Orientador: Gustavo Luiz Chagas Manhães de Abreu / Resumo: Fluidos magnetoreológicos (MR) são fluidos capazes de alterar suas propriedadesreológicas quando um campo magnético é aplicado sobre ele. Uma das aplicações maisimportantes do fluido MR é em amortecedores de vibrações, utilizados principalmente naconstrução civil, veículos automotivos e outros sistemas mecânicos sujeitos a excitaçõesque provocam vibrações indesejáveis. Na indústria automobilística, atualmente atecnologia dos amortecedores que utilizam fluido MR vem se destacando como umasolução que pode trazer benefícios de conforto e segurança aos usuários de veículos emgeral. Este trabalho discute a modelagem não-linear de um veículo que considera adinâmica vertical, lateral e longitudinal, simulado em diferentes condições de conduçãoque buscam avaliar o conforto, a aderência à estrada, a dirigibilidade, a rolagem e adeflexão da suspensão. Pretende, também, contribuir com a área de controle de vibraçõesem suspensões veiculares que utilizam amortecedores MR, avaliando o desempenho doscontroladores ótimo (LQR), nebuloso e FEB (Frequency-Estimation-Based ) projetadosem 1/4 de veículo e aplicados ao modelo não-linear do veículo. O trabalho terminacomentando as potencialidades da metodologia apresentada, discutindo as facilidadese dificuldades encontradas na sua implementação e aponta propostas para a suacontinuidade. / Doutor

Submitted by THIAGO GALBIATI LAGOIN null (thiagolagoin@yahoo.com.br) on 2017-10-26T18:00:51Z No. of bitstreams: 1 lagoin_tg_dr_ilha.pdf: 10803891 bytes, checksum: 11fc5487a11d931ddaa2432fe578288e (MD5) / Approved for entry into archive by Monique Sasaki (sayumi_sasaki@hotmail.com) on 2017-10-31T18:40:57Z (GMT) No. of bitstreams: 1 lagoin_tg_dr_ilha.pdf: 10803891 bytes, checksum: 11fc5487a11d931ddaa2432fe578288e (MD5) / Made available in DSpace on 2017-10-31T18:40:57Z (GMT). No. of bitstreams: 1 lagoin_tg_dr_ilha.pdf: 10803891 bytes, checksum: 11fc5487a11d931ddaa2432fe578288e (MD5) Previous issue date: 2017-09-28 / Fluidos magnetoreológicos (MR) são fluidos capazes de alterar suas propriedades reológicas quando um campo magnético é aplicado sobre ele. Uma das aplicações mais importantes do fluido MR é em amortecedores de vibrações, utilizados principalmente na construção civil, veículos automotivos e outros sistemas mecânicos sujeitos a excitações que provocam vibrações indesejáveis. Na indústria automobilística, atualmente a tecnologia dos amortecedores que utilizam fluido MR vem se destacando como uma solução que pode trazer benefícios de conforto e segurança aos usuários de veículos em geral. Este trabalho discute a modelagem não-linear de um veículo que considera a dinâmica vertical, lateral e longitudinal, simulado em diferentes condições de condução que buscam avaliar o conforto, a aderência à estrada, a dirigibilidade, a rolagem e a deflexão da suspensão. Pretende, também, contribuir com a área de controle de vibrações em suspensões veiculares que utilizam amortecedores MR, avaliando o desempenho dos controladores ótimo (LQR), nebuloso e FEB (Frequency-Estimation-Based ) projetados em 1/4 de veículo e aplicados ao modelo não-linear do veículo. O trabalho termina comentando as potencialidades da metodologia apresentada, discutindo as facilidades e dificuldades encontradas na sua implementação e aponta propostas para a sua continuidade. / Magnetorheological fluids (MR) are capable of changing their rheological properties when a magnetic field is applied. One of the most important applications of the MR fluid is in vibration dampers, mainly used in construction, automobiles and other mechanical systems subjected to excitations that cause unwanted vibrations. In the automotive industry, nowadays the technology of dampers using MR fluid has emerged as a solution which can bring benefits of comfort and safety to overall vehicle users. This work discusses the non-linear modeling of a vehicle which considers the vertical, lateral and longitudinal dynamics, simulated in different driving conditions aiming evaluate the comfort, the road holding, the handling, the roll and the suspension deflection. It also aims to contribute to the field of vibration control in vehicular suspensions that use magnetoreological dampers, evaluating the performance of controllers optimal (LQR), fuzzy and FEB (Frequency-Estimation-Based ) designed in 1/4 of vehicle and applied to the non-linear model of the vehicle. This work is concluded presenting the potentialities of the design methodology proposed and future developments to be implemented.

Mitigation of displacement and acceleration responses of a three story benchmark structure excited by seismic motions is pursued in this study. Multiple 20-kN magnetorheological (MR) dampers are installed in the three-story benchmark structure and managed by a global fuzzy logic controller to provide smart damping forces to the benchmark structure. Two configurations of MR damper locations are considered to display multiple-input, single-output and multiple-input, multiple-output control capabilities. Characterization tests of each MR damper are performed in a laboratory to enable the formulation of fuzzy inference models. Prediction of MR damper forces by the fuzzy models shows sufficient agreement with experimental results. A controlled-elitist multi-objective genetic algorithm is utilized to optimize a set of fuzzy logic controllers with concurrent consideration to four structural response metrics. The genetic algorithm is able to identify optimal passive cases for MR damper operation, and then further improve their performance by intelligently modulating the command voltage for concurrent reductions of displacement and acceleration responses. An optimal controller is identified and validated through numerical simulation and fullscale experimentation. Numerical and experimental results show that performance of the controller algorithm is superior to optimal passive cases in 43% of investigated studies. Furthermore, the state-space model of the benchmark structure that is used in numerical simulations has been improved by a modified version of the same genetic algorithm used in development of fuzzy logic controllers. Experimental validation shows that the state-space model optimized by the genetic algorithm provides accurate prediction of response of the benchmark structure to base excitation.

Submitted by Renata Lopes (renatasil82@gmail.com) on 2017-03-02T15:20:50Z No. of bitstreams: 1 catarinavieiranagahama.pdf: 3773576 bytes, checksum: ae1c4d72f2626e67a0fa112d4f0ee815 (MD5) / Approved for entry into archive by Adriana Oliveira (adriana.oliveira@ufjf.edu.br) on 2017-03-06T19:53:32Z (GMT) No. of bitstreams: 1 catarinavieiranagahama.pdf: 3773576 bytes, checksum: ae1c4d72f2626e67a0fa112d4f0ee815 (MD5) / Made available in DSpace on 2017-03-06T19:53:32Z (GMT). No. of bitstreams: 1 catarinavieiranagahama.pdf: 3773576 bytes, checksum: ae1c4d72f2626e67a0fa112d4f0ee815 (MD5) Previous issue date: 2013-09-10 / CAPES - Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / Sistemas de controle de vibrações estruturais são formas de proteção que visam reduzir vibrações excessivas e/ou indesejáveis induzidas por cargas dinâmicas atuantes em estruturas civis, de forma a atender os critérios de segurança, funcionalidade e conforto aos usuários. Estes sistemas podem ser classificados em passivo, ativo, híbrido e semiativo. Dentre os sistemas de controle de vibrações, os semiativos têm se mostrado bastante atrativos por aliar a confiança e a simplicidade típicas de sistemas passivos `a adaptabilidade dos sistemas ativos, destacando-se, dentre eles, os amortecedores magnetorreológicos (MR). Os amortecedores MR são dispositivos semiativos que possuem a capacidade de mudar, de maneira reversível e quando exposto a um campo magnético, de um estado líquido para um estado semi-sólido em milissegundos. Essa característica faz dos amortecedores MR uma ferramenta ideal para o controle de sistemas estruturais, pois com eles é possível administrar forças de amortecimento de forma rápida e segura, utilizando pequenas quantidades de energia. Neste trabalho, a eficiência do controle semiativo utilizando amortecedores MR aplicados a um modelo de um edifício de dois andares submetido a uma aceleração na base é estudada. Os resultados obtidos são validados através da comparação com resultados já publicados por outros autores. A metodologia adotada consiste basicamente em avaliar o comportamento dos amortecedores MR em três situações distintas: 1) funcionando como um amortecedor passivo, ou seja, aplicando-se uma voltagem constante e, portanto, sem variações de suas propriedades amortecedoras; 2) funcionando como um controlador semiativo em que a voltagem de comando dos amortecedores MR é determinada pelo algoritmo clipped optimal baseado em um regulador linear quadrático (LQR), podendo assumir o valor 0V ou voltagem máxima; e 3) funcionando como controlador semiativo com voltagem de comando otimizada, podendo assumir valores intermediários de voltagem entre 0V e voltagem máxima. Esta última estratégia é original e consiste na principal contribuição do presente trabalho. Para efeitos comparativos, o modelo estudado também foi submetido ao controle puramente ativo, supondo-se um atuador mecânico exercendo forças de controle diretamente na estrutura. Esta última estratégia permite confrontar os desempenhos do amortecedor MR com um controlador ativo. / Structural vibration control systems are means of protection which aim the reduction of excessive or undesirable vibrations caused by dynamic loads on civil structures in other to assure safety and comfort criteria. Those systems are classified into passive, active, hybrid and semi-active. Semi-active controls systems are among the most used due to reliability and simplicity reasons. The magnetorheological dampers (MR) are semi-active devices capable of changing from liquid to semisolid state in milliseconds, in a reversible manner, when exposed to a magnetic field. Due to this capability, MR dampers are considered as a perfect tool for controlling structural systems - allowing the management of damping load fast and safely, with small amount of energy. This work studies the efficiency of semi-active control systems, by means of MR dampers applied to a two-story building model subject to accelerations applied to its basis. The obtained results are compared to data available in the literature, showing good agreement. The adopted methodology consists in evaluating the MR dampers behavior in three distinct situations: 1) applied as a passive damper: under constant voltage, with no variation of dampers properties; 2) applied as a semiactive controller for which the activation voltage is determined by the clipped optimal algorithm based on a linear quadratic regulator - with voltages of 0V or a maximum value; 3) applied as a semiactive controller with optimized activation voltage, assuming voltage values ranging from 0V to a maximum value. This strategy is original and consists in the main contribution of the present work. For comparison purposes, the studied model was also subjected to purely active control, by assuming a mechanical actuator exerting control forces directly on the structure. This strategy allows confronting the performance of the MR damper with an active controller.

As Civil Engineering structures, e.g., tall buildings, long span bridges, deep water offshore platforms, nuclear power plants, etc., have become more costly, complex and serve more critical functions, the consequences of their failure are catastrophic. Therefore, the protection of these structures against damage induced by large environmental loads, e.g., earthquakes, strong wind gusts and waves, etc., is without doubt, a worldwide priority. However, structures cannot be designed to withstand all possible external loads and some extraordinary loading episodes do occur, leading to damage or even failure of the structure. Protection of a structure against hazards can be achieved by various means such as modifying structural rigidities, increasing structural damping, and by attaching external devices, known as control devices. Control devices can be deployed either to isolate the structure from external excitation or to absorb input seismic energy to the structure (absorber) so as to mitigate vibration in the primary structure. Seismic base isolation is one such mechanism which isolates a structure from harmful ground excitations. Seismic base isolation is a widely accepted and implemented structural control mechanism due to its robustness and ease in deployment. Following the Northridge earthquake (1994), and Kobe earthquake (1995), the interest of structural engineers in understanding near-source ground motions has enhanced. Documents published after these earthquakes emphasized the issue of large base displacements because of the use of none or little isolation damping (of viscous type only) prior to these events. More recent studies have investigated analytically and experimentally, the efficiency of various dissipative mechanisms to protect seismic isolated structures from recorded near-source long period, pulse-type, high velocity ground motions. Consequently, hybrid isolation systems, seismic base isolation supplemented with damping mechanisms, have become the focus of current research trend in structural vibration control. Hybrid base isolation system incorporating passive supplemental damping devices like, viscous fluid dampers, etc., performs satisfactorily in minimizing isolator displacement but at the same time increases superstructure acceleration response. Furthermore, the passive system can be tuned to a particular frequency range and its performance decreases for frequencies of excitation outside the tunning bandwidth. In such a scenario, active control devices in addition to base isolation mechanism provide better performance in reducing base displacement and superstructure acceleration for a broad range of excitation frequencies. Tremendous power requirement and the possibility of power failure during seismic hazards restrict the usage of active systems as a supplemental device. Semi-active devices provide the robustness of passive devices and adaptive nature of active devices. These characteristics make them better suited for structural control applications. The recent focus is on the development of magnetorheological (MR) dampers as semi-active device for structural vibration control applications. MR dampers provide hysteretic damping and can operate with battery power. The thrust of this thesis is on developing a hybrid base isolation mechanism using MR dampers as a supplemental damping device. The use of MR damper as a semi-active device involves two steps; development of a model to describe the MR damper hysteretic behaviour; development of a proper nonlinear control algorithm to monitor MR damper current / voltage supply. Existing parametric models of MR damper hysteretic behaviour, e.g., Bouc-Wen model, fail to consider the effect of amplitude and frequency of excitation on the device. Recently reported literature has demonstrated the necessity of incorporating amplitude and frequency dependence of MR damper models. The current/voltage supply as the input variable to the MR damper restricts the direct use of any control algorithms developed for active control of structures. The force predicted by the available control algorithms should be mapped to equivalent current/voltage and then to be fed into the damper. Available semi-active algorithms in the literature used ‘on-off’ or ‘bang-bang’ strategy for MR applications due to nonlinear current/voltage-force relation of MR damper. The ‘on-off’ nature of these algorithms neither provides smooth change in MR damper current/voltage input nor considers all possible current/ voltage values within its minimum to maximum range. Secondly, these algorithms fail to consider the effect of the MR damper applied and commanded current/voltage dynamics. The thrust of this dissertation is to develop semi-active control algorithms to monitor MR damper supply current/voltage. The study develops a Bouc-Wen based model to characterize the MR damper hysteretic phenomenon. Experimental results and modeling details have been documented. A fuzzy based intelligent control and two model-based nonlinear control algorithms based on optimal dynamic inversion and integral backstepping have been developed. Performance of the fuzzy logic based intelligent control has been explored using experimental investigation on a three storey base isolated building. Further the application of the proposed controllers on a benchmark building; a benchmark highway bridge and a stay cable vibration reduction have been discussed. Experimental study has revealed that the performance of optimal FLC is better than manually designed FLC in terms of reducing base displacement and storey accelerations. The performance of both the FLCs (simple FLC and genetic algorithm based optimal FLC) is better than ‘passive-off’ (zero ampere current supply) and ‘passive-on’ (one ampere current supply) condition of MR damper applications. The ‘passive-off’ results have shown higher base displacements with lower storey accelerations, whereas, the ‘passive-on’ results have reduced base displacement to the least but at the same time increased the storey acceleration too much. The FLC monitored MR damper show a compromise between the two passive conditions. Analytical results confirm these observations. Numerical simulations of the base isolated building with the two model based MR damper control algorithms developed have shown a better performance over FLC and widely used clipped optimal algorithms. The applications of the proposed semi-active control algorithms (FLC, dynamic inversion and integral backstepping) have shown better performance in comparison to that of control algorithms provided with the benchmark studies.

Los sistemas tales como edificios y veh¨ªculos est¨¢n sujetos a vibraciones que pueden causar mal funcionamiento, incomodidad o colapso. Para mitigar estas vibraciones, se suelen instalar amortiguadores. Estas estructuras se convierten en sistemas adaptr¨®nicos cuando los amortiguadores son controlables. Esta tesis se enfoca en la soluci¨®n del problema de vibraciones en edificios y veh¨ªculos usando amortiguadores magnetoreol¨®gicos (MR). Estos son unos amortiguadores controlables caracterizados por una din¨¢mica altamente no lineal. Adem¨¢s, los sistemas donde se instalan se caracterizan por la incertidumbre param¨¦trica, la limitaci¨®n de medidas y las perturbaciones desconocidas, lo que obliga al uso de t¨¦cnicas complejas de control. En esta tesis se usan Backstepping, QFT y H2/H¡Þ mixto para resolver el problema. Las leyes de control se verifican mediante simulaci¨®n y experimentaci¨®n. / Buildings and vehicle systems are subject to vibrations that may cause malfunctioning, discomfort or collapse. It is an extended practice to install damping devices in order to mitigate such vibrations. With controllable dampers, structures act as adaptronic systems. This dissertation focuses on solving the vibration mitigation problem in buildings and vehicles making use of magnetorheological (MR) dampers which are controllable devices characterized by a highly nonlinear dynamics. Additionally, the systems where they are installed, are characterized by parametric uncertainties, limited measurement availability and unknown disturbances. This implies the use of complex control techniques in order to get a reliable performance of the control system. This research makes use of Backstepping, QFT and Mixed H2/H¡Þ control techniques for achieving the proposed goal. These are verified thorugh simulations and experimentation.

Magnetorheological (MR) dampers are adaptive devices whose properties can be adjusted through the application of a controlled voltage signal. A semi-active suspension system incorporating MR dampers combines the advantages of both active and passive suspensions. For this reason, there has been a continuous effort to develop control algorithms for MR-damped vehicle suspension systems to meet the requirements of the automotive industry. The overall aims of this thesis are twofold: (i) The investigation of non-parametric techniques for the identification of the nonlinear dynamics of an MR damper. (ii) The implementation of these techniques in the investigation of MR damper control of a vehicle suspension system that makes minimal use of sensors, thereby reducing the implementation cost and increasing system reliability. The novel contributions of this thesis can be listed as follows: 1- Nonparametric identification modelling of an MR damper using Chebyshev polynomials to identify the damping force from both simulated and experimental data. 2- The neural network identification of both the direct and inverse dynamics of an MR damper through an experimental procedure. 3- The experimental evaluation of a neural network MR damper controller relative to previously proposed controllers. 4- The application of the neural-based damper controller trained through experimental data to a semi-active vehicle suspension system. 5- The development and evaluation of an improved control strategy for a semi-active car seat suspension system using an MR damper. Simulated and experimental validation data tests show that Chebyshev polynomials can be used to identify the damper force as an approximate function of the displacement, velocity and input voltage. Feed-forward and recurrent neural networks are used to model both the direct and inverse dynamics of MR dampers. It is shown that these neural networks are superior to Chebyshev polynomials and can reliably represent both the direct and inverse dynamic behaviours of MR dampers. The neural network models are shown to be reasonably robust against significant temperature variation. Experimental tests show that an MR damper controller based a recurrent neural network (RNN) model of its inverse dynamics is superior to conventional controllers in achieving a desired damping force, apart from being more cost-effective. This is confirmed by introducing such a controller into a semi-active suspension, in conjunction with an overall system controller based on the sliding mode control algorithm. Control performance criteria are evaluated in the time and frequency domains in order to quantify the suspension effectiveness under bump and random road excitations. A study using the modified Bouc-Wen model for the MR damper, and another study using an actual damper fitted in a hardware-in-the-loop- simulation (HILS), both show that the inverse RNN damper controller potentially gives significantly superior ride comfort and vehicle stability. It is also shown that a similar control strategy is highly effective when used for a semi-active car seat suspension system incorporating an MR damper.