisation généralistes (dans le sens non spécifique à un domaine industriel donné) comme SysML et MARTE bien que bénéfique, n'a pas encore trouvé une manière d'être pleinement exploité par les constructeurs automobiles. Cela concerne en particulier la modélisation d’une spécification analysable et l'optimisation des préoccupations qui permettrait d’effectuer des analyses et optimisations à base de modèles. Ce travail définit une méthodologie et les concepts nécessaires à la construction de modèles d'analyse et d'optimisation de ces systèmes. / Hardware/software based solutions play significant role in the automotive domain. It is common that the implementation of certain functions that was done in a mechanical manner, in nowadays cars is done through the software and hardware. This tendency lead to the substantial number of functions operating as a set of software components deployed into hardware entities, i.e. Electronic Control Units (ECU). As a consequence the capacity of the overall code is estimated as tens of gigabytes and the number of ECUs reaches more than 100. Consequently the industrial state of the practice development approaches become inefficient. The objective of this thesis is to add to the current efforts trying to employ the Model Driven Engineering (MDE) in the context of the automotive SW/HW architectures design. First set of contributions relates to the guided strategies supporting the key engineering activities of the automotive methodology established by the EAST-\ADL2 language and the AUTOSAR standard. The main is the integration of the software architecture with the hardware platform. Although the amount of work on the synthesis is substantial, this thesis presents shortcomings of the existing approaches that disable them to fully support the EAST-ADL2/AUTOSAR methodology and delivers new techniques overcoming the current deficiencies. Second contribution concerns approaches for the modeling. Surprisingly the usage of general purpose modeling languages such as the SysML and MARTE although beneficial, haven’t found its way yet to be fully exploited by the automotive OEMs (Original Equipment Manufacturer). This especially relates to the modeling of the analyzable input and the optimization concerns which would enable triggering of the analysis and optimization directly from the models level. This work shows a way and defines additional concepts, necessary to construct analysis and optimization models.

In the building automation domain, there are gaps among various tasks related to design engineering. As a result created system designs must be adapted to the given requirements on system functionality, which is related to increased costs and engineering effort than planned. For this reason standards are prepared to enable a coordination among these tasks by providing guidelines and unified artifacts for the design. Moreover, a huge variety of prefabricated devices offered from different manufacturers on the market for building automation that realize building automation functions by preprogrammed software components. Current methods for design creation do not consider this variety and design solution is limited to product lines of a few manufacturers and expertise of system integrators. Correspondingly, this results in design solutions of a limited quality. Thus, a great optimization potential of the quality of design solutions and coordination of tasks related to design engineering arises. For given design requirements, the existence of a high number of devices that realize required functions leads to a combinatorial explosion of design alternatives at different price and quality levels. Finding optimal design alternatives is a hard problem to which a new solution method is proposed based on heuristical approaches. By integrating problem specific knowledge into algorithms based on heuristics, a promisingly high optimization performance is achieved. Further, optimization algorithms are conceived to consider a set of flexibly defined quality criteria specified by users and achieve system design solutions of high quality. In order to realize this idea, optimization algorithms are proposed in this thesis based on goal-oriented operations that achieve a balanced convergence and exploration behavior for a search in the design space applied in different strategies. Further, a component model is proposed that enables a seamless integration of design engineering tasks according to the related standards and application of optimization algorithms.

Water distribution networks are a fundamental part of any modern city and their daily operations constitute a significant expenditure in terms of energy and maintenance costs. Careful scheduling of pump operations may lead to significant energy savings and prevent wear and tear. By means of computer simulation, an optimal schedule of pumps can be found by an optimisation algorithm. The subject of this thesis is the study of pump scheduling as an optimisation problem. New representations of pump schedules are investigated for restricting the number of potential schedules. Recombination and mutation operators are proposed, in order to use the new representations in evolutionary algorithms. These new representations are empirically compared to traditional representations using different network instances, one of them being a large and complex network from UK. By means of the new representations, the evolutionary algorithm developed during this thesis finds new best-known solutions for both networks. Pump scheduling as the multi-objective problem of minimising energy and maintenance costs in terms of Pareto optimality is also investigated in this thesis. Two alternative surrogate measures of maintenance cost are considered: the minimisation of the number of pump switches and the maximisation of the shortest idle time. A single run of the multi-objective evolutionary algorithm obtains pump schedules with lower electrical cost and lower number of pump switches than those found in the literature. Alternatively, schedules with very long idle times may be found with slightly higher electrical cost. Finally, ant colony optimisation is also adapted to the pump scheduling problem. Both Ant System and Max-Min Ant System are tested. Max-Min Ant System, in particular, outperforms all other algorithms in the large real-world network instance and obtains competitive results in the smallest test network. Computation time is further reduced by parallel simulation of pump schedules.

La prise de décision est une partie intégrante de notre vie quotidienne où le décideur est confronté à des problèmes composés de plusieurs objectifs habituellement contradictoires. Dans ce travail, nous traitons des problèmes d'optimisation multiobjectif dans des espaces de recherche continus ou discrets. Nous avons développé plusieurs nouveaux algorithmes basés sur les métaheuristiques hybrides évolutionnaires, en particulier sur l'algorithme MOEA/D. Nous avons proposé l'algorithme HEMH qui utilise l'algorithme DM-GRASP pour construire une population initiale de solutions de bonne qualité dispersées le long de l'ensemble des solutions Pareto optimales. Les résultats expérimentaux montrent la supériorité de toutes les variantes hybrides proposées sur les algorithmes originaux MOEA/D et SPEA2. Malgré ces bons résultats, notre approche possède quelques limitations, levées dans une version améliorée de HEMH : HEMH2 et deux autres variantes HEMHde et HEMHpr. Le Adaptive Binary DE inclus dans les HEMH2 et HEMHde a de meilleures capacités d'exploration qui pallient aux capacités de recherche locale contenues dans la HEMH, HEMH2 et HEMHde. Motivés par ces résultats, nous avons proposé un nouvel algorithme baptisé HESSA pour explorer un espace continu de recherche où le processus de recherche est réalisé par différentes stratégies de recherche. Les résultats expérimentaux montrent la supériorité de HESSA à la fois sur MOEA/D et dMOPSO. Tous les algorithmes proposés ont été vérifiés, testé et comparés à certaines méthodes MOEAs. Les résultats expérimentaux montrent que toutes les propositions sont très compétitives et peuvent être considérés comme une alternative fiable / Many real-world decision making problems consist of several conflicting objectives, the solutions of which is called the Pareto-optimal set. Hybrid metaheuristics proved their efficiency in solving these problems. They tend to enhance search capabilities by incorporating different metaheuristics. Thus, we are concerned with developing new hybrid schemes by incorporating different strategies with exploiting the pros and avoiding the drawback of the original ones. First, HEMH is proposed in which the search process includes two phases DMGRASP obtains an initial set of efficient solutions in the 1st phase. Then, greedy randomized path-relinking with local search or reproduction operators explore the non-visited regions. The efficient solutions explored over the search are collected. Second, a comparative study is developed to study the hybridization of different metaheuristics with MOEA/D. The 1st proposal combines adaptive discrete differential Evolution with MOEA/D. The 2nd combines greedy path-relinking with MOEA/D. The 3rd and the 4th proposals combine both of them in MOEA/D. Third, an improved version of HEMH is presented. HEMH2 uses inverse greedy to build its initial population. Then, differential evolution and path-relink improves these solutions by investigating the non-visited regions in the search space. Also, Pareto adaptive epsilon concept controls the archiving process. Motivated by the obtained results, HESSA is proposed to solve continuous problems. It adopts a pool of search strategies, each of which has a specified success ratio. A new offspring is generated using a randomly selected one. Then, the success ratios are adapted according to the success of the generated offspring. The efficient solutions are collected to act as global guides. The proposed algorithms are verified against the state of the art MOEAs using a set of instances from literature. Results indicate that all proposals are competitive and represent viable alternatives

This thesis deals with the principles of ultrasonic perfusion analysis and methods for determining perfusion parameters. It examines Evolutionary algorithms and their ability to optimize the approximation of dilution curves from ultrasond tissue scannig. It compares the optimization performance of three evolutionary algorithms. Continuous genetic algorithm GA, algorithm SOMA and PSO. Methods are evaluated on simulated and clinical data.

This diploma thesis deals with evolutionary algorithms used for travelling salesman problem (TSP). In the first section, there are theoretical foundations of a graph theory and computational complexity theory. Next section contains a description of chosen optimization algorithms. The aim of the diploma thesis is to implement an application that solve TSP using evolutionary algorithms.

This master’s thesis proposes a method for the evolutionary design of rewriting systems. In particular, genetic algorithm will be applied to design rewriting rules for a specific variant of Lindenmayer system. The evolved rules of such grammar will be applied to generate growing sorting networks. Some distinct approaches to the rewriting process and construction of the sorting networks will be investigated. It will be shown that the evolution is able to successfully design rewriting rules for the proposed variants of rewriting processes. The results obtained exhibit abilities to successfully create partially growing sorting networks, which was evolved to grow for fewer inputs and in subsequent iterations grows up to 36 inputs.

In the context of evolutionary robotics and neurorobotics, artificial neural networks, used as controllers for animats, are examined to identify principles of neuro-control, network organization, the interaction between body and control, and other likewise properties. Before such an examination can take place, suitable neuro-controllers have to be identified. A promising and widely used technique to search for such networks are evolutionary algorithms specifically adapted for neural networks. These allow the search for neuro-controllers with various network topologies directly on physically grounded (simulated) animats. This neuro-evolution approach works well for small neuro-controllers and has lead to interesting results. However, due to the exponentially increasing search space with respect to the number of involved neurons, this approach does not scale well with larger networks. This scaling problem makes it difficult to find non-trivial, larger networks, that show interesting properties. In the context of this thesis, networks of this class are called mid-scale networks, having between 50 and 500 neurons. Searching for networks of this class involves very large search spaces, including all possible synaptic connections between the neurons, the bias terms of the neurons and (optionally) parameters of the neuron model, such as the transfer function, activation function or parameters of learning rules. In this domain, most evolutionary algorithms are not able to find suitable, non-trivial neuro-controllers in feasible time. To cope with this problem and to shift the frontier for evolvable network topologies a bit further, a novel evolutionary method has been developed in this thesis: the Interactively Constrained Neuro-Evolution method (ICONE). A way to approach the problem of increasing search spaces is the introduction of measures that reduce and restrict the search space back to a feasible domain. With ICONE, this restriction is realized with a unified, extensible and highly adaptable concept: Instead of evolving networks freely, networks are evolved within specifically designed constraint masks, that define mandatory properties of the evolving networks. These constraint masks are defined primarily using so called functional constraints, that actively modify a neural network to enforce the adherence of all required limitations and assumptions. Consequently, independently of the mutations taking place during evolution, the constraint masks repair and readjust the networks so that constraint violations are not able to evolve. Such functional constraints can be very specific and can enforce various network properties, such as symmetries, structure reuse, connectivity patterns, connectivity density heuristics, synaptic pathways, local processing assemblies, and much more. Constraint masks therefore describe a narrow, user defined subset of the parameter space -- based on domain knowledge and user experience -- that focuses the search on a smaller search space leading to a higher success rate for the evolution. Due to the involved domain knowledge, such evolutions are strongly biased towards specific classes of networks, because only networks within the defined search space can evolve. This, surely, can also be actively used to lead the evolution towards specific solution approaches, allowing the experimenter not only to search for any upcoming solution, but also to confirm assumptions about possible solutions. This makes it easier to investigate specific neuro-control principles, because the experimenter can systematically search for networks implementing the desired principles, simply by using suitable constraints to enforce them. Constraint masks in ICONE are built up by functional constraints working on so called neuro-modules. These modules are used to structure the networks, to define the scope for constraints and to simplify the reuse of (evolved) neural structures. The concept of functional, constrained neuro-modules allows a simple and flexible way to construct constraint masks and to inherit constraints when neuro-modules are reused or shared. A final cornerstone of the ICONE method is the interactive control of the evolution process, that allows the adaptation of the evolution parameters and the constraint masks to guide evolution towards promising domains and to counteract undesired developments. Due to the constraint masks, this interactive guidance is more effective than the adaptation of the evolution parameters alone, so that the identification of promising search space regions becomes easier. This thesis describes the ICONE method in detail and shows several applications of the method and the involved features. The examples demonstrate that the method can be used effectively for problems in the domain of mid-scale networks. Hereby, as effects of the constraint masks and the herewith reduced complexity of the networks, the results are -- despite their size -- often easy to comprehend, well analyzable and easy to reuse. Another benefit of constraint masks is the ability to deliberately search for very specific network configurations, which allows the effective and systematic exploration of distinct variations for an evolution experiment, simply by changing the constraint masks over the course of multiple evolution runs. The ICONE method therefore is a promising novel evolution method to tackle the problem of evolving mid-scale networks, pushing the frontier of evolvable networks a bit further. This allows for novel evolution experiments in the domain of neurorobotics and evolutionary robotics and may possibly lead to new insights into neuro-dynamical principles of animat control.