Para aplicações de extração de petróleo de poços em terra foi desenvolvido pelo Laboratório de Eletromagnetismo Aplicado (LMAG) da Escola Politécnica da Universidade de São Paulo um protótipo de motor de indução linear tubular (MILT), onde o movimento axial do secundário do motor aciona diretamente a bomba de extração situada no fundo do poço. Numa segunda etapa, foi prevista a substituição dos mancais mecânicos por dois mancais magnéticos ativos (AMBs), que permitem melhor movimentação e praticamente nenhum atrito, reduzindo o desgaste causado por impurezas contidas no petróleo extraído, e consequentemente os esforços de manutenção. Todavia, o protótipo atual possui apenas um mancal magnético, e o outro é mecânico. O presente trabalho apresenta a instalação do segundo mancal magnético ao protótipo do MILT, e propõe realizar o controle do sistema de mancais magnéticos para o MILT, tendo em vista um problema multivariável, onde as posições do secundário do motor em relação aos dois mancais são correlacionadas, bem como as ações de controle sobre os mesmos. O trabalho faz uma revisão do sistema atual com um AMB, abordando sua concepção física, modelagem e o controlador, e tal controlador é replicado para o segundo AMB. Um novo modelo é apresentado, considerando o comportamento multivariável dos dois AMBs, e um sistema de controle robusto multivariável é projetado, através da técnica LQG/LTR. Resultados de simulação do novo controlador são analisados e comparados com os resultados experimentais do controlador atual aplicado aos dois AMBs, e apresentam-se as conclusões. / For onshore oil extraction applications, a tubular linear induction motor (TLIM) prototype was developed by Applied Electromagnetism Laboratory (LMAG) of Escola Politécnica da Universidade de São Paulo, on which the axial movement of the motor secondary drives the suction pump, placed in the down hole of the oil well. In a second step, it was planned to replace the mechanical bearings by two Active Magnetic Bearings (AMB), in order to have better movement and practically no friction, reducing damages caused by impurities in the oil, thus reducing maintenance effort. Nevertheless, the actual prototype has only one AMB, being the other one a mechanical bearing. This paper presents the installation of the second AMB onto TLIM prototype, and a proposal to implement the control algorithm for the TLIM magnetic bearing system, considering now a multivariable problem, where the position of the motor secondary for both AMB are related, as well as control efforts. The present work review the actual system with only one AMB, approaching its physical construction, mathematical model and applied control system; and this control system is applied to the second AMB. A new model is presented, considering the AMB system multivariable behavior, and a multivariable robust control system is then designed, using LQG/LTR approach. Simulation results for the new controller are analyzed and compared to experimental results from the actual controller applied to both AMB, and some conclusions are presented.

Made available in DSpace on 2016-08-17T14:52:52Z (GMT). No. of bitstreams: 1 Carlos Cesar Teixeira Ferreira.pdf: 819618 bytes, checksum: d77aa555d25af8f110fac9381a4b6c4d (MD5) Previous issue date: 2004-03-25 / This work presents a proposal for Eigenstructure Assignment in dynamic stochastic multivariable systems, using LQG/LTR Robust Controllers and Evolutionary Computation. It shows the importance and influence of the Eigenstructure Assignment on the systems dynamic response. It solves the Eigenstructure Assignment problem of the control problem through Linear Quadratic Regulator Design and Genetic Algorithm, developed to perform the search of the state and control weighting matrices in order to assign the Eigenstructure. The Eigenstructure Assignment problem of stochastic state estimation is formulated and solved by using Kalman Filter Design and Genetic Algorithm, now developed to search the state disturbance and measurement noise covariances matrices in order to find an adequate estimator. The Eigenstructure Assignment problem for the stochastic multivariable control with observer is formulated as LQG/LTR Robust Control problem with Loop Transfer Recovery at the Input. This proposal for Eigenstructure Assignment using LQG/LTR Controllers is examined on dynamical system model that representing an aircraft. / Apresenta-se neste trabalho um método para Alocação de Auto-estrutura em sistemas dinâmicos estocásticos multivariáveis, utilizando-se o Projeto de Controladores Robustos LQG/LTR e Computação Evolutiva. Mostra-se a importância e influência da Alocação de Auto-estrutura na resposta de sistemas dinâmicos. Resolve-se o problema da Alocação de Auto-estrutura do problema de controle através do Projeto do Regulador Linear Quadrático e Algoritmo Genético, desenvolvido para realizar a busca das matrizes de ponderação do estado e do controle de forma a alocar a Auto-estrutura. Formula-se e resolve-se o problema de Alocação de Auto-estrutura do estimador de estado estocástico, utilizando-se Filtro de Kalman e Algoritmo Genético, agora desenvolvido para buscar as matrizes de covariâncias da perturbação no estado e do ruído de medida, de forma a encontrar um estimador adequado. Propõe-se e implementa-se a solução do problema de Alocação de Auto-estrutura para o projeto de controle multivariável com observador estocástico, formulado como um problema de Controle Robusto LQG/LTR com Recuperação da Malha de Transferência na Entrada. Verifica-se esta proposta de Alocação de Auto-estrutura via Controladores LQG/LTR em um modelo de um sistema dinâmico que representa uma aeronave.

Short interspersed nuclear elements (SINEs) are small non-autonomous and heterogeneous retrotransposons, widespread in animals and plants and usually differentially propagated in related species resulting in genome-specific copy numbers. Within the monocots, the Poaceae (sweet grasses) is the largest and economically most important plant family. The distribution of 24 Poaceae SINE (PoaS) families, five of which showing a subfamily structure, was analyzed in five important cereals (Oryza sativa, Triticum aestivum, Hordeum vulgare, Sorghum bicolor, Zea mays), the energy crop Panicum virgatum and the model grass Brachypodium distachyon. The comparative investigation of SINE abundance and sequence diversity within Poaceae species provides insights into their species‐specific diversification and amplification. The PoaS families and subfamilies fall into two length and structural categories: simple SINEs of up to 180 bp and dimeric SINEs larger than 240 bp. Of 24 PoaS families, 20 are structurally related across species, in particular either in their 5′ or 3′ regions. Hence, reshuffling between SINEs, likely caused by nested insertions of full-lengh and truncated copies, is an important evolutionary mechanism of SINE formation. Most striking, the recently evolved homodimeric SINE family PoaS‐XIV occurs exclusively in wheat (T. aestivum) and consists of two tandemly arranged PoaS‐X.1 copies. Exemplary for deciduous tree species, the evolutionary history of SINE populations was examined in six Salicaceae genomes (Populus deltoides, Populus euphratica, Populus tremula, Populus tremuloides, Populus trichocarpa, Salix purpurea). Four of eleven Salicaceae SINE (SaliS) families exhibit a subfamily organization. The SaliS families consist of two groups, differing in their phylogenetic distribution pattern, sequence similarity and 3’ end structure. These groups probably emerged at different evolutionary periods of time: during the ‘salicoid duplication’ (~ 65 million years ago) in the Salix-Populus progenitor, and during the separation of the genus Salix (~ 45 - 65 million years ago), respectively. Similar to the PoaS families, the majority of the 20 SaliS families and subfamilies share regions of sequence similarity, providing evidence for SINE emergence by reshuffling. Furthermore, they also contain an evolutionarily young dimeric SINE family (SaliS-V), amplified only in two poplar genomes. The special feature of the Salicaceae SINEs is the contrast of the conservation of 5’ start motifs across species and SINE families compared to the high variability of 3’ ends within the SINE families, differing in sequence and length, presumably resulting from mutations in the poly(A) tail as a possible route for SINE elongation. Periods of increased transpositional activity promote the dissemination of novel 3’ ends. Thereby, evolutionarily older motifs are displaced leading to various 3’ end subpopulations within the SaliS families. Opposed to the PoaS families with a largely equal ratio of poly(A) to poly(T) tail SINEs, the SaliS families are exclusively terminated by adenine stretches. Among retrotransposon-based markers, SINEs are highly suitable for the development of molecular markers due to their unidirectional insertion and random distribution mainly in euchromatic genome regions, together with an easy and fast detection of the heterogeneous SINE families. As a prerequisite for the development of SINE-derived inter-SINE amplified polymorphism (ISAP) markers, 13 novel Theaceae SINE families (TheaS-I - TheaS-VII, TheaS-VIII.1 and TheaS-VIII.2, TheaS-IX - TheaS-XIII) were identified in the angiosperm tree species Camellia japonica. Moreover, six Pinaceae SINE families (PinS-I.1 and PinS-I.2, PinS-II – PinS-VI) were detected in the gymnosperm species Larix decidua. Compared to the SaliS and PoaS families, structural relationships are less frequent within the TheaS families and absent in the PinS families. The ISAP analysis revealed the genetic identity of Europe’s oldest historical camellia (C. japonica) trees indicating their vegetative propagation from the same ancestor specimen, which was probably the first living camellia on European ground introduced to England within the 18th century. Historical sources locate the native origin of this ancestral camellia specimen either in the Chinese province Yunnan or at the Japanese Gotō Islands. Comparative ISAPs showed no accordance to the Gotō camellia sample pool and appropriate Chinese reference samples were not available. However, the initial experiments demonstrated the potential of ISAP to resolve variations among natural populations. The ISAP application on angiosperm trees also concerned fast growing Populus clones grown in short rotation coppice plantations for energy production. The species-specific P. tremula ISAP primers might also be applied for the discrimination of hybrid poplar clones involving P. tremuloides genome portions, since SINEs of these two species are highly related. However, due to lineage-specific SINE evolution during speciation, cross-species applications are generally only successful to limited extent. The analysis of poplar hybrids composed of P. maximowiczii with either P. trichocarpa or P. nigra based on P. tremula ISAP primers showed a strongly reduced resolution. In forestry, hybrid larch (e.g. Larix × eurolepis) genotypes have to be selected from the offspring of Japanese (Larix kaempferi) and European larch (Larix decidua) crosses, as they exhibit superior growth rates compared to the parental species. Initial ISAP-based examinations of European larch genotypes provided less polymorphic banding patterns, probably resulting from general high levels of synteny and collinearities reported for gymnosperm species. Hence, the ISAP was combined with the AFLP technique to the novel marker system inter-SINE-restriction site amplified polymorphism (ISRAP). The amplicons originating from genomic regions between SINEs and EcoRI cleavage sites were visualized with the sensitive capillary gel electrophoresis. The ISRAP assays, based on EcoRI adapter primers combined with two different SINE-derived primers, resulted in a sufficient number of polymorphic peaks to distinguish the L. decidua genotypes investigated. Compared to ISAPs, the ISRAP approach provides the required resolution to differentiate highly similar larch genotypes.

Die Gattung Beta gehört zur Familie der Chenopodiaceae und wird in die vier Sektionen Beta, Corollinae, Nanae und Procumbentes unterteilt, wobei die Zuckerrübe der Sektion Beta zugeordnet wird. Aus dem Genom der Zuckerrübe und verwandter Wildarten konnten bereits eine Vielzahl von repetitiven DNA-Familien kloniert und untersucht werden. Mit der monosomen Fragmentadditionslinie PRO1 stand eine Chromosomenmutante zur Verfügung, die neben den 18 B. vulgaris-Chromosomen ein Chromosomenfragment der Wildrübe Beta procumbens enthält. Da dieses als Minichromosom bezeichnete Fragment mitotische Stabilität aufweist, muss es ein funktionelles Centromer besitzen, das auch im genetischen Hintergrund von Beta vulgaris aktiv ist. Mit der Erstellung einer BAC (bacterial artifical chromosome)-Bank von PRO1 wurde die molekulare Charakterisierung von Ty3-gypsy-Retrotransposons eines einzelnen Wildrüben-Centromers möglich. Die für die Wildrübe Beta procumbens spezifischen Satellitenrepeats pTS5 und pTS4.1 dienten der Selektion von BACs aus der Centromer-Region des PRO1-Minichromosoms. Die Identifizierung eines unikalen genomischen Locus, mit einer Verschachtelung von zwei nicht homologen LTR-Retrotransposons, ermöglichte die gerichtete Isolation der LTR-Retrotransposons Beetle1 und Beetle2. Das Retrotransposon Beetle1 hat eine Gesamtlänge von 6736 bp und wird von LTR-Sequenzen begrenzt, die eine Länge von 1091 bp (5’-LTR) bzw. 1089 bp (3’-LTR) aufweisen. Das LTR-Retrotransposon Beetle2 weist mit 6690 bp eine ähnliche Gesamtlänge wie Beetle1 auf. Es wird von deutlich kürzeren LTR-Sequenzen mit einer Länge von 774 bp begrenzt. Aufgrund der Reihenfolge der Polyproteingene lassen sich Beetle1 und Beetle2 in die Gruppe der Ty3-gypsy-Retrotransposons (Metaviridae) einordnen. Beide Retrotransposon-Familien besitzen ein einziges offenes Leseraster (open reading frame; ORF) mit fusionierten gag- und pol-Genen. Datenbankrecherchen zeigten hohe Homologien von Beetle1 und Beetle2 mit den centromerischen Ty3-gypsy-Retrotransposons CRM aus Zea mays, CRR aus Oryza sativa und cereba aus Hordeum vulgare. Diese centromerischen Retrotransposons (CRs) sind in den Poaceae stark konserviert und stellen neben Satellitenrepeats eine hochabundante Sequenzklasse der Centromere der Süßgräser dar. Da sie im 3’-Bereich des gag-pol-Polyproteins eine Chromodomäne aufweisen, werden sie der eigenständigen Gruppe der Chromoviren zugeordnet. Chromodomänen sind zur Bindung von Proteinen und DNA befähigt und spielen eine wichtige Rolle in der Chromatin-Modifikation und der Bildung von Heterochromatin-Regionen. Beetle1 und Beetle2 besitzen Motive einer Chromodomäne, die vermutlich für eine gerichtete Transposition in die Centromer-Region verantwortlich ist. Neben der geringen Divergenz von Beetle1- und Beetle2-Sequenzen sowohl im Genom von Beta procumbens als auch in den anderen Arten der Sektion Procumbentes spricht auch das junge Alter von 100 000 bis 350 000 Jahren und die Transkriptionsaktivität für eine Einordnung dieser Ty3-gypsy-Retrotransposons in die Gruppe der Chromoviren. Sowohl die Southern-Hybridisierung als auch die Fluoreszenz-in situ-Hybridisierung zeigten, dass Beetle1 und Beetle2 nur für die Sektion Procumbentes spezifisch sind und dort in hoher Kopienzahl vorkommen. Untersuchungen mit methylierungssensitiven Restriktionsendonukleasen veranschaulichten den hohen Grad an Cytosin-Methylierung von Beetle1 und Beetle2.

Le promoteur viral module la transcription et par conséquent la réplication virale. Le promoteur du VIH-1 est polymorphe pour une variété de sites de fixation de facteurs de transcription, toutefois l'impact de ce polymorphisme sur la réplication virale in vivo n'est pas élucidé. Nous avons construit des virus chimères isogèniques portant le promoteur minimal des sous-types B, C, et E du VIH-1 dans le génome du VISmac239. Dans quatre macaques rhésus co-infectés avec les trois chimères, une discordance entre les génotypes des virus associés aux rPBMCs et des virions du sérum est observée lors de la primo-infection. Un changement dans le génotype viral est également visible entre la primo-infection et la phase asymptomatique. Nos résultats montrent que le polymorphisme du promoteur des sous-types du VIH-1 constitue un déterminant majeur pour la réplication et la dissémination virale, et que la virémie en primo-infection est majoritairement due à la réplication virale dans l'intestin.

La méthylation de l'ADN et les modifications post-traductionnelles des histones sont desmodifications épigénétiques qui interviennent dans la régulation des éléments transposables(ET) chez de nombreuses espèces. La proportion des ET dans les génomes varie selon lesespèces considérées et pose la question des mécanismes de régulation de ces ET. Au sein del'espèce Drosophila simulans, les populations naturelles présentent un polymorphisme uniquedans le nombre de copies des ET, ce qui en fait un excellent modèle pour étudier cettequestion. L'étude de la méthylation d'ADN et des modifications post-traductionnelles deshistones associées au rétrotransposon à LTR tirant dans la lignée germinale des populationsnaturelles a permis de montrer l'influence d'une copie d'ET sur la structure de la chromatineau site d'insertion. Dans un second volet, nous avons cherché à caractériser la méthylation del'ADN chez la drosophile, chez laquelle la fonction est encore mal connue. Nous avons, pardes approches spécifiques et globales, mesuré l'abondance de cette marque épigénétique chezla drosophile. Nous concluons que les taux de méthylation de l'ADN sont très faibles maisvariables entre espèces. Notre travail n'a pas permis de mettre en évidence un rôle de laméthylation de l'ADN dans le contrôle des ET, toutefois, nous ne pouvons pas exclure cesystème de régulation.

FundaÃÃo Cearense de Apoio ao Desenvolvimento Cientifico e TecnolÃgico / Na presente tese, propÃe-se um controlador robusto LQG/LTR (Linear Quadratic Gaussian with Loop Transfer Recovery) com aÃÃo integral (LQG/LTRI) em uma nova aplicaÃÃo para o ajuste dos controladores do conversor do lado da mÃquina e do conversor do lado rede elÃtrica em um sistema de conversÃo eÃlica utilizando um gerador de induÃÃo duplamente alimentado (Doubly-Fed Induction Generator - DFIG). A metodologia de controle proposta assegura a robustez em relaÃÃo à rejeiÃÃo do erro de rastreamento, insensibilidade a variaÃÃes paramÃtricas, alÃm de permitir que erros de medida e modelagem sejam incorporados no projeto. Testes de robustez e desempenho foram realizados para variaÃÃes dos parÃmetros internos da mÃquina e variaÃÃes de referÃncia de velocidade. Resultados de simulaÃÃo e experimentais, obtidos em um protÃtipo de laboratÃrio com uma mÃquina de 2kW sÃo apresentados para validar e demonstrar o bom desempenho e robustez do controlador proposto comparado com os controladores clÃssicos Proporcional-Integral (PI) em um sistema de geraÃÃo eÃlica com mÃquinas DFIG. / It is proposed in this thesis a LQG/LTR robust controller (Linear Quadratic Gaussian with Loop Transfer Recovery) with integral action (LQG/LTRI) in a new application for gain tuning for the rotor side converter and grid side converter in a wind energy conversion system (WECS) with Doubly-Fed Induction Generator(DFIG). The proposed control method assures the robustness with respect to the tracking error rejection, insensitivity to parameter variations and permits that measurement and modeling errors are incorporated in the project. Robustness and performance tests were performed for variations of machine internal parameters and speed. Simulation and experimental results, obtained from a laboratory prototype that uses a 2 kW machine are presented to validate and demonstrate the robustness and performance of the proposed controller through the comparison with the proportional and integral (PI) controller employed in a wind energy conversion system with DFIG.

Les travaux de cette thèse porte sur le calcul d’un espace d’état navigable d’un véhicule routier, ainsi que sur l’observation et l’estimation de son état, à l’aide des méthodes ensemblistes par intervalles. Dans la première partie de la thèse, nous nous intéressons aux problèmes d’estimation d’état relevant de la dynamique du véhicule. Classiquement, l’estimation se fait en utilisant le filtrage de Kalman pour des problèmes d’estimation linéaires ou le filtrage de Kalman étendu pour les cas non-linéaires. Ces filtres supposent que les erreurs sur le modèle et sur les mesures sont blanches et gaussiennes. D’autre part, les filtres particulaires (PF), aussi connus comme Méthodes de Monte-Carlo séquentielles, constituent souvent une alternative aux filtres de Kalman étendus. Par contre, les performances des filtres PF dépendent surtout du nombre de particules utilisées pour l’estimation, et sont souvent affectées par les bruits de mesures aberrants. Ainsi, l’objectif principal de cette partie de travail est d’utiliser une des méthodes à erreurs bornées, qui est le filtrage par boites particulaires (Box Particle Filter (BPF)), pour répondre à ces problèmes. Cette méthode généralise le filtrage particulaire à l’aide des boites remplaçant les particules. A l’aide de l’analyse par intervalles, l’estimation de certains variables fortement reliées à la dynamique du véhicule comme le transfert de charge latérale, le roulis et la vitesse de roulis est donnée, à chaque instant, sous forme d’un intervalle contenant la vraie valeur simulée. Dans la deuxième partie de la thèse, une nouvelle formalisation du problème de calcul de l’espace navigable de l’état d’un véhicule routier est présentée. Un algorithme de résolution est construit, basé sur le principe de l’inversion ensembliste par intervalles et sur la satisfaction des contraintes. Nous cherchons à caractériser l’ensemble des valeurs de la vitesse longitudinale et la dérive au centre de gravité qui correspondent à un comportement stable du véhicule : pas de renversement ni dérapage. Pour décrire le risque de renversement, nous avons utilisé l’indicateur de transfert de charge latéral (LTR). Pour décrire le risque de dérapage, nous avons utilisé les dérives des roues. Toutes les variables sont liées géométriquement avec le vecteur d’état choisi. En utilisant ces relations, l’inversion ensembliste par intervalles est appliquée afin de trouver l’espace navigable de l’état tel que ces deux risques sont évités. L’algorithme Sivia est implémenté, approximant ainsi cet espace. Une vitesse maximale autorisée au véhicule est déduite. Elle est associée à un angle de braquage donné sur une trajectoire connue. / In this thesis, we aim to characterize a vehicle stable state domain, as well as vehicle state estimation, using interval methods.In the first part of this thesis, we are interested in the intelligent vehicle state estimation.The Bayesian approach is one of the most popular and used approaches of estimation. It is based on the calculated probability of the density function which is neither evident nor simple all the time, conditioned on the available measurements.Among the Bayesian approaches, we know the Kalman filter (KF) in its three forms(linear, non linear and unscented). All the Kalman filters assume unimodal Gaussian state and measurement distributions. As an alternative, the Particle Filter(PF) is a sequential Monte Carlo Bayesian estimator. Contrary to Kalman filter,PF is supposed to give more information about the posterior even when it has a multimodal shape or when the noise follows non-Gaussian distribution. However,the PF is very sensitive to the imprecision due by bias or noise, and its efficiency and accuracy depend mainly on the number of propagated particles which can easily and significantly increase as a result of this imprecision. In this part, we introduce the interval framework to deal with the problems of the non-white biased measurements and bounded errors. We use the Box Particle Filter (BPF), an estimator based simultaneously on the interval analysis and on the particle approach. We aim to estimate some immeasurable state from the vehicle dynamics using the bounded error Box Particle algorithm, like the roll angle and the lateral load transfer, which are two dynamic states of the vehicle. BPF gives a guaranteed estimation of the state vector. The box encountering the estimation is guaranteed to encounter thereal value of the estimated variable as well.In the second part of this thesis, we aim to compute a vehicle stable state domain.An algorithm, based on the set inversion principle and the constraints satisfaction,is used. Considering the longitudinal velocity and the side slip angle at the vehicle centre of gravity, we characterize the set of these two state variables that corresponds to a stable behaviour : neither roll-over nor sliding. Concerning the roll-over risk,we use the lateral transfer ratio LTR as a risk indicator. Concerning the sliding risk, we use the wheels side slip angles. All these variables are related geometrically to the longitudinal velocity and the side slip angle at the centre of gravity. Using these constraints, the set inversion principle is applied in order to define the set ofthe state variables where the two mentioned risks are avoided. The algorithm of Sivia is implemented. Knowing the vehicle trajectory, a maximal allowed velocityon every part of this trajectory is deduced.

Main motive of this master thesis was the need of good bioinformatics tools for genome comparison and improvement of one of the existing tools - RepeatExplorer. This work offers an overview of transposable elements in DNA, existing tools for identification and analysis of repetitions in sequenced genomes, summary of currently used genome sequencing methods. This work describes shortcomings of RepeatExplorer tool with focus on comparative analysis of genomes. Two solutions to remove these problems were designed and implemented. The first solution is designed for comparing pairs of genomes. The principle of this solution is based on comparison of similarity of distribution of contigs coverages using Kolmogorov-Smirnov test, thanks to which we are able to determine different parts in the genomes.The second solution, which is used to compare multiple genomes, is based on the method of mapping reads from compared genomes to the reference genome contigs and provides contigs coverage graphs, by which we are able to determine the variability of the repeats.Their functionality was verified on real NGS data of organism Silene latifolia.

<p>Докторска дисертација припада области информационих система, са јасним акцентом на употребу неуронских мрежа за решавање проблема вишеструких зависних временских серија&nbsp; који је у&nbsp; овом докторату дефинисан.Основни циљ дисертације је креирање система у форми едукативног софтвера путем којег ће се обучавати младе кошаркашке судије</p><p>Један од кључих елемената овог&nbsp; доктората јесте симулација хоризонталног видног поља на основу којег се&nbsp; утврђује да ли је резоновано кретање кошаркашких судија било адекватно&nbsp; или није. Стога развијени софтвер поседује споменуту едукативну примену.</p><p>Како би се реализовао споменути софтвер спроведено је истраживање које је обухватило обучавање великог броја традиционалних вишеслојних перцептрона као и формирање посебне&nbsp; LTR&nbsp; &ndash;&nbsp; MDTS структуре неуронске мреже за коју се сматра да је погодна за решавање постојећег проблема. За реализацију симулације хоризонталног видног поља разматрано је више алгоритама из области рачунарске графике&nbsp; а&nbsp; Sweep and Prune&nbsp; алгоритам је парцијално пружио основу за развијени и тренутно&nbsp; имплементирани алгоритам.</p> / <p>Doktorska disertacija pripada oblasti informacionih sistema, sa jasnim akcentom na upotrebu neuronskih mreža za rešavanje problema višestrukih zavisnih vremenskih serija&nbsp; koji je u&nbsp; ovom doktoratu definisan.Osnovni cilj disertacije je kreiranje sistema u formi edukativnog softvera putem kojeg će se obučavati mlade košarkaške sudije</p><p>Jedan od ključih elemenata ovog&nbsp; doktorata jeste simulacija horizontalnog vidnog polja na osnovu kojeg se&nbsp; utvrđuje da li je rezonovano kretanje košarkaških sudija bilo adekvatno&nbsp; ili nije. Stoga razvijeni softver poseduje spomenutu edukativnu primenu.</p><p>Kako bi se realizovao spomenuti softver sprovedeno je istraživanje koje je obuhvatilo obučavanje velikog broja tradicionalnih višeslojnih perceptrona kao i formiranje posebne&nbsp; LTR&nbsp; &ndash;&nbsp; MDTS strukture neuronske mreže za koju se smatra da je pogodna za rešavanje postojećeg problema. Za realizaciju simulacije horizontalnog vidnog polja razmatrano je više algoritama iz oblasti računarske grafike&nbsp; a&nbsp; Sweep and Prune&nbsp; algoritam je parcijalno pružio osnovu za razvijeni i trenutno&nbsp; implementirani algoritam.</p> / <p>Doctoral dissertation belongs to the field of information systems, with a clear&nbsp; emphasis on the use of neural networks for solving the problem of multiple dependent time series, which is defined in this doctorate. The main objective of the&nbsp; thesis is to create a system in the form of educational software that will be used druring the training of young basketball referees.One of the key elements of this doctorate is a simulation of a horizontal field of&nbsp; vision&nbsp; on the basis of which it is determined whether the movement of&nbsp; reasoned&nbsp; basketball referees was adequate &nbsp; or not.&nbsp; Therefore developed software has&nbsp; aforementioned educational use. In order&nbsp; to realize the aforementioned software, a&nbsp; research&nbsp; was conducted that included training of a large number of traditional multilayer perceptron neural networks and the&nbsp; formation of special LTR&nbsp; -&nbsp; MDTS&nbsp; neural network&nbsp; structure which is considered to be&nbsp; suitable&nbsp; for solving the presented problem. For the realization of the simulation&nbsp; of the horizontal field of vision a large number of algorithms in the field of computer graphis was considered&nbsp; and Sweep and Prune algorithm partially provided the basis for the developed and&nbsp; currently implemented algorithm.</p>