The research presented in this thesis examines the area of control systems for robots or animats (animal-like robots). Existing systems have problems in that they require a great deal of manual design or are limited to performing jobs of a single type. For these reasons, a better solution is desired. The system studied here is an Artificial Nervous System (ANS) which is biologically inspired; it is arranged as a hierarchy of layers containing modules operating in parallel. The ANS model has been developed to be flexible, scalable, extensible and modular. The ANS can be implemented using any suitable technology, for many different environments. The implementation focused on the two lowest layers (the reflex and action layers) of the ANS, which are concerned with control and rhythmic movement. Both layers were realised as Artificial Neural Networks (ANN) which were created using Evolutionary Algorithms (EAs). The task of the reflex layer was to control the position of an actuator (such as linear actuators or D.C. motors). The action layer performed the task of Central Pattern Generators (CPG), which produce rhythmic patterns of activity. In particular, different biped and quadruped gait patterns were created. An original neural model was specifically developed for assisting in the creation of these time-based patterns. It is shown in the thesis that Artificial Reflexes and CPGs can be configured successfully using this technique. The Artificial Reflexes were better at generalising across different actuators, without changes, than traditional controllers. Gaits such as pace, trot, gallop and pronk were successfully created using the CPGs. Experiments were conducted to determine whether modularity in the networks had an impact. It has been demonstrated that the degree of modularization in the network influences its evolvability, with more modular networks evolving more efficiently.

The marketing literature so far only considers attraction models with strict functional forms. Greater exibility can be achieved by the neural net based approach introduced which assesses brands' attraction values by means of a perceptron with one hidden layer. Using log-ratio transformed market shares as dependent variables stochastic gradient descent followed by a quasi-Newton method estimates parameters. For store-level data the neural net model performs better and implies a price response qualitatively different from the well-known MNL attraction model. Price elasticities of these competing models also lead to specific managerial implications. (author's abstract) / Series: Working Papers SFB "Adaptive Information Systems and Modelling in Economics and Management Science"

The contributions of this thesis are an implicit modelling method for the coupled constitutive relations and an energy-based method for material modelling. The two developed methods utilise implicit models to represent material constitutive relations without the requirement of physical parameters. The first method is developed to model coupled constitutive relations using state-space representation with neural networks. State-space representation is employed to express the desired relations in a compact fashion while simultaneously providing the capability of modelling rate- and/or path-dependent behaviour. The employment of neural networks with the generalised state-space representation results in a single implicit model that can be adapted for a broad range of constitutive behaviours. The performance and applicability of the method are highlighted through the applications for various constitutive behaviour of piezoelectric materials, including the effects of hysteresis and cyclic degradation. An energy-based method is subsequently developed for implicit constitutive modelling by utilising the energy principle on a deformed continuum. Two formulations of the proposed method are developed for the modelling of materials with varying nature in directional properties. The first formulation is based on an implicit strain energy density function, represented by a neural network with strain invariants as input, to derive the desired stress-strain relations. The second formulation consists of the derivation of an energy-based performance function for training a neural network that represents the stress-strain relations. The requirement of deriving stress is eliminated in both formulations and this facilitates the use of advanced experimental setup, such as multi-axial load tests or non-standard specimens, to produce the most information for constitutive modelling from a single experiment. A series of numerical studies -- including validation problems and practical cases with actual experimental setup -- have been conducted, the results of which demonstrate the applicability and effectiveness of the proposed method for constitutive modelling on a continuum basis.

Στην παρούσα διπλωματική εργασία μελετάμε την ικανότητα των τεχνητών νευρωνικών δικτύων στη σύνθεση μουσικής και την αναγνώριση μουσικού συνθέτη. Συγκεκριμένα, στο πρώτο κεφάλαιο κάνουμε μία εισαγωγή στα τεχνητά νευρωνικά δίκτυα και ειδικά σε αυτά που χρησιμοποιούνται στα επόμενα κεφάλαια. Γίνεται αναφορά στα βασικά είδη των ΤΝΔ που υπάρχουν, εμπρόσθιας τροφοδότησης και αναδραστικά και περιγράφονται οι αλγόριθμοι εκπαίδευσής τους. Εξηγούμε την ικανότητα των αναδραστικών νευρωνικών δικτύων να έχουν δυναμική μνήμη, σε αντίθεση με αυτά που είναι εμπρόσθιας τροφοδότησης, πράγμα που τα καθιστά ικανά στην πρόβλεψη χρονοσειρών. Αυτή η ικανότητα των αναδραστικών δικτύων σε συνδυασμό με το γεγονός ότι ένα μουσικό κομμάτι μπορεί να χαρακτηριστεί σαν μία αλληλουχία γεγονότων χρονικής συνοχής (χρονοσειρά) δημιούργησε ένα ερευνητικό ρεύμα προς την κατεύθυνση της σύνθεσης μουσικής με τη χρήση ανδραστικών τεχνητών νευρωνικών δικτύων. Στο δεύτερο κεφάλαιο κάνουμε μία αναφορά στην αλγοριθμική σύνθεση μουσικής, ιδιαίτερα με χρήση πινάκων μετάβασης. Έπειτα ακολουθεί η περιγραφή του CONCERT, ενός αναδραστικού νευρωνικού δικτύου που κατασκευάστηκε για να συνθέτει μουσική με πρόβλεψη νότας προς νότα. Αναλύουμε επίσης την μοντελοποίηση των μουσικών αντικειμένων για την επεξεργασία και αναπαράστασή τους από το CONCERT η οποία βασίζεται σε ψυχοακουστικούς περιορισμούς αντίληψης των μουσικών αντικειμένων από τους ανθρώπους. Εξηγούμε τον τρόπο που εκπαιδεύεται το CONCERT έτσι ώστε να έχει όσο το δυνατόν μεγαλύτερη μνήμη και περιγράφουμε τις επιδόσεις του σε διάφορες δοκιμές που έγιναν, από την εκμάθηση μίας διατονικής κλίμακας μέχρι ενός κομματιού του J. S. Bach. Παρατηρώντας την ικανότητα του CONCERT να αντιλαμβάνεται τοπικές δομές (μοτίβα φράσεις) μα όχι καθολικές (μέρη του μουσικού κομματιού) αναφερόμαστε στην τεχνική της περιορισμένης περιγραφής που αποτελεί μια προσπάθεια για εκπαίδευση του δικτύου έτσι ώστε να αντιλαμβάνεται το μουσικό κομμάτι σε μία μεγαλύτερη κλίμακα. Στο τέλος του δευτέρου κεφαλαίου εξετάζουμε τη συνολική επίδοση του CONCERT και αναλύουμε τις κατευθύνσεις προς τις οποίες θα μπορούσαμε να κινηθούμε για τη βελτίωση των αποτελεσμάτων. Στο τρίτο κεφάλαιο αναφερόμαστε στην αναγνώριση του συνθέτη ενός μουσικού κομματιού με τη χρήση τεχνητών νευρωνικών δικτύων πάνω στην παρτιτούρα του κομματιού αυτού. Αρχικά γίνεται μία συζήτηση γύρω από το ποια στοιχεία της παρτιτούρας θεωρούμε σημαντικά, ποια από αυτά μπορούμε και ποια έχει νόημα να μοντελοποιήσουμε έτσι ώστε ένα ΤΝΔ να μπορεί να κάνει πρόβλεψη. Αναλύονται οι τεχνικές λεπτομέρειες των στοιχείων που χρειαζόμαστε για τη μοντελοποίηση μιας παρτιτούρας στον υπολογιστή και στη συνέχεια αναφερόμαστε στα δύο πειράματα που ελέγχουν την ορθότητα και αποτελεσματικότητα της παραπάνω προσέγγισης. Το ποια κομμάτια χρησιμοποιήθηκαν και από ποιους συνθέτες δε θα μπορούσε να είναι τυχαίο καθώς πρέπει να ικανοποιούνται διάφορες συνθήκες στατιστικής ομοιομορφίας έτσι ώστε η απάντηση του ΤΝΔ να είναι όσο το δυνατόν πιο αμερόληπτη. Αυτές οι συνθήκες, καθώς και οι κίνδυνοι που υπάρχουν σε πιθανή παράληψή τους εξηγούνται πριν τα πειράματα. Το πρώτο πείραμα πραγματεύεται την αναγνώριση συνθέτη ενός κομματιού που συντέθηκε από τον Chopin ή όχι (δηλαδή από τους Beethoven ή Mozart) ενώ στο δεύτερο οι εμπλεκόμενοι συνθέτες είναι οι Bach και Handel. Δοκιμάζονται διάφορες αρχιτεκτονικές ΤΝΔ και μετρούμε τη μέση και τη βέλτιστη επίδοσή τους. Τέλος συζητάμε τα αποτελέσματα των δύο πειραμάτων καθώς και τροποποιήσεις είτε του ΤΝΔ είτε της μοντελοποίησης που διαλέξαμε για την αναπαράσταση της παρτιτούρας στον υπολογιστή έτσι ώστε να έχουμε καλύτερα αποτελέσματα. / In this work we study the capability of artificial neural networks for composing music and musical composer recognition. To this end, in the first chapter the neural networks are introduced, especially the forms of those that are used later on. A reference is being made to the basic forms of neural networks, feedforward (FNN) and recursive (RNN), and their training algorithms. We explain the ability of the RNNs to have dynamic memory, in contrast to FNNs, which makes them suitable for predicting time series. This ability combined to the fact that a musical piece can be considered as a time series has urged researchers to explore music composition through RNNs. In the second chapter algorithmic music composition is being described, especially with the use of Markov chains. Then we describe CONCERT, a RNN constructed for composing music with note by note prediction. We also analyze the representation of musical objects which is based in how humans perceive them. CONCERT is trained with different musical patterns (from diatonic scales to Bach pieces) and its composing ability is being discussed. The fact that CONCERT lacks in capturing the global structure of a piece is not changed with the use of reduced description, which is thoroughly described. The second chapter concludes with thoughts on how a RNN could capture the global structure of a piece. The third chapter is devoted to composer recognition with the use of FNNs. Firstly we discuss which elements of a score are useful and which of them we can represent such that a FNN can identify a composer. The techniques that we use for the computer modeling of the problem and the manipulation of the pieces are thoroughly described. Two experiments are presented, in the first one the FNN is called to recognize Chopin from Mozart and Beethoven and in the second Bach from Handel. Finally a discussion is made on the results of the above experiments and how we could optimize them.

The human olfactory system can classify new odors in a dynamic environment with varying odor complexity and concentration, while simultaneously reducing the influence of stable background odors. Replication of this capability has remained an active area of research over the past 3 decades and has great potential to advance medical diagnostics, environmental monitoring and industrial monitoring, among others. New methods for rapid dynamic temporal evaluation of chemical sensor arrays for the monitoring of analytes is explored in this work. One such method is high and low bandpass filtering of changing sensor responses; this is applied to reduce the effects of background noise and sensor drift over time. Processed sensor array responses, coupled with principal component analysis (PCA), will be used to develop a novel approach to classify odors in the presence of changing sensor responses associated with evolving odor concentrations. These methods will enable the removal of noise and drift, as well as facilitating the normalization to decouple classification patterns from intensity; lastly, PCA and artificial neural networks (ANNs) will be used to demonstrate the capability of this approach to function under dynamic conditions, where concentration is changing temporally. / February 2016

The operation of large industrial scale combustion systems, such as furnaces and boilers is increasingly dictated by emission legislation and requirements for improved efficiency. However, it can be exceedingly difficult and time consuming to gather the information required to improve original designs. Mathematical modelling techniques have led to the development of sophisticated furnace representations that are capable of representing combustion parameters. Whilst such data is ideal for design purposes, the current power of computing systems tends to generate simulation times that are too great to embed the models into online control strategies. The work presented in this thesis offers the possibility of replacing such mathematical models with suitably trained Artificial Neural Networks (ANNs) since they can compute the same outputs at a fraction of the model's speed, suggesting they could provide an ideal alternative in online control strategies. Furthermore, artificial neural networks have the ability to approximate and extrapolate making them extremely robust when encountering conditions not met previously. In addition to improving operational procedures, another approach to increasing furnace system efficiency is to minimise the waste heat energy produced during the combustion process. One very successful method involves the implementation of a heat exchanger system in the exiting gas flue stream, since this is predominantly the main source of heat loss. It can be exceptionally difficult to determine which heat exchanger is best suited for a particular application and it can prove an even more arduous task to control it effectively. Furthermore, there are many factors that alter the performance characteristics of a heat exchanger throughout the duration of its operational life, such as fouling or unexpected systematic faults. This thesis investigates the modelling of an experimental heat exchanger system via artificial neural networks with a view to aiding the design and selection process. Moreover, the work presented offers a means to control heat exchangers subject to varying operating conditions more effectively, thus promoting savings in both waste energy and time.

Master of Science / Department of Computer Science / William H. Hsu / Cryptocurrencies are digital currencies that have garnered significant investor attention in the financial markets. The aim of this project is to predict the daily price, particularly the daily high and closing price, of the cryptocurrency Bitcoin. This plays a vital role in making trading decisions. There exist various factors which affect the price of Bitcoin, thereby making price prediction a complex and technically challenging task. To perform prediction, we trained temporal neural networks such as time-delay neural networks (TDNN) and recurrent neural networks (RNN) on historical time series – that is, past prices of Bitcoin over several years. Features such as the opening price, highest price, lowest price, closing price, and volume of a currency over several preceding quarters were taken into consideration so as to predict the highest and closing price of the next day. We designed and implemented TDNNs and RNNs using the NeuroSolutions artificial neural network (ANN) development environment to build predictive models and evaluated them by computing various measures such as the MSE (mean square error), NMSE (normalized mean square error), and r (Pearson’s correlation coefficient) on a continuation of the training data from each time series, held out for validation.

Orientador: Maria de Lourdes Bueno Trindade Galo / Resumo: A utilização de dados de Sensoriamento Remoto em conjunto com técnicas de processamento digital e análise de imagens tem possibilitado o desenvolvimento de estudos integrados, com vistas ao monitoramento dos recursos naturais. Uma maneira de representar esses dados é através de mapas temáticos, obtidos por métodos de classificação multiespectral. Para a classificação de dados de Sensoriamento Remoto, a utilização de Redes Neurais Artifíciais tem se apresentado como uma alternativa vantajosa em relação aos classificadores baseados em conceitos estatísticos, uma vez que nenhuma hipótese prévia sobre a distribuição dos dados a serem classificados é exigida. Assim, esse trabalho teve como objetivo detectar a ocorrência e mapear a dispersão espacial de plantas aquáticas emersas em cinco reservatórios ao longo do rio Tietê-SP (Barra Bonita, Bariri, Ibitinga, Promissão e Nova Avanhandava) através da classificação por Redes Neurais Artifíciais...(Resumo completo, clicar acesso eletrônico abaixo) / Mestre

Conselho Nacional de Desenvolvimento CientÃfico e TecnolÃgico / Estimar o recalque em estacas à um problema muito complexo, incerto e ainda nÃo totalmente compreendido, devido Ãs muitas incertezas associadas aos fatores que afetam a magnitude desta deformaÃÃo. As RNA sÃo ferramentas que funcionam analogamente ao cÃrebro humano, e sua unidade principal, o neurÃnio artificial, trabalha de maneira semelhante ao neurÃnio biolÃgico. Esta ferramenta alternativa vem sendo aplicada com sucesso em muitos problemas de engenharia geotÃcnica, podendo, portanto ser utilizadas como uma ferramentas alternativas para avaliar recalques em estacas isoladas. Nessa pesquisa as RNA utilizadas foram do tipo perceptron de mÃltiplas camadas, empregando um treinamento supervisionado utilizando o algoritmo de retropropagaÃÃo do erro. O modelo desenvolvido relaciona o recalque em estacas isoladas com as propriedades geomÃtricas das estacas (diÃmetro e comprimento), a estratigrafia e as caracterÃsticas de compacidade, ou consistÃncia dos solos por meio dos resultados obtidos nos ensaios SPT, e a carga atuante, obtidas em provas de carga realizadas em estacas hÃlice contÃnua, cravada metÃlica e escavada. O conjunto de aprendizagem foi composto por 1947 exemplos de entrada e saÃda. Com auxilio do programa QNET2000 foram treinadas e validadas vÃrias arquiteturas de redes neurais. ApÃs comparar o desempenho da curva carga x recalque elaborada com os recalques estimados pelo modelo proposto com a curva carga x recalque resultante da prova de carga estÃtica e com a curva carga x recalque gerada pelo emprego do programa comercial baseado em elementos finitos tridimensionais PLAXIS 3D Foundation, constatou-se que as RNA foram capazes de entender o comportamento das fundaÃÃes profundas do tipo estacas hÃlice contÃnua, escavada e cravada metÃlica, possibilitando dentre outras coisas, a definiÃÃo das cargas de trabalho e cargas limites nas estacas. / Predicting the settlement in deep foundation is a very complex, uncertain and not yet fully understood, due to the many uncertainties associated with factors that affect the magnitude of this deformation. Artificial Neural Network (ANN) is a tool that works similarly to the human brain, its main unit, the artificial neuron, works in a similar way to the biological neuron. This alternative tool has been successfully applied in many geotechnical engineering problems and can therefore be used as an alternative tool to evaluate the behavior of settlement in isolated piles. In this paper, the ANN used were the multilayer perceptron type, employing a supervised training that uses the error back propagation algorithm. The model developed relates settlement in isolated piles with the type and the geometrical properties of the piles (diameter and length), the stratigraphy and characteristics of compactness or consistency of soils by means of the SPT tests results, and the load applied, obtained in static pile load tests performed in continuous helix, steel driven and excavated pile types. The data set used to model consisted of 1.947 samples of input and output. QNET 2000 was the program used to assist the training and validation of various architectures of neural networks. The architecture formed by 10 nodes in the input layer, 28 neurons distributed in 4 intermediate layers and one neuron in the output layer, corresponding to the measured discharge for cutting (A10: 14:8:4:2:1) was the one that showed the best performance, with the correlation coefficient between the estimated settlements and settlements measured during the validation phase of 0.94, such value can be considered satisfactory when considering the prediction of a complex phenomenon. After comparing the performance of the applied load x settlement estimated by model proposed curve with the applied load x settlement measured in static pile load test curve and the applied load x settlement estimated by an elasto-plastic model thru numerical simulation, it was found that the ANN were able to understand the behavior of deep foundations of continuous helix, steel driven and excavated piles type, allowing among other things, the definition of workloads and load limits at the pile.

Orientadores: Edson Bim, Fernando José Von Zuben / Dissertação (mestrado) - Universidade Estadual de Campinas, Faculdade de Engenharia Elétrica e de Computação / Made available in DSpace on 2018-08-18T01:48:41Z (GMT). No. of bitstreams: 1 Marchi_RodrigoAndreolide_M.pdf: 5160130 bytes, checksum: 8e043b48f5cd0e86cc078b3adc27f74b (MD5) Previous issue date: 2011 / Resumo: Neste trabalho é apresentada a estratégia de Controle Direto de Potência para o Gerador de Indução Duplamente Alimentado utilizando um controlador Perceptron de Múltiplas Camadas. O controlador tem a função de gerar os sinais das componentes de eixo direto e quadratura da tensão do rotor, sem a necessidade de controladores de corrente. A estratégia de controle apresentada permite operar o conversor de potência, conectado aos terminais do rotor, com frequência de chaveamento constante. A rede neural foi treinada off-line, a partir de um algoritmo de otimização de segunda ordem baseado no gradiente conjugado estendido, utilizando um conjunto de amostras obtido por meio da simulação digital de uma máquina de rotor bobinado de potência igual a 2 MW. Resultados de simulação digital com os dados dessa máquina, operando no modo gerador e com dupla alimentação, são apresentados para vários valores de potência ativa e reativa, e para velocidades fixas e variáveis, compreendidas na faixa de -15% a +15% da velocidade síncrona. Com o controlador implementado por uma rede neural artificial e treinada para uma máquina de 2 MW, testes de simulação digital e experimentais para uma máquina de 2,2 kW, operando na velocidade subsíncrona, são apresentados para validar a proposta / Abstract: This work presents a direct power control strategy for the doubly fed induction generator using a controller artificial neural networks, more specifically a multilayer perceptron. The controller has the role of generating the direct and quadrature-axis component signals of the rotor voltage, without the need of current controllers. The proposed control strategy allows to operate the converter, connected to the rotor terminals, with a fixed switching frequency. The multilayer perceptron was subject to an off-line training procedure using a second order algorithm based on an extend version of the conjugate gradient algorithm, using a set of samples produced by a 2 MW machine's digital simulation. Results of digital simulation for this machine are presented for several values of active and reactive power, with the generator operating on fixed and variable speed, in the range of -15% and +15% of the synchronous speed, considering the parameters of 2 MW machine. With the artificial neural network controller designed for this machine, digital simulation tests and experimental tests for a 2,2 kW machine, operating in a sub-synchronous speed, arc presented to validate the proposal / Mestrado / Energia Eletrica / Mestre em Engenharia Elétrica