<p>Artificial intelligence is a broad research area and there are many different reasons why it is interesting to study artificial intelligence. One of the main reasons is to understand how information might be represented in the human brain. The Recursive Auto Associative Memory (RAAM) is a connectionist architecture that with some success has been used for that purpose since it develops compact distributed representations for compositional structures.</p><p>A lot of extensions to the RAAM architecture have been developed through the years in order to improve the performance of RAAM; Bi coded RAAM (B-RAAM) is one of those extensions. In this work a modified B-RAAM architecture is tested and compared to RAAM regarding: Training speed, ability to learn with smaller internal representations and generalization ability. The internal representations of the two network models are also analyzed and compared. This dissertation also includes a discussion of some theoretical aspects of B-RAAM.</p><p>It is found here that the training speed for B-RAAM is considerably lower than RAAM, on the other hand, RAAM learns better with smaller internal representations and is better at generalize than B-RAAM. It is also shown that the extracted internal representation of RAAM reveals more structural information than it does for B-RAAM. This has been shown by hieratically cluster the internal representation and analyse the tree structure. In addition to this a discussion is added about the justifiability to label B-RAAM as an extension to RAAM.</p>

<p>Efter en omfattande SAP implementering finns det oftast ännu mer datamaterial i en</p><p>organisation. Moderna ERP-system som SAP skapar å ena sidan en stor transparens, men å</p><p>andra sidan förändras processerna och det är viktigt för en organisation att kunna analysera</p><p>affärsinformation snabbt och enkelt och att kunna reagera på förändringar på marknaden.</p><p>Detta arbete ska ge en överblick över olika verktyg för ett BI-system i samband med</p><p>analysen av omsättningen och faktureringsprocessen vid ett aluminiumvalsverk.</p><p>Processerna har förändrats vid aluminiumvalsverket p g a SAP implementeringen och det</p><p>är en akut och viktig fråga att bygga upp ett informationssystem som baserar sig på SAP</p><p>information. Därför valde Man att implementera BI-verktyget Cubeware samt att</p><p>programmera några SAP Query för att kontrollera faktureringsprocessen. Men första steget</p><p>vid starten av denna informationsprocess av BI-verktyg var ett Excel-program som på ett</p><p>enkelt sätt kan analysera data som kommer från faktureringsprocessen. Det visade sig dock</p><p>snabbt att det fanns problem i denna informationsprocess. Å ena sidan fanns det felaktig</p><p>information inom SAP systemet, därför utvecklades SAP Query; och å andra sidan har</p><p>Excel-filen blivit allt för stor och nästan oanvändbar för att kunna göra snabba analyser.</p><p>Därför implementerades Cubeware i samband med databasen MS SQL Server 2005.</p><p>I mitt arbete ska jag beskriva denna utvecklingsprocess för informationshanteringen och</p><p>samtidigt analysera olika felorsaker och beskriva informationskvalitetsutvecklingen.</p>

Thesis (S.T.M.)--Concordia Seminary, 1999. / Includes bibliographical references (leaves 138-144). Includes additional title p. and some bibliographical references in Chinese.

In recent decades, climate change has become one of the most crucial challenges for humanity. Climate change has a direct correlation with global warming, caused mainly by the green house gas emissions (GHG). The Environmental Protection Agency in the U.S. (EPA) attributes carbon dioxide to account for approximately 82\% of the GHG emissions. Unfortunately, the energy sector is the main producer of carbon dioxide, with China and the U.S. as the highest emitters. Therefore, there is a strong (positive) correlation between energy production, global warming, and climate change. Stringent carbon emissions reduction targets have been established in order to reduce the impacts of GHG. Achieving these emissions reduction goals will require implementation of policies like as cap-and-trade and carbon taxes, together with transformation of the electricity grid into a smarter system with high green energy penetration. However, the consideration of policies solely in view of carbon emissions reduction may adversely impact other market outcomes such as electricity prices and consumption. In this dissertation, a two-layer mathematical-statistical framework is presented, that serves to develop carbon policies to reduce emissions level while minimizing the negative impacts on other market outcomes. The bottom layer of the two layer model comprises a bi-level optimization problem. The top layer comprises a statistical model and a Pareto analysis. Two related but different problems are studied under this methodology. The first problem looks into the design of cap-and-trade policies for deregulated electricity markets that satisfy the interest of different market constituents. Via the second problem, it is demonstrated how the framework can be used to obtain levels of carbon emissions reduction while minimizing the negative impact on electricity demand and maximizing green penetration from microgrids. In the aforementioned studies, forecasts for electricity prices and production cost are considered. This, this dissertation also presents anew forecast model that can be easily integrated in the two-layer framework. It is demonstrated in this dissertation that the proposed framework can be utilized by policy-makers, power companies, consumers, and market regulators in developing emissions policy decisions, bidding strategies, market regulations, and electricity dispatch strategies.

A bi-polar world emerged at the end of World War II. The United States and the Soviet Union were the world's superpowers and tensions between them spiraled consequently bringing about the Cold War. United States foreign policy during the Cold War revolved around containment policy. The Middle East during the Cold War was a region that the bi-polar world's superpowers wanted to influence, and protect. The United States during the Cold War warned the Soviet Union through presidential doctrines that it would fight to keep the Middle East from communism, and the Soviet Union's influence. The bi-polar international power structure did not allow the United States the ability to pick its battles. The power structure that constrained the Cold War forced the United States to react to the Soviet Union, and it forced foreign policy makers to always consider the Soviet Union's response to its policies. United States foreign policy in the Middle East during the Cold War threatened wit h military methods to solve local and regional instabilities. However, the United States was constrained by the bi-polar world thus, it was cautious of committing military troops in the region permanently and upsetting the region's delicate balance of power. United States foreign policy toward the Middle East has changed between 1981 and 2006. This change is in the direction of greater use of military methods to resolve what various administrations have perceived to be local and regional instability. This change in policy is partly attributable to changes in the United States power position in the world. A United States foreign policy shift in the Middle East occurred due to a change in the distribution of political power within the interstate system. This change has had the following result: the United States is no longer constrained by the bi-polar international power structure that characterized the Cold War period. The collapse of the Soviet Union created the uni-polar internationa l power structure. United States foreign policy is now capable of deploying the military to resolve local and regional instabilities in the Middle East, and that deployment has tended to become increasingly permanent in nature.

Over the past twenty-five years, condensed matter physics has been developing materials with novel electronic characteristics for a wide range of future applications. Two research directions have shown particular promise: topological insulators, and high temperature copper based superconductors (cuprates). Topological insulators are a newly discovered class of materials that can be manipulated for spintronic or quantum computing devices. However there is a poor spectroscopic understanding of the current topological insulators and emerging topological insulator candidates. In cuprate superconductors, the challenge lies in raising the superconducting transition temperature to temperatures accessible in non-laboratory settings. This effort has been hampered by a poor understanding of the superconducting mechanism and its relationship with a mysterious pseudogap phase. In this thesis, I will describe experiments conducted on topological insulators and cuprate superconductors using scanning tunneling microscopy and spectroscopy, which provide nanoscale spectroscopic information in these materials. / Physics

In this thesis, a capacitor semi-active hybrid energy storage system for electric vehicle is proposed. A DC-to-DC bi-directional converter is required to couple the supercapacitor to the system DC bus. Through literature reviews, it was decided that a dual-bridge resonant converter with HF transformer isolation is best suited for the hybrid energy storage application. First, a dual-bridge series resonant converter with capacitive output filter is proposed. Modified gating scheme is applied to the converter instead of the 50% duty cycle gating scheme. Comparing to the 50% duty cycle gating scheme where only four switches work in ZVS, The modified gating scheme allows all eight switches working in ZVS at design point with high load level, and seven switches working in ZVS under other conditions. Next, a dual-bridge LCL-type series resonant converter with capacitive output filter is proposed. Similarly, the modified gating scheme is applied to the converter. This converter shows further improvement in ZVS ability. Operating principles, design examples, simulation results and experimental results of the two newly proposed converters are also presented. In the last part of the thesis, a capacitor semi-active hybrid energy storage system is built to test if the proposed converters are compatible to the system. The dual-bridge LCL-type series resonant converter is placed in parallel to the supercapacitor. The simulation and experimental results of the hybrid energy storage system match closely to the theoretical waveforms. / Graduate

Η μελέτη συνήθων διαφορικών εξισώσεων συχνά χρησιμοποιεί μεθόδους γνωστές από την κλασική Μηχανική. Η πιο γνωστή από αυτές ϕέρει το όνομα του εμπνευστή της, του Ιρλανδού Sir William Rowan Hamilton (1805 - 1865), κι αποτελεί μία μαθηματικά πλήρη ϑεωρία για τα λεγόμενα συστήματα Hamilton. Πρόσφατα, όμως, δομές τύπου Hamilton άρχισαν να μελετώνται και σε συστήματα μερικών διαφορικών εξισώσεων, συγκεκριμένα εξισώσεων εξέλιξης. Σκοπός της παρούσας εργασίας είναι η ανάπτυξη της ϑεωρίας Hamilton για τα συστήματα αυτά και ιδιαίτερα για τις περιπτώσεις εκείνες που εμφανίζουν ολοκληρωσιμότητα. Η γραμμή που ϑα ακολουθήσουμε έχει ως κύριο οδηγό τις συμμετρίες των διαφορικών εξισώσεων, ένα πολύ χρήσιμο εργαλείο για την επίλυση οποιασδήποτε διαφορικής εξίσωσης, που πρώτος ανέδειξε ο Νορβηγός Marius Sophus Lie (1842 - 1899). Στο πρώτο κεφάλαιο λοιπόν γίνεται μία εισαγωγή στην ϑεωρία των (γεωμετρικών) συμμετριών, ενώ επίσης παρουσιάζονται τρόποι επίλυσης και γενικότερα αντιμετώπισης ξεχωριστά συνήθων και μερικών διαφορικών εξισώσεων με την χρήση των ομάδων συμμετρίας τους. Το δεύτερο κεφάλαιο ϕιλοδοξεί να αναδείξει την αντιστοιχία μεταξύ των συμμετριών ενός συστήματος διαφορικών εξισώσεων και των νόμων διατήρησης στους οποίους υπακούει το ϕυσικό σύστημα που περιγράφουν. Αυτό είναι και το περιεχόμενο του ϑεωρήματος που διατύπωσε η Γερμανίδα Amalie Emmy Noether (1882 - 1935), το οποίο ισχύει και στην ειδική περίπτωση των συστημάτων Hamilton. Το πρώτο, λοιπόν, ϐήμα προς αυτήν την κατεύθυνση είναι η επέκταση της έννοιας της συμμετρίας στις λεγόμενες γενικευμένες συμμετρίες, με ιδιαίτερη έμφαση στις εξισώσεις εξέλιξης. Το δεύτερο είναι ουσιαστικά μια μικρή εισαγωγή στην ϑεωρία μεταβολών, απαραίτητη όμως και για τα επόμενα κεφάλαια. Την γνωστή ϑεωρία Hamilton για πεπερασμένα συστήματα, συστήματα δηλαδή συνήθων διαϕορικών εξισώσεων πραγματεύεται το τρίτο κεφάλαιο. Σκοπός του κεφαλαίου αυτού δεν είναι η πλήρης περιγραφή της ϑεωρίας, αλλά η διατύπωση των εννοιών εκείνων που μπορούν να γενικευτούν και στην περίπτωση των απειροδιάστατων συστημάτων. Για τον λόγο αυτό έχει προτιμηθεί η κάπως πιο αφηρημένη και σίγουρα όχι τόσο συνηθισμένη περιγραφή στο πλαίσιο της γεωμετρίας Poisson. Αντιμετωπίζοντας τις συμπλεκτικές δομές, οι οποίες επικρατούν στην ϐιβλιογραφία, ως μια υποπερίπτωση των γενικότερων δομών Poisson, έχουμε ουσιαστικά αποφύγει τελείως την χρήση διαφορικών μορφών, στρέφοντας περισσότερο την προσοχή στις ομάδες συμμετρίας Hamilton, μία έννοια-κλειδί για την ολοκληρωσιμότητα των συστημάτων αυτών. Στο τέταρτο κεφάλαιο παρουσιάζουμε το κεντρικό ϑέμα αυτής της εργασίας, δηλαδή τη ϑεωρία Hamilton για απειροδιάστατα συστήματα εξισώσεων εξέλιξης, και ειδικότερα την ολοκληρωσιμότητα τους. Τα ϐασικά μας εργαλεία είναι αυτά που παρουσιάστηκαν νωρίτερα, δηλαδή οι (γενικευμένες) συμμετρίες και οι νόμοι διατήρησης από την μια, και τα διανυσματικά πεδία Hamilton από την άλλη που μας επιτρέπουν την μεταξύ τους αντιστοιχία. Με ϐάση αυτά τα εργαλεία ϐλέπουμε πως η μελέτη πολλών μερικών διαφορικών εξισώσεων ϑυμίζει εκείνων των κλασικών συστημάτων Hamilton της Μηχανικής. Στην παραπάνω αντιστοιχία ϐασίζεται και η έννοια των δι-Χαμιλτονικών συστημάτων, την οποία μελετάμε στο πέμπτο κεφάλαιο. Μέσα από το παράδειγμα της εξίσωσης Korteweg-de Vries αναδεικνύονται τα πλεονεκτήματα της εύρεσης δύο διαφορετικών, ανεξάρτητων εκφράσεων Hamilton, που οδηγούν στην κατασκευή άπειρων συμμετριών ή ακόμα και νόμων διατήρησης. Η διπλή αυτή δομή Hamilton των απειροδιάστατων συστημάτων συνδέεται, όπως ϑα δούμε, με την ολοκληρωσιμότητα είτε με την έννοια του Liouville, είτε με διάφορα άλλα κριτήρια. Γνωστά παραδείγματα παραθέτονται, πέρα από την KdV, όπως η εξίσωση Schroedinger, η modified KdV, κι άλλες μη γραμμικές κυματικές εξισώσεις. Στο έκτο και τελευταίο κεφάλαιο παρουσιάζουμε την περίπτωση, όπου ένα σύστημα επιδέχεται πολλαπλή δομή Hamilton. Τέτοιου είδους συστήματα μας επιτρέπουν να δούμε προϋπάρχουσες έννοιες από την ϑεωρία Hamilton, αλλά κι όχι μόνο, κάτω από μία άλλη σκοπιά. Γι΄ αυτό κι έχουν απασχολήσει την σύγχρονη ϐιβλιογραφία, πάνω στην οποία κάνουμε μία σύντομη επισκόπηση, τόσο στο κομμάτι εκείνο που ασχολείται με τις πρόσφατες εξελίξεις της ϑεωρίας Hamilton, όσο και με την μελέτη γενικότερα της ολοκληρωσιμότητας των μερικών διαφορικών εξισώσεων. / The study of ordinary differential equations has often borrowed well known methods from Classical Mechanics. The most popular one is due to Sir William Rowan Hamilton (1805-1865), which has become a complete mathematical theory for the so-called Hamiltonian systems. Recently, Hamiltonian structures have been developed for systems of partial differential equations, particularly evolution equations. The purpose of this master thesis is to present the Hamiltonian theory for this type of systems, and especially for integrable equations. Our description is based on Symmetries, a useful tool for solving any differential equation, first discovered by Marius Sophus Lie (1842-1899). Thus, an introduction to his theory of point or geometrical symmetries is given in the first chapter, along with some applications, such as integration of ordinary differential equations and group-invariant solutions of partial differential equations. In the second chapter we discuss the connection between the symmetries of a system of differential equations and the conservation laws of the physical problem that they describe. That is the content of Noether’s theorem, which also holds in the particular case of Hamiltonian systems. The first step towards this direction is the generalization of the basic symmetry concept, and the second one is a small introduction to variational problems, also necessary for the next chapters. The well known Hamilton’s theory for finite systems is presented in the third chapter. We do not wish to describe the whole theory in full detail but only focus on these points that will be needed to handle the infinite-dimensional case. Therefore, we introduce the general notion of a Poisson structure, instead of the more familiar symplectic one. Avoiding the use of differential forms almost entirely, we concentrate on the Hamiltonian symmetries and their key role in the reduction theory of these systems. In Chapter 4 lies the heart of the subject, the Hamiltonian approach to a system of evolution equations. We start off by drawing an analogy between first order ordinary differential equations and evolution equations, and then we establish the fundamental concepts of the Hamiltonian franework, i.e. the Poisson bracket and Hamiltonian vector fields. Through another version of Noether’s theorem, we are able to explore, once again, the correspondence between (generalized) symmetries and conservation laws. Thus, we see that the study of several partial differential equations is in some way very close to the one of classical mechanical Hamiltonian systems. Evolution equations possessing, not just one, but two Hamiltonian structures, called bi-Hamiltonian systems, are discussed in the next chapter. The advantages of finding two different, independent Hamiltonian expressions are pointed out through the example of the Korteweg-de Vries equation. We show that such systems have an infinite number of symmetries and, subject to a mild compatibility condition, they also have an infinite number of conservation laws. Therefore they are completely integrable in Liouville’s sense. Several examples are presented, besides the KdV equation, such as the nonlinear Schroedinger, the modified KdV and other nonlinear wave equations. The final chapter is devoted to some of the recent publications, regarding multi-Hamiltonian evolution equations. This type of systems puts the classical Hamiltonian theory of ordinary differential equations in a new perspective and at the same time allows us to draw some connections with other integrability criteria used in the field of partial differential equations.

No description available.

The rise of the Environmentalism in every sector of the Industry has lead the aircraft and engine manufacturing companies to develop new generations of more environmentally friendly engines. The companies, encouraged to this purpose, are in a constant research for new manufacturing and production techniques, in order to improve their products, from the environmental point of view, by gaining efficiency in the manufacturing techniques and reduce the fuel consumption and emissions in-flight. Having in mind this scenario, the sponsor of this Project is interested in understanding how changing the materials of the blades, titanium alloys currently, for other lighter materials, such as composites, is going to have an effect in the overall gas turbine efficiency. In the particular case of this Project, it will be studied the influence of the Stiffness and coating Thickness of those non-metallic materials suitable to be employed as coatings on gas turbine fan blades, from the icing point of view. The work procedure will be based on a study of Linear Elastic Fracture Mechanics of bi-material junctions and will extrapolate the general problem to the ice-coatings case, by getting experimental data from tests carried out in an Icing Tunnel. It will be observed that the coating Stiffness has an influence on the Adhesion Level of ice to less stiff materials, if compared with the Adhesion Level of ice to metals. Besides, it will be described how a 0.5 millimetres thin polymeric coating placed over a metallic substrate is enough to reduce the Adhesion Level of ice, hiding any effect that the underneath materials might have on the Adhesion Level.