No description available.

No description available.

A customers propensity to buy a certain product is a widely researched field and is applied in multiple industries. In this thesis it is showed that using binary classification on data from Scandinavian Airlines can predict their customers propensity to book a flight within the next coming seven days. A comparison between logistic regression and support vector machine is presented and logistic regression with reduced number of variables is chosen as the final model, due to it’s simplicity and accuracy. The explanatory variables contains exclusively booking history, whilst customer demographics and search history is showed to be insignificant. / En kunds benägenhet att göra ett visst köp är ett allmänt undersökt område som applicerats i flera olika branscher. I den här studien visas det att statistiska binära klassificeringsmodeller kan användas för att prediktera Scandinavian Airlines kunders benägenhet att köpa en resa de kommande sju dagarna. En jämförelse är presenterad mellan logistisk regression och stödvektormaskin och logistisk regression med reducerat antal parametrar väljs som den slutgiltiga modellen tack vare sin enkelhet och träffsäkerhet. De förklarande variablerna är uteslutande bokningshistorik medan kundens demografi och sökdata visas vara insignifikant.

No description available.

[ES] Hoy en día, los materiales cementicios están presentes en la gran mayoría de las infraestructuras de nuestro entorno, como pueden ser el hormigón y el mortero, debido a su bajo coste y sus características mecánicas estructurales y de durabilidad. Pese a todo, estas se ven degradadas por factores externos e internos, reduciendo la viabilidad de estos con el paso del tiempo. Para inspeccionar estos materiales se han creado múltiples ensayos destructivos (ED) y ensayos no destructivos (END) que indican mediante ciertos parámetros el estado de los materiales de construcción. Dentro de los no destructivos, encontramos los ultrasonidos cuya propagación en estos materiales otorga información sobre su estado y estructura interna. Entre los múltiples ensayos ultrasónicos se encuentra la tomografía ultrasónica cuya base nace gracias a las Tomografías Computarizadas (TC): se ilumina un objeto mediante una fuente y se reciben las señales mediante los receptores. Se rota entorno al objeto bajo estudio combinando las señales mediante los algoritmos tomográficos y obteniendo una reconstrucción del objeto interno sin producirle ningún tipo de daño. No obstante, aunque para determinadas longitudes de onda podemos asumir una trayectoria de rayo recto, los ultrasonidos son ondas dispersivas que se difractan y se reflejan alejándose de esta condición de idoneidad afectando negativamente a las reconstrucciones. En esta tesis se estudia la tomografía de ultrasonidos aplicada a probetas de mortero. Para ello, previamente se realiza un estudio de los algoritmos de reconstrucción tomográfica donde se hace un recorrido por los principales algoritmos convencionales. Los transformados (FBP y DFT) cuyos resultados son excelentes en caso de que tengamos un nivel elevado de rayos y direcciones que conforman las proyecciones. Los algoritmos de redes neuronales (BPE y RBF) y métodos algebraicos (ART, CART, SART y SIRT) presentan buenos resultados en aquellas situaciones donde se tenga un bajo número de rayos y direcciones o alta presencia de ruido. Se comparan entre ellos mediante proyecciones obtenidas con señales simuladas y se obtienen los mejores resultados para el algoritmo FBP, con lo que las siguientes reconstrucciones reales se llevan a cabo con este método. La aplicación en la que se centra este trabajo consiste en la detección del frente de carbonatación en probetas cementicias. Es por ello que se diseñan diferentes casos de probetas con daño y sin daño para validar el funcionamiento de un sistema tomográfico. Se diseña y se construye el sistema hardware capaz de la toma automatizada de medidas empleando una configuración de rayos paralelos o de rayos en abanico. Además, se ha adaptado para que sea capaz de inspeccionar tanto con transductores acoplados por aire como inspeccionar el objeto en inmersión (acoplamiento por agua). Se concluye que la tomografía por inmersión ofrece una solución de compromiso entre transferencia de energía y proceso de automatización. Además se implementan dos modelos de redes neuronales entrenados mediante sinogramas simulados para posteriormente reconstruir casos reales. Todos los algoritmos y casos son evaluados tanto en calidad de reconstrucción como en prestaciones. / [CA] Avui dia, els materials cimentants són presents a la majoria de les infraestructures del nostre entorn com poden ser el formigó i el morter, donat el seu baix cost i les seues característiques mecànic estructurals i la seua durabilitat. Malgrat tot, aquestes es veuen degradades per factors externs i interns, reduint la seua viabilitat amb el pas del temps. Per inspeccionar dits materials s'han creat múltiples assajos destructius (AD) i assajos no destructius (AND) que indiquen mitjançant certs paràmetres l'estat dels materials de construcció. Dins del no destructius trobem els ultrasons, la propagació dels quals per aquests materials ens aporta informació sobre el seu estat i estructura interna. Entre els múltiples assajos ultrasònics, es troba la tomografia ultrasònica, la base de la qual neix gràcies a les Tomografies Computeritzades (TC): s'il·lumina un objecte per mitjà d'una font i es reben les senyals a través dels receptors. Es rota entorn l'objecte en estudi combinant les senyals mitjançant els algoritmes tomogràfics i obtenint una reconstrucció de l'objecte intern sense produir-li cap tipus de dany. No obstant això, i encara que per a determinades longituds d'ona podem assumir una trajectòria recta del raig, els ultrasons són ones dispersives que es difracten i reflecteixen, allunyant-se d'aquesta condició d'idoneïtat i afectant negativament les reconstruccions. En aquesta tesi s'estudia la tomografia d'ultrasons aplicada a provetes de morter. Amb aquesta finalitat, prèviament es realitza un estudi dels algoritmes de reconstrucció tomogràfica on es fa un recorregut pels principals algoritmes convencionals. Els transformats (FDB i DFT) els resultats dels quals son excel·lents en cas que tinguem un nivell elevat de raigs i direccions que conformen les projeccions. Els algoritmes de xarxes neuronals (BPE i RBF) i mètodes algebraics (ART, CART, SART i SIRT) presenten bons resultats en aquelles situacions on es tingui un baix número de raigs i direccions o una alta presència de soroll. Es comparen entre ells per mitjà de projeccions obtingudes amb senyals simulades i s'obtenen els millors resultats per a l'algoritme FBP, duent-se a terme les següents reconstruccions amb aquest mètode. L'aplicació en la que es centra aquest treball consisteix en la detecció del front de carbonatació en provetes cimentants. És per això que es dissenyen diferents casos de provetes amb desperfectes i sense desperfectes per validar el funcionament d'un sistema tomogràfic. Es dissenya i es construeix el sistema hardware capaç de la presa automatitzada de mesures emprant una configuració de raigs paral·lels o de raigs en ventall. A més, s'ha adaptat per a que sigui capaç d'inspeccionar tant amb transductors acoblats per aire com inspeccionar l'objecte en immersió (acoblament per aigua). Es conclou que la tomografia per immersió ofereix una solució de compromís entre la transferència d'energia i el procés d'automatització. A més s'implenten dos models de xarxes neuronals entrenats per mitjà de sinogrames simulats per a posteriorment reconstruir casos reals. Tots els algoritmes i casos són avaluats tant en qualitat de reconstrucció com en prestacions. / [EN] Nowadays, cementitious materials are present in the great majority of our surrounding infrastructures such as concrete and mortar, due to its low cost mechanic-structural features and its lasting. Nevertheless, this characteristics are degraded because of external and internal factors, reducing its viability over time. In order to inspect this materials, multiple destructive testing (DT) and non-destructive testing (NDT) have been created. This trials show construction materials conditions with certain parameters. In the non-destructive group, we found ultrasounds whose spreading in this materials gives us information about their condition and internal structure. Among the multiple ultrasonic tests, we can find the ultrasonic tomography which is based in the Computed Tomography Scans (CT) basis: an object is illuminated by a source and signals are received through receivers. Rotation is made around the object under study combining the signals using tomographic algorithms for the purpose of obtaining an internal object reconstruction without damaging it. However we can assume a straight beam path for certain wavelengths, ultrasound are dispersive waves that diffract and reflect, making them less suitable because they worsen the quality of reconstructions. In this thesis, the ultrasonic tomography applied to mortar specimens is studied. For that, a study of tomographic reconstruction algorithms is carried out and the main conventional algorithms are reviewed. The transforms (FBD and DFT) whose results are excellent in case we have a high level of beams and directions that make up the projections. The neuronal network algorithms (BPE and RBF) and the ones for algebraic methods (ART, CART, SART and SIRT) have good results in situations where a low number of beams and directions or high noise presence are found. A comparation between them is made using projections obtained with simulated signals and the best FBP algorithm results are extracted. The following real reconstructions are carried out with this method. The application on which this work focuses consists of the detection of the carbonation front in cementitious specimens. That is why different types of specimens with damage and without damage are designed to validate a tomographic system function. A hardware system capable of taking automated measures using a configuration of parallel beams or fan beams is designed and built. Moreover, it has been adapted to be able to inspect both with air-coupled transducers and to inspect the object while submerged (water coupled). It is concluded that immersion tomography offers a compromise solution between energy transfer and automation process. Two models of neuronal networks trained through simulated sinograms are also implemented to reconstruct real cases afterwards. All the algorithms and cases are evaluated both in reconstruction quality and in features. / Gallardo Llopis, C. (2023). Tomografía ultrasónica para la evaluación de daño por gradiente en materiales cementantes [Tesis doctoral]. Universitat Politècnica de València. https://doi.org/10.4995/Thesis/10251/194552

Processor architectures are becoming increasingly complex and hence architects have to evaluate a large design space consisting of several parameters, each with a number of potential settings. In order to assist in guiding design decisions we develop simple and accurate models of the superscalar processor design space using a detailed and validated superscalar processor simulator. Firstly, we obtain precise estimates of all significant micro-architectural parameters and their interactions by building linear regression models using simulation based experiments. We obtain good approximate models at low simulation costs using an iterative process in which Akaike’s Information Criteria is used to extract a good linear model from a small set of simulations, and limited further simulation is guided by the model using D-optimal experimental designs. The iterative process is repeated until desired error bounds are achieved. We use this procedure for model construction and show that it provides a cost effective scheme to experiment with all relevant parameters. We also obtain accurate predictors of the processors performance response across the entire design-space, by constructing radial basis function networks from sampled simulation experiments. We construct these models, by simulating at limited design points selected by latin hypercube sampling, and then deriving the radial neural networks from the results. We show that these predictors provide accurate approximations to the simulator’s performance response, and hence provide a cheap alternative to simulation while searching for optimal processor design points.

Computational tools such as finite element analysis and simulation are commonly used for system performance analysis and validation. It is often impractical to rely exclusively on the high-fidelity simulation model for design activities because of high computational costs. Mathematical models are typically constructed to approximate the simulation model to help with the design activities. Such models are referred to as “metamodel.” The process of constructing a metamodel is called “metamodeling.” Metamodeling, however, faces eminent challenges that arise from high-dimensionality of underlying problems, in addition to the high computational costs and unknown function properties (that is black-box functions) of analysis/simulation. The combination of these three challenges defines the so-called high-dimensional, computationally-expensive, and black-box (HEB) problems. Currently there is a lack of practical methods to deal with HEB problems. This dissertation, by means of surveying existing techniques, has found that the major deficiency of the current metamodeling approaches lies in the separation of the metamodeling from the properties of underlying functions. The survey has also identified two promising approaches - mapping and decomposition - for solving HEB problems. A new analytic methodology, radial basis function–high-dimensional model representation (RBF-HDMR), has been proposed to model the HEB problems. The RBF-HDMR decomposes the effects of variables or variable sets on system outputs. The RBF-HDMR, as compared with other metamodels, has three distinct advantages: 1) fundamentally reduces the number of calls to the expensive simulation in order to build a metamodel, thus breaks/alleviates exponentially-increasing computational difficulty; 2) reveals the functional form of the black-box function; and 3) discloses the intrinsic characteristics (for instance, linearity/nonlinearity) of the black-box function. The RBF-HDMR has been intensively tested with mathematical and practical problems chosen from the literature. This methodology has also successfully applied to the power transfer capability analysis of Manitoba-Ontario Electrical Interconnections with 50 variables. The test results demonstrate that the RBF-HDMR is a powerful tool to model large-scale simulation-based engineering problems. The RBF-HDMR model and its constructing approach, therefore, represent a breakthrough in modeling HEB problems and make it possible to optimize high-dimensional simulation-based design problems.

Computational tools such as finite element analysis and simulation are commonly used for system performance analysis and validation. It is often impractical to rely exclusively on the high-fidelity simulation model for design activities because of high computational costs. Mathematical models are typically constructed to approximate the simulation model to help with the design activities. Such models are referred to as “metamodel.” The process of constructing a metamodel is called “metamodeling.” Metamodeling, however, faces eminent challenges that arise from high-dimensionality of underlying problems, in addition to the high computational costs and unknown function properties (that is black-box functions) of analysis/simulation. The combination of these three challenges defines the so-called high-dimensional, computationally-expensive, and black-box (HEB) problems. Currently there is a lack of practical methods to deal with HEB problems. This dissertation, by means of surveying existing techniques, has found that the major deficiency of the current metamodeling approaches lies in the separation of the metamodeling from the properties of underlying functions. The survey has also identified two promising approaches - mapping and decomposition - for solving HEB problems. A new analytic methodology, radial basis function–high-dimensional model representation (RBF-HDMR), has been proposed to model the HEB problems. The RBF-HDMR decomposes the effects of variables or variable sets on system outputs. The RBF-HDMR, as compared with other metamodels, has three distinct advantages: 1) fundamentally reduces the number of calls to the expensive simulation in order to build a metamodel, thus breaks/alleviates exponentially-increasing computational difficulty; 2) reveals the functional form of the black-box function; and 3) discloses the intrinsic characteristics (for instance, linearity/nonlinearity) of the black-box function. The RBF-HDMR has been intensively tested with mathematical and practical problems chosen from the literature. This methodology has also successfully applied to the power transfer capability analysis of Manitoba-Ontario Electrical Interconnections with 50 variables. The test results demonstrate that the RBF-HDMR is a powerful tool to model large-scale simulation-based engineering problems. The RBF-HDMR model and its constructing approach, therefore, represent a breakthrough in modeling HEB problems and make it possible to optimize high-dimensional simulation-based design problems.

Στην παρούσα διπλωματική εργασία θα γίνει μελέτη και εφαρμογή μεθόδων επίλυσης του προβλήματος αναγνώρισης γεωμετρικών χαρακτηριστικών ανθρώπινων ερυθρών αιμοσφαιρίων από προσομοιωμένες εικόνες σκέδασης ΗΜ ακτινοβολίας ενός He-Ne laser 632.8 μm. Στο πρώτο κεφάλαιο γίνεται μια εισαγωγή στις ιδιότητες και τα χαρακτηριστικά του ερυθροκυττάρου καθώς, επίσης, παρουσιάζονται διάφορες ανωμαλίες των ερυθροκυττάρων και οι μέχρι στιγμής χρησιμοποιούμενοι τρόποι ανίχνευσής των. Στο δεύτερο κεφάλαιο της εργασίας γίνεται μια εισαγωγή στις ιδιότητες της ΗΜ ακτινοβολίας, περιγράφεται το φαινόμενο της σκέδασης και παρουσιάζεται το ευθύ πρόβλημα σκέδασης ΗΜ ακτινοβολίας ανθρώπινων ερυθροκυττάρων. Το τρίτο κεφάλαιο αποτελείται από δύο μέρη. Στο πρώτο μέρος γίνεται εκτενής ανάλυση της θεωρίας των τεχνητών νευρωνικών δικτύων και περιγράφονται τα νευρωνικά δίκτυα ακτινικών συναρτήσεων RBF. Στη συνέχεια, αναφέρονται οι μέθοδοι εξαγωγής παραμέτρων και, πιο συγκεκριμένα, δίνεται το θεωρητικό και μαθηματικό υπόβαθρο των μεθόδων που χρησιμοποιήθηκαν οι οποίες είναι ο αλογόριθμος Singular Value Decomposition (SVD), o Angular Radial μετασχηματισμός (ART) και φίλτρα Gabor. Στο δεύτερο μέρος περιγράφεται η επίλυση του αντίστροφου προβλήματος σκέδασης. Παρουσιάζεται η μεθοδολογία της διαδικασίας επίλυσης όπου εφαρμόστηκαν ο αλογόριθμος συμπίεσης εικόνας SVD, o περιγραφέας σχήματος ART και ο περιγραφέας υφής με φίλτρα Gabor για την εύρεση των γεωμετρικών χαρακτηριστικών και νευρωνικό δίκτυο ακτινικών συναρτήσεων RBF για την ταξινόμηση των ερυθροκυττάρων. Στο τέταρτο και τελευταίο κεφάλαιο γίνεται δοκιμή και αξιολόγηση της μεθόδου και συνοψίζονται τα αποτελέσματα και τα συμπεράσματα που εξήχθησαν κατά τη διάρκεια της εκπόνησης αυτής της διπλωματικής. / In this thesis we study and implement methods of estimating the geometrical features of the human red blood cell from a set of simulated light scattering images produced by a He-Ne laser beam at 632.8 μm. Ιn first chapter an introduction to the properties and the characteristics of red blood cells are presented. Furthermore, we describe various abnormalities of erythrocytes and the until now used ways of detection. In second chapter the properties of electromagnetic radiation and the light scattering problem of EM radiation from human erythrocytes are presented. The third chapter consists of two parts. In first part we analyse the theory of neural networks and we describe the radial basis function neural network. Then, we describe the theoritical and mathematical background of the methods that we use for feature extraction which are Singular Value Decomposition (SVD), Angular Radial Transform and Gabor filters. In second part the solution of the inverse problem of light scattering is described. We present the methodology of the solution process in which we implement a Singular Value Decomposition approach, a shape descriptor with Angular Radial Transform and a homogenous texture descriptor which uses Gabor filters for the estimation of the geometrical characteristics and a RBF neural network for the classification of the erythrocytes. In the forth and last chapter the described methods are evaluated and we summarise the experimental results and conclusions that were extracted from this thesis.

Dans le domaine d’optimisation de forme de structures, la réduction des coûts et l’amélioration des produits sont des défis permanents à relever. Pour ce faire, le procédé de mise en forme doit être optimisé. Optimiser le procédé revient alors à résoudre un problème d’optimisation. Généralement ce problème est un problème d’optimisation multicritère très coûteux en terme de temps de calcul, où on cherche à minimiser plusieurs fonctions coût en présence d’un certain nombre de contraintes. Pour résoudre ce type de problème, on a développé un algorithme robuste, efficace et fiable. Cet algorithme, consiste à coupler un algorithme de capture de front de Pareto (NBI ou NNCM) avec un métamodèle (RBF), c’est-à-dire des approximations des résultats des simulations coûteuses. D’après l’ensemble des résultats obtenus par cette approche, il est intéressant de souligner que la capture de front de Pareto génère un ensemble des solutions non dominées. Pour savoir lesquelles choisir, le cas échéant, il est nécessaire de faire appel à des algorithmes de sélection, comme par exemple Nash et Kalai-Smorodinsky. Ces deux approches, issues de la théorie des jeux, ont été utilisées pour notre travail. L’ensemble des algorithmes sont validés sur deux cas industriels proposés par notre partenaire industriel. Le premier concerne un modèle 2D du fond de la canette (elasto-plasticité) et le second est un modèle 3D de la traverse (élasticité linéaire). Les résultats obtenus confirment l’efficacité de nos algorithmes développés. / One of the current challenges in the domain of the multiobjective shape optimization is to reduce the calculation time required by conventional methods. The high computational cost is due to the high number of simulation or function calls required by these methods. Recently, several studies have been led to overcome this problem by integratinga metamodel in the overall optimization loop. In this thesis, we perform a coupling between the Normal Boundary Intersection -NBI- algorithm and The Normalized Normal constraint Method -NNCM- algorithm with Radial Basis Function -RBF- metamodel in order to have asimple tool with a reasonable calculation time to solve multicriteria optimization problems. First, we apply our approach to academic test cases. Then, we validate our method against two industrial cases, namely, shape optimization of the bottom of a can undergoing nonlinear elasto-plastic deformation and an optimization of an automotive twist beam. Then, in order to select solutions among the Pareto efficient ones, we use the same surrogate approach to implement a method to compute Nash and Kalai-Smorodinsky equilibria.