CNPq - Conselho Nacional de Desenvolvimento Científico e Tecnológico / Esse trabalho propõe a modelagem e o controle neuro-fuzzy aplicados na estabilidade estática de um veículo de duas rodas do tipo pêndulo invertido, utilizando como método de otimização a evolução diferencial. Durante a fase de modelagem, determinam-se as incertezas relacionadas aos parâmetros e também à resposta do modelo neuro-fuzzy. Verifica-se que este é capaz de se ajustar satisfatoriamente aos dados extraídos experimentalmente do veículo. Na determinação do controlador neuro-fuzzy, testam-se três estratégias de ajuste de parâmetros, sendo duas delas propostas neste texto, e os resultados são comparados entre si e aos obtidos através de controladores clássicos, e verifica-se experimentalmente e por meio de testes estatísticos que as abordagens propostas apresentam grande capacidade de adaptação às restrições impostas à planta, garantindo a estabilidade estática e a eficiência energética do sistema. / This work proposes the neuro-fuzzy modeling and control applied to the static stability of a two-wheeled inverted pendulum vehicle, using differential evolution as optimization technique. During the modeling phase, the uncertainties related to the parameters and also to the neuro-fuzzy model response are determined. It is possible to verify that the neuro-fuzzy system is capable of satisfactorily adjusts to the data experimentally extracted from the vehicle. In the determination of the neuro-fuzzy controller, three strategies of parameter adjustment are tested, two of them being proposed in this text, and the results are compared between them and those obtained through classical controllers, and it is verified experimentally and through tests that the proposed approaches present a great capacity to adapt to the constraints imposed on the plant, guaranteeing the static stability and the energy efficiency of the system. / Dissertação (Mestrado)

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / Um Estudo para a solução numérica do modelo de difusão com retenção, proposta por Bevilacqua et al. (2011), é apresentado, bem como uma formulação implícita para o problema inverso para a estimativa dos parâmetros envolvidos na formulação matemática do modelo. Através de um estudo minucioso da análise de sensibilidade e do cálculo do coeficiente de correlação de Pearson, são identificadas as chances de se obter sucesso na solução do problema inverso através do método determinístico de Levenberg-Marquardt e dos métodos estocásticos Algoritmo de Colisão de Partículas (Particle Collision Algorithm - PCA) e Evolução Diferencial (Differential Evolution - DE). São apresentados os resultados obtidos através destes três métodos de otimização para três casos de conjunto de parâmetros. Foi observada uma forte correlação entre dois destes três parâmetros, o que dificultou a estimativa simultânea dos mesmos. Porém, foi obtido sucesso nas estimativas individuais de cada parâmetro. Foram obtidos bons resultados para os fatores que multiplicam os termos diferenciais da equação que modela o fenômeno de difusão com retenção. / A Study for the numerical solution of the diffusion model with retention, proposed by Bevilacqua et al.(2011), using the finite difference method is presented, as well as an implicit formulation for the inverse problem to estimate the parameters involved in the formulation of the mathematical model. Through of a thorougth study of sensitivity analysis and calculating the Pearson correlation coefficient, are identified the chances of success in solving the inverse problem using the deterministic method of Levenberg-Marquardt and stochastic methods Particle Collision Algorithm - PCA and Differential Evolution - DE. Presents the results obtained from these three methods of optimization for three cases of parameter set. We observed a strong correlation between two of these three parameters, making it difficult to estimate simultaneously the same. However, success was obtained in the individual estimates for each parameter. Good results were obtained for the factors that increase the terms of the differential equation that models the phenomenon of diffusion with retention.

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / Sistemas estruturais em suas variadas aplicações incluindo-se veículos espaciais, automóveis e estruturas de engenharia civil tais como prédios, pontes e plataformas off-shore, acumulam dano durante suas vidas úteis. Em muitas situações, tal dano pode não ser visualmente observado. Do ponto de vista da segurança e da performance da estrutura, é desejável monitorar esta possível ocorrência, localizá-la e quantificá-la. Métodos de identificação de sistemas, que em geral, são classificados numa categoria de Técnicas de Avaliação Não-Destrutivas, podem ser utilizados para esta finalidade. Usando dados experimentais tais como frequências naturais, modos de vibração e deslocamentos estáticos, e um modelo analítico estrutural, parâmetros da estrutura podem ser identificados. As propriedades estruturais do modelo analítico são modificadas de modo a minimizar a diferença entre os dados obtidos por aquele modelo e a resposta medida. Isto pode ser definido como um problema inverso onde os parâmetros da estrutura são identificados. O problema inverso, descrito acima, foi resolvido usando métodos globais de otimização devido à provável presença de inúmeros mínimos locais e a não convexidade do espaço de projeto. Neste trabalho o método da Evolução Diferencial (Differential Evolution, DE) foi utilizado como ferramenta principal de otimização. Trata-se de uma meta-heurística inspirada numa população de soluções sucessivamente atualizada por operações aritméticas como mutações, recombinações e critérios de seleção dos melhores indivíduos até que um critério de convergência seja alcançado. O método da Evolução Diferencial foi desenvolvido como uma heurística para minimizar funções não diferenciáveis e foi aplicado a estruturas planas de treliças com diferentes níveis de danos. / Structural systems in a variety of applications including aerospace vehicles, automobiles and civil engineering structures such as tall buildings, bridges and offshore platforms, accumulate damage during their service life. In several situations, such damage may not be visually observable. From the standpoint of both safety and performance, it is desirable to monitor the occurrence, location and extent of such damage.System identification methods, which may be classified in a general category of nondestructive evaluation techniques, can be employed for this purpose. Using experimental data, such as eigenmodes, eigenvectors and static displacements, and an analytical structural model, parameters of the structures can be identified. The approach used in the present work is one where the structural properties of the analytical model are varied to minimize the difference between the analytically predicted and empirically measured response. This is an inverse problem where the structural parameters are identified. In this work a reduced number of vibration modes were used as the measured response. For the damage assessment problem a close analytical model of the structural system is available and the model of the damaged structure will be identified. Damage will be represented by a reduction in the elastic stiffness properties of the structure.The problem described above was solved using global methods of optimization due to the fact that depending on the number of variables or the location of damage the resulting design space is nonconvex presenting several local minima. In the present work, the Differential Evolution Optimization Technique (DE) was used. It is a metaheuristic inspired by a population of solutions that is successively updated by arithmetic operations such as mutation and recombination, until convergence. The approach was applied to simple truss structures with different levels of damage.

Submitted by João Arthur Martins (joao.arthur@ufpe.br) on 2015-03-12T17:38:41Z No. of bitstreams: 2 Dissertaçao Tiago de Lima.pdf: 1469834 bytes, checksum: 95a0326778b3d0f98bd35a7449d8b92f (MD5) license_rdf: 1232 bytes, checksum: 66e71c371cc565284e70f40736c94386 (MD5) / Approved for entry into archive by Daniella Sodre (daniella.sodre@ufpe.br) on 2015-03-13T14:23:38Z (GMT) No. of bitstreams: 2 Dissertaçao Tiago de Lima.pdf: 1469834 bytes, checksum: 95a0326778b3d0f98bd35a7449d8b92f (MD5) license_rdf: 1232 bytes, checksum: 66e71c371cc565284e70f40736c94386 (MD5) / Made available in DSpace on 2015-03-13T14:23:38Z (GMT). No. of bitstreams: 2 Dissertaçao Tiago de Lima.pdf: 1469834 bytes, checksum: 95a0326778b3d0f98bd35a7449d8b92f (MD5) license_rdf: 1232 bytes, checksum: 66e71c371cc565284e70f40736c94386 (MD5) Previous issue date: 2013-02-26 / In this dissertation, we present a methodology that aims the automatic construction of multi-classifiers systems based on the combination of selection and fusion. The presented method initially finds an optimum number of clusters for training data set and subsequently determines an ensemble for each cluster found. For model evaluation, the testing data set are submitted to clustering techniques and the nearest cluster to data input will emit a supervised response through its associated ensemble. Self-organizing maps were used in the clustering phase and multilayer perceptrons were used in the classification phase. Adaptive differential evolution has been used in this work in order to optimize the parameters and performance of the different techniques used in the classification and clustering phases. The proposed method, called SFJADE - Selection and Fusion (SF) via Adaptive Differential Evolution (JADE), has been tested on data compression of signals generated by artificial nose sensors and well-known classification problems, including cancer, card, diabetes, glass, heart, horse, soybean and thyroid. The experimental results have shown that the SFJADE method has a better performance than some literature methods while significantly outperforming most of the methods commonly used to construct Multi-Classifier Systems. / Nesta dissertação, nós apresentamos uma metodologia que almeja a construção automática de sistemas de múltiplos classificadores baseados em uma combinação de seleção e fusão. O método apresentado inicialmente encontra um número ótimo de grupos a partir do conjunto de treinamento e subsequentemente determina um comitê para cada grupo encontrado. Para avaliação do modelo, os dados de teste são submetidos à técnica de agrupamento e o grupo mais próximo do dado de entrada irá emitir uma resposta supervisionada por meio de seu comitê associado. Mapas Auto Organizáveis foi usado na fase de agrupamento e Perceptrons de múltiplas camadas na fase de classificação. Evolução Diferencial Adaptativa foi utilizada neste trabalho a fim de otimizar os parâmetros e desempenho das diferentes técnicas utilizadas nas fases de classificação e agrupamento de dados. O método proposto, chamado SFJADE – Selection and Fusion (SF) via Adaptive Differential Evolution (JADE), foi testado em dados gerados para sensores de um nariz artificial e problemas de referência em classificação de padrões, que são: cancer, card, diabetes, glass, heart, heartc e horse. Os resultados experimentais mostraram que SFJADE possui um melhor desempenho que alguns métodos da literatura, além de superar a maioria dos métodos geralmente usados para a construção de sistemas de múltiplos classificadores.

The thesis studies selected evolution systems to use in planning of high intensity focused ultrasound surgeries. Considered algorithms are statistically analyzed and compared by appropriate criteria to find the one that adds the most value to the potential real world medical problems.

The dissertation thesis deals with Evolution optimization of control algorithms. The first part of the thesis describes the principles and partial methods of evolution optimization methods especially those used in two-level transplant evolution method. Later the grammatical evolution method is described, which modified algorithm became impulse for creation of transplant evolution method. The transplant evolution method and its two-level modification are new evolutionary algorithms proposed in this work, which were used for optimization of structure and parameters of general controllers control algorithms. The transplant evolution algorithm and its extended two-level modification are described in detail in next chapters. The proper settings of evolutionary algorithms are important for minimization the time of optimization and for finds results approaching the global optimum. For proper setting the parameters of differential evolution was created meta-evolution algorithm that is described in chapter named meta-evolution. The basic concepts of control, chosen methods of system identification and controller parameters settings are described in next part. This part describes algorithms of digital controllers and some specific methods uses in digital control. The demonstrations of control algorithm optimizations of various types of controllers are showed in experimental part. The optimized algorithms of general controllers are compared with various types of PSD controllers which were set by various algebraic methods or differential evolution for various models of systems. In the conclusion of this work is stated a recommendation for further development of evolutionary optimization of controllers are focusing on parallel and distributed computing.

No description available.

The electricity grids have become a key player in the society. An increased usage of electricity is both a result from the more electrified society, but also as a main solver in reaching the climate goals by reducing emissions. This thesis work explores some of the new features for the electricity grid from integration of electrification from renewable energy resources (RES) and from strategies for energy optimization related to the loads and specifically from thermal heat pumps. These strategies lie in the field of demand response, which takes advantage of the flexibility of loads in terms of willingness to switch or decrease their consumption at a particular moment of the day. This research proposes a three-step framework to harness the flexibility of Thermo-Statically controlled loads (TCLs) based on a simulated grey-box building model that uses historical outside temperature and prices data and learns the thermal parameters such as Thermal Resistance, Thermal Capacitance, but also price responsiveness (pth ) through a Differential Evolution (DE) based optimization algorithm. The price responsiveness is used to provide further insight into the flexibility of the loads and is employed in the last step to propose a price policy estimation algorithm also based on DE that minimizes the gap between supply and demand while preserving supplier and customer profitability. The proposed approach has proven to be accurate for a large number of parameters but also effective with reduced training data (prediction errors around 2.5% on the power average and standard deviation), as well as to be successful in providing a Day-Ahead Real-Time Price policy that maximizes supplier and customer utility. The price policy provides a lower total price for the customer compared with a tariff without demand response (reduction up to 53.63 %), reduces the gap between RES-based energy sources and heating demand, and respects grid technical constraints. / Elnätet har blivit en nyckelaktör i samhället. En ökad elanvändning är både ett resultat av det mer elektrifierade samhället, men också som en huvudlösare för att nå klimatmålen genom att minska utsläpp. Detta examensarbete undersöker några av de nya funktionerna för elnätet från integration av elektrifiering från förnybara energiresurser (RES) och från strategier för energioptimering relaterade till ellaster och specifikt från termiska värmepumpar. Dessa strategier ligger inom området för efterfrågerespons, som drar fördel av lasternas flexibilitet när det gäller viljan att byta eller minska sin konsumtion vid en viss tidpunkt på dagen. Den här forskningen föreslår ett ramverk i tre steg för att utnyttja flexibiliteten hos termostatiskt kontrollerade laster (TCL) baserat på en simulerad gråbox-byggnadsmodell som använder historisk utomhustemperatur och prisdata och lär sig de termiska parametrarna som termisk motstånd, termisk kapacitans , men också priskänslighet (pth) genom en Differential Evolution (DE) baserad optimeringsalgoritm. Priskänsligheten används för att ge ytterligare insikt om lasternas flexibilitet och används i det sista steget för att föreslå en prispolitisk uppskattningsalgoritm också baserad på DE som minimerar klyftan mellan utbud och efterfrågan samtidigt som leverantörs- och kundlönsamheten bevaras. Det föreslagna tillvägagångssättet har visat sig vara korrekt för ett stort antal parametrar men också effektivt med reducerad träningsdata (förutsägelsefel runt 2,5% på effektmedelvärde och standardavvikelse), samt vara framgångsrik i att tillhandahålla en Day-Ahead Real -Tidsprispolicy som maximerar leverantörs- och kundnyttan. Prispolicyn ger ett lägre totalpris för kunden jämfört med en tariff utan efterfrågerespons (reduktion upp till 53, 63 %), minskar gapet mellan RES-baserade energikällor och värmebehov samt följer nättekniska begränsningar.

No / Evolutionary membrane computing is an important research direction of membrane computing that aims to explore the complex interactions between membrane computing and evolutionary computation. These disciplines are receiving increasing attention. In this paper, an overview of the evolutionary membrane computing state-of-the-art and new results on two established topics in well defined scopes (membrane-inspired evolutionary algorithms and automated design of membrane computing models) are presented. We survey their theoretical developments and applications, sketch the differences between them, and compare the advantages and limitations. (C) 2014 Elsevier Inc. All rights reserved.

Η παρούσα διατριβή ασχολείται με τη μελέτη, την ανάπτυξη και τη θεμελίωση νέων μεθόδων Υπολογιστικής Νοημοσύνης και Ευφυούς Βελτιστοποίησης. Συνοπτικά οργανώνεται στα ακόλουθα τρία μέρη: Αρχικά παρουσιάζεται το πεδίο της Υπολογιστικής Νοημοσύνης και πραγματοποιείται μία σύντομη αναφορά στους τρεις κύριους κλάδους της, τον Εξελικτικό Υπολογισμό, τα Τεχνητά Νευρωνικά Δίκτυα και τα Ασαφή Συστήματα. Το επόμενο μέρος αφιερώνεται στην παρουσίαση νέων, καινοτόμων οικογενειών των αλγορίθμων Βελτιστοποίησης Σμήνους Σωματιδίων (ΒΣΣ) και των Διαφοροεξελικτικών Αλγόριθμων (ΔΕΑ), για την επίλυση αριθμητικών προβλημάτων βελτιστοποίησης χωρίς περιορισμούς, έχοντας είτε ένα, είτε πολλαπλούς ολικούς βελτιστοποιητές. Οι αλγόριθμοι ΒΣΣ και ΔΕΑ αποτελούν τις βασικές μεθοδολογίες της παρούσας διατριβής. Όλες οι οικογένειες μεθόδων που προτείνονται, βασίζονται σε παρατηρήσεις των κοινών δομικών χαρακτηριστικών των ΒΣΣ και ΔΕΑ, ενώ η κάθε προτεινόμενη οικογένεια τις αξιοποιεί με διαφορετικό τρόπο, δημιουργώντας νέες, αποδοτικές μεθόδους με αρκετά ενδιαφέρουσες ιδιότητες και δυναμική. Η παρουσίαση του ερευνητικού έργου της διατριβής ολοκληρώνεται με το τρίτο μέρος στο οποίο περιλαμβάνεται μελέτη και ανάπτυξη μεθόδων ολικής βελτιστοποίησης για την εκπαίδευση Τεχνητών Νευρωνικών Δικτύων Υψηλής Τάξης, σε σειριακά και παράλληλα ή / και κατανεμημένα υπολογιστικά συστήματα. Η διδακτορική διατριβή ολοκληρώνεται με βασικά συμπεράσματα και τη συνεισφορά της. / The main subject of the thesis at hand revolves mainly around the development and foundations of new methods of computational intelligence and intelligent optimization. The thesis is organized into the following three parts: Firstly, we briefly present an overview of the field of Computational Intelligence, by describing its main categories, the Evolutionary Computation, the Artificial Neural Networks and the Fuzzy Systems. In the second part, we provide a detailed description of the newly developed families of algorithms for solving unconstrained numerical optimization problems in continues spaces with at least one global optimum. The proposed families are based on two well-known and widely used algorithms, namely the Particle Swarm Optimization (PSO) and the Differential Evolution (DE) algorithm. Both DE and PSO are the basic components for almost all methodologies proposed in the thesis. The proposed methodologies are based on common observations of the dynamics, the structural and the spacial characteristics of DE and PSO algorithms. Four novel families are presented in this part which exploit the aforementioned characteristics of the DE and the PSO algorithms. The proposed methodologies are efficient methods with quite interesting properties and dynamics. The presentation and description of our research contribution ends with the third and last part of the thesis, which includes the study and the development of novel global optimization methodologies for training Higher order Artificial Neural Networks in serial and parallel / distributed computational environments. The thesis ends with a brief summary, conclusions and discussion of the contribution of this thesis.