Repositório RCAAP

The objective of this work is to develop real time dynamic simulators to use them as supporting tools in instrumentation and process control technical courses. The developed simulators are as following : - simulator for the equipment to test control valves showing all the acessories necessary to test a control valve. In this phase the concepts of Cv, curves inherent and installed of a control valve are introduced. - simulator in real time of a level control using P+I+D controller for a tank with the form of cylinder in the vertical position, in this phase the Ziegler-Nichols tunning methods are presented for systems with different response times and which are the influence of the control valve gain in the calculated tunning parameters. In this phase a system with fast response are presented. The same simulation is made using the same system with larger response time. During this simulation the influence of the gain of the valve is shown in the process response of the control and what is procedure to solve this problem. At this part, the advantages and disadvantages of the Ziegler-Nichols tunning methods in a practical applications. - simulator in real time of three elements level control for a boiler. In this phase a experimental method is showing for controller's tunning parameters. The experimental method presents advantages of practical order in reference of Ziegler- Nichols method. - This work was developed to permits to analyse the behaviour of dynamic system as well to study solutions that can do in the classroom without the concerns about safety which normally occurs in the industrial environment. The great target of this work is to show in the classroom experimental methods, as the own name indicates, only can be done in the real equipment. Also we believe that sigficant coast reduction can be achieve if simulators are used instead pilots plants.

This dissertation presents an educational software which explores Virtual Reality techniques (VR) through the development of virtual educational environments. 3D Geometry has been taken as a case study. The objective of this work is the extension of the cut for convex and non-convex polyhedron. The virtual environment and the interaction techniques have been developed using the VRML (Virtual Reality Modeling Language) and JavaScript languages. After making the system available to potential users, they were able to do some experiments and confirmed that the techniques proposed are satisfactory and helpful in the learning process and in the visualization of non-convex polyhedron.

When the suply chain of medicine logistic is analysed, critical points can be found in the actual model of deliveries to the drugstores, which are the direct clients of the dealer. The clients demand full requests to be delivered on time. In this context, this work starts with the evaluation of the logistic process of the dealer. A field study has been conducted in order to diagnose and develop a core system for vehicle routing. The idea is that an optimized delivering process will reduce costs and, most importantly, the delay on deliveries. Considering the complexity and relevance of the problem in the context of logistics, the chosen approach uses genetic algorithms as the metaheuristic. It is robust enough to tackle the given problem. It is an interactive method with some intelligence in the process of searching for solutions which does not stop on the first local optimum that is found. The problem is to optimize the allocation of deliveries to the available vehicles, taking into account the restrictions of each vehicle, in such a way that the total travelled distance will be minimized. The representation of the solutions and the genetic operators used are based on the travelling salesman problem. In order to produce competitive results on this class of problems, Genetic Algorithms need to be associated with a local method of search. This should be done at each iteration of the algorithm. This allows the generation of better solutions when compared to conventional heuristic methods. The system has been implemented in the Ocaml programming language. Ocaml is a modern language which allow quick development and the code is less error prone, when compared to other programming languages common on the software market, like C, C++ and Java. Known problems from the literature have been used to test the system. The results show that it is possible to obtain optimized routers to solve the problem of delivery of medicines to the clients, taking into account restrictions like capacity of vehicles and transported value limitations.

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior

Conselho Nacional de Desenvolvimento Científico e Tecnológico

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior

Conselho Nacional de Desenvolvimento Científico e Tecnológico

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior

Nowadays, the high level of complexibility designated by duties in many areas are expecting more from the man his natural senses can offered him. The use of Virtual Reality techniques can auxiliate the development of perception, interaction and consequently the every day productivity. Basing it in this idea, this research has as objectivity to create an application to offer the user information about the virtual urban rotes. This dissertation present a architecture for construction of a virtual environment which reproduce virtual urban rotes having as instrument the search of the best rote, a algorithm based on evolutive computation, denominated Genetic Algorithm, and, as instrument of visualization of the scene of the virtual rote, streets of a city in three dimensions, a Virtual Reality. A prototype was built having as references a neighborhood of a city seen under two points of view. The first one in two dimensions allowing the choice of a start point (origin) and an end point (destiny) and a way to be covered between the points with the shortest distance possible. For treating of a probabilistic problem, where there are innumerous possibilities of solution, the genetic algorithm were choose for making framing in this type of problem possible. The second point of view, in three dimensions, beyond offer to be virtual environment with possibilities of free navigation on the scene of a city, proportionate to the user navigation for ways built by the model 2D. The 3D model was built with the support of the graphic library OpenGL and the geometric model of the city was drawn with the use of a specific tool of graphic modeling been imported to the prototype. The system use for both points of view (2D and 3D) the same model, which proportionate a portability in relation to the scenes (cities), so it s enough to have a geometric model of a neighborhood or city to the application functioning. The system was evaluated by researchers and specific users and the results reached allowed concluding that the same is efficient and applicable.

The visualization of a 3D image obtained from computerized tomography examinations has shown itself to be an important factor for increasing the quality of medical diagnoses and, consequently, treatment efficacy. There already exist on the market, several visualization softwares, which use different techniques to show the 3D tomography image. However, to show a high quality 3D image, sophisticated devices must be used to obtain slices, close to one another, thus increasing the incidence of X-ray given to the patient. An interpolation slice method which resulted from the TC examination produces good results, and is able to reduce the X-ray incidence upon the patient. This method must reconstruct the curvature from the patient s internal structures without using slices in close proximity. This work proposes a method of 3D image interpolation, composed of a juxtaposition of the slices from CT examination results. The goal of this method is to increase the quality of 3D visualization through the production of sharp and precise structure contours. This thesis proposes the division of the interpolation method into two steps. In the first step, the goal is to obtain an initial representation of the image in 3D, which is composed of real slices as well as virtual slices which are referred to in this work as initial virtual slices. In the second step, the empty spaces of the structure are recovered by the 3D image inpainting process. This work also proposes a method to obtain the initial virtual slice and two different methods for inpainting the 3D image. These inpainting methods are the transversal slice line prolongation method and the transportation and diffusion of information. Both methods use the differential equation theory. The transportation and diffusion of information method shows better results than other methods proposed in this work, besides this, this method presents better results than the linear interpolation and Goshtasby et al. [1] methods also implemented in this work. Visual and numerical comparisons are used to obtain this conclusion. The numerical measures used are statistical correlation, the PSNR and the Hausdorff distance [2]. The transportation and diffusion of information method shows itself able to produce better results than all the other tested methods. Besides this principal contribution, this work also developed a KIT to implement 2D and 3D CT visualize applications.

OCaml and other ML like languages does not provide developer with a native GUI. Therefore one must use a third party GUI, like TCL/TK or GTK. This requires final users to search the Internet for a GUI library, download it, install it, etc. Most final users are not up to the task. Therefore we decided to create a GUI interface entirely written in OCaml, that does not require any external library to work, and that has a very small footprint. Compiled programs occupy as little as 300Kbytes on disk. The system works equaly well on Windows, Linux, or any other operational system with an OCaml port. It has four kinds of widgets: Buttons, refresh devices, parsers, and Emacs-like mini-editors. The Emacs widgets work with plain text or mask-driven forms; they offer a practical and simple method for data input. Since OCaml is a strong typed functional language, this application is pratically bug free.

The purpose of this work is to generate a better Quantization Table for a group of natural images and medical images using the Genetic Algorithm method in the process of compression JPEG. The fitness function of the population will be analyzed and the probability of selection the better pairs of chromosomes or matrix, and we will also breach means to improve the results through modifications of the input parameters. In the previous processing of the coded image it is necessary to know very well the Quantization Table to achieve an image with less loss and therefore, better final quality. The method of Genetic Algorithm applied to this program is based on mechanisms of natural selection and reproduction for a set of natural and medical images and furthermore it allowed for the creation of not one but various Tables with SNR higher than those produced by the JPEG Table since it stimulated the principal characteristics of this method in programming such as: codification of defined parameters; the process of search for better matrix starting from a group of matrix and not a simple matrix; information obtained of an objective determined function (fitness function) and the use of the probabilistic rules of transition. Based on the results obtained in this work, we can recommend the use of Genetic Algorithm for natural images and human eye images with a compression rate up to 30:1 and a great reconstructed image quality.

This work presents as main goal to implement a computational application that allows the user create MIDI musical sequences for guitar. The difference of the musical sequences generated by the software resides in the fact that each musical note of the MIDI sequence can be modified for the inclusion of humanized articulation, individualized and with timbral qualities like an acoustic instrument. The sound resulted of these sequences is render in other musical formats, such as Wave. The application interface and functions are developed in pure functional paradigm (based on lambda calculus). This choice avoids the use of dlls and other resources strongly dependent of operational system version. The choice of the functional language Clean is justified in this work, by the reason of being one of the two more efficient languages of the actuality (according to analysis of specialized benchmarks). In the implementation of the manipulation functions of SMF files and interface generated libraries that will allow extending this work for another soloists (monophonic) instruments. This application presents an adherent interface with all kinds of users. In order to manipulate SMF files, many libraries were developed, allowing extending this work for other soloists (monophonic) instruments. This application presents an adherent interface with all levels of users.

The extraction of signal components from Auditory Brainstem Response (ABR), referred to as Jewett waves, is an important step in studies regarding the investigation of the behavior of the auditory system. However, this evaluation is complex and depends on the experience of the specialist. In this context, the objectives of this research are to compare ABR analyses provided by different examiners, and also to verify the possibility of discrimination of Jewett waves through the analysis of instantaneous attributes (IA) of time-series, estimated by the Hilbert Transform (HT). The ABR data were collected from healthy subjects, following the criterion of auditory normality, and analyzed by four examiners. A statistical study aiming to verify the coherence among these analyses was performed. The HT was then employed for calculating the IA from ABR signals. The compatibility among the results provided by the examiners was verified. The analysis of IA suggests that they can be used as discriminative features for the different waves that compose the ABR. These results are relevant for the development of tools for automatic ABR analysis. Furthermore, it contributes to investigations that aim to discriminate ABR related to distinct disorders.

Universidade Federal de Uberlândia

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior

This dissertation presents and analyzes the results of theoretical investigations and experimental of the thermal behavior of transformers the dry, isolated with resin epoxy, under conditions sinusoidal and non sinusoidal of tension and current. Development a study of the losses in the several parts of the transformer, standing out the losses for currents parasites proposed by the norm ANSI/IEEE [3] and was compared with other [5], more exact, that takes in consideration the geometry of the transformer. Inside of this context present new equations for the temperature elevations, rendiment, useful life and derating of the transformer. A thermal model is used to estimate the temperature elevations in the transformer. Experimental studies are led in a prototype of a transformer trifásico the dry, isolated with resin epoxy, delta/estrela, 5 kVA especially built for the accomplishment of temperature tests, with sensor thermal located in several points of the equipment. This procedure makes possible the comparison among the results computations, with views to the validation of the proposal of the thermal modelling of first order.

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior

This thesis applies computer graphics to two important fields of Biomedical Engineering. Firstly, in order to provide new software for interdisciplinary education, an interactive Virtual-based platform was developed, presenting a tridimensional visualization of the neuron and of its major microestructures. Such platform enables an overall view of the neural cell, including particular details in a microcellular level. Several issues regarding the tridimensional reconstruction of neuroanatomical strucutures based on photos taken from real-life tissues are discussed, leading to the implementation of a bidimensional atlas, that establishes connections between neuroanatomy and the clinical practice. Secondly, a system for the efficient and accurate estimation of the time intervals associated with swallow phases was developed, supposing videofluoroscopic images as inputs. The system performs pre-processing of such images by means of file conversions, followed by image analysis, that leads to the final estimation. A clinical experiment was carried out in order to validate the platform, yielding (allowing) a preliminar proposition for a normal pattern of swallow , which is based on the average-estimated swallow times. Results pointed out that the system may be considered, flexible, requiring simple hardware/software configurations, and also providing details on fractioning.