Unmanned Underwater Vehicles (UUV) in open-seas are highly nonlinear with system motions. Because of the complex interaction of the body with environment it is difficult to control them efficiently. Linearization is applied to system in order to design controllers developed for linear systems. To overcome the effects of disturbances, a mathematical model which will compensate all disturbances and effects of linearization is required. In this study first a mathematical model is formed wherein the linear and nonlinear hydrodynamic coeffi- cients are calculated with strip theory. After the basic mathematical model is developed, it is simplified and decoupled into speed, steering and diving subsystems. Consequently PID (Proportional Derivative Integral), SMC (SlidingMode Control) and LQR (Linear Quadratic Regulator)/LQG (Linear Quadratic Gaussian) control methods can be applied on each subsystem to design controllers. Some of the system parameters can be estimated from state vector data based on measurements using the methods of linear sequential estimation and genetic algorithms. As for the final part of the study, an online obstacle avoidance algorithm which avoids local optimums using Boolean operators is presented. In addition a simple guidance algorithm is suggested for waypoint navigation. Due to the fact that ULISAR UUV is still on construction phase, we were unable to test our algorithms. But in the near future, we plan to study all these algorithms on the UUV ULISAR.

In vector control of induction motors it is essential to know the parameters of the motor. Known approaches to this problem have some drawbacks. This thesis work is planned to develop solutions to the existing problems. The proposed solutions will be implemented and tested.

High dynamic performance, which is obtained from dc motors, became achievable from induction motors with the recent advances in power semiconductors, digital signal processors and development in control techniques. By using field oriented control, torque and flux of the induction motors can be controlled independently as in dc motors. The control performance of field oriented induction motor drive greatly depends on the correct stator flux estimation. In this thesis voltage model is used for the flux estimation. Stator winding resistance is used in the voltage model. Also leakage inductance, mutual inductance and referred rotor resistance values are used in vector control calculations. Motor control algorithms use motor models, which depend on motor parameters, so motor parameters should be measured accurately. Induction motor parameters may be measured by conventional no load and locked rotor test. However, an intelligent induction motor drive should be capable of identifying motor parameters itself. In this study parameter estimation algorithms are implemented and motor parameters are calculated. Then these parameters are used and rotor flux oriented vector control is implemented. Test results are presented.

Made available in DSpace on 2017-07-21T20:43:49Z (GMT). No. of bitstreams: 1 Pedro Henrique F Oliveira.pdf: 13888256 bytes, checksum: 45c80d3205c28d94e4c7c0eb2987187f (MD5) Previous issue date: 2016-09-23 / The aim of this study was to produce deposit of the alloy AA 6351-T6, filled with alumina particles, on an AA5052-32 alloy substrate using friction-surfacing technology. The depositions were performed with holes in the AA6351-T6 consumable rods, which were filled with particulate alumina (Al2O3). A conventional milling machine, KONE KFE-3 / BR, available by SENAI Ponta Grossa – Paraná, was used to carry out the deposition. The control parameter used for the making the deposits was the rod feed rate being calculated from the relationship between the upward axial displacement in the Z axis (Dz) of the machine table and the total time (t) of deposition. The deposits obtained were evaluated from their physical aspects such as width, thickness and length. In addition, bending tests were conducted for a qualitative assessment of the influence of the alumina particles in the adhesion of the deposit to the substrate and microhardness profiles were obtained to evaluate the influence of the alumina particles produced in the hardness of the deposits. Images of optical microscopy images (OM) and scanning electron microscopy (SEM) were obtained to evaluate the microstructure and morphology of the deposits produced. The volume fraction of alumina particles in the deposits were obtained by digital processing of the images by optical microscopy. EBSD analysis of deposits in specific regions were performed in order to support the microstructural analysis, enabling verify grain size distribution in different regions of deposits showing the existence of substructures indicating the occurrence of dynamic recrystallization phenomenon. The results showed that it was possible to produce deposits with a refined microstructure and with a certain dispersion of alumina particles. Depositions with two holes in the rod produced deposits with higher volume fraction, and this increase in volume fraction was accompanied by higher hardness values. / O presente trabalho buscou depositar a liga AA 6351-T6, preenchida com partículas de alumina (Al2O3), sobre um substrato de liga AA5052-32 através da tecnologia de deposição superficial por atrito (friction surfacing). As deposições foram realizadas com um e dois furos nas hastes de liga AA6351-T6, que foram preenchidos com partículas de alumina. Para a realização das deposições foi utilizada uma fresadora convencional KONE KFE-3/BR disponibilizada pelo SENAI Ponta Grossa – Paraná. O parâmetro de controle utilizado para a confecção dos depósitos foi a taxa de alimentação da haste, sendo calculada a partir da relação entre o deslocamento axial ascendente no eixo Z (Dz) da mesa da máquina com o tempo total (t) de deposição. Os depósitos obtidos após o processamento foram avaliados a partir de seus aspectos físicos tais quais largura, espessura e comprimento. Também foram realizados ensaios de dobramento para uma avaliação qualitativa da influência das partículas de alumina na adesão do depósito no substrato e foram obtidos perfis de microdureza para avaliação da influência das partículas de alumina na dureza dos depósitos produzidos. Foram obtidas imagens de microscopia ótica (MO) e microscopia eletrônica de varredura (MEV) para avaliação da microestrutura e morfologia dos depósitos produzidos. Ainda foi obtida a fração volumétrica das partículas de alumina nos depósitos produzidos através de processamento digital das imagens obtidas por microscopia óptica. Análises de EBSD em regiões específicas dos depósitos foram realizadas com o intuito de reforçar a análise microestrutural possibilitando verificar a distribuição de tamanho de grão em diferentes regiões dos depósitos, mostrando a existência de subestruturas indicando a ocorrência do fenômeno de recristalização dinâmica. Os resultados mostraram que foi possível produzir depósitos com uma microestrutura refinada e com uma certa dispersão de partículas de alumina. As deposições com dois furos na haste produziram depósitos com maior fração volumétrica, sendo que esse aumento da fração volumétrica foi acompanhada pelo aumento dos valores de dureza.

Thesis (Ph. D.)--Ohio State University, 2005. / Available online via OhioLINK's ETD Center; full text release delayed at author's request until 2006 Aug 15.

This thesis is focused on problematics of control of brushless DC motor in the sensor and also in the sensorless mode. Also it interprets possibilities of BLDC motor control with one faulty sensor and derivation and simulation of mathematical model. First part mentions options of rotor position sensing as well as existing methods of sensorless BLDC motor control. Second part describes control algorithms of sensor and sensorless motor control realised on device dSPACE and also realisation of faulty sensor control algorithm. Third part deals with derivation of mathematical model, its realisation using Matlab Simulink software and identification of its parameters. Last part concludes results and compares control methods used on the real system.