RCAAP Repository

Conselho Nacional de Desenvolvimento Científico e Tecnológico

Fundação de Amparo a Pesquisa do Estado de Minas Gerais

Fundação de Amparo a Pesquisa do Estado de Minas Gerais

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior

Conselho Nacional de Desenvolvimento Científico e Tecnológico

Conselho Nacional de Desenvolvimento Científico e Tecnológico

Conselho Nacional de Desenvolvimento Científico e Tecnológico

This dissertation deals with the dynamic modeling of flexible multibody systems subjected to viscoelastic damping for the purpose of passive vibration control. Accounting for the fact that the large majority of previous research work has been focused on viscoelastic vibration control of fixed-geometry structures, the main goals of this essay are: i) to integrate, sistematically, the formulation pertaining to the various aspects addressed in the derivation of the numerical models of viscoelastic flexible multibody systems, namely: Lagrangean approach as applied to flexible multibody systems, finite element discretization, viscoelastic constitutive models, algorithms for numerical solution of differential-algebraic systems of equations; ii) to implement and validate computer codes intended for the dynamics simulation of flexible multibody systems containing viscoelastic elements; iii) to appraise, through numerical simulations, the effectiveness of viscoelastic treatments in terms of vibration mitigation of two types of mechanical systems, namely: closed-chain plane mechanisms and spacecraft containing flexible appendages. A constitutive law based of fractional derivatives is considered for the modeling of the viscoelastic behavior. Both the Finite Element Method, associated to floating reference frames, and the Assumed Modes Method are used to perform spatial discretization of the equations of motion. Numerical simulations are accomplished for two different multibody systems: a plane flexible four bar linkage, and an artifitial satellite containing flexible appendages and attitude control based on a reaction wheel. The results obtained from numerical integration of the nonlinear dynamic equations of motion enabled to evaluate the influence of the components compliance on the dynamic responses. They also prove that viscoelastic treatments can be effectively implemented.

Conselho Nacional de Desenvolvimento Científico e Tecnológico

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior

Nowadays, one of the main problems faced by oil exploitation industries is the formation of cracks in offshore production units. In these classified areas, the Friction Hydro Pillar Processing has some advantages when compared to the arc-voltaic welding, which is widely used to perform these repairs. Some of them are the lower temperatures during the process and also the reduction of the surroundings influences in the quality of these repairs. This work presents the results concerning the optimization of some important parameters of the process performed by a Friction Pillar Processing unit able to apply axial loads up to 245 kN. This technique consists of filling a hole, through the use of a consumable rod under a controlled rotation speed and pressed against the hole. The friction between the surfaces increases their temperatures reducing the yield strength of the materials involved allowing a metallurgical bond between the parts. Among the process parameters to be optimized there are: axial load, rotation speed and burn length. For this, the PID control techniques were applied and a specific study of the controller boards was also accomplished. After this, some tests were performed to evaluate proposed solutions and also to identify the operational ranges of the equipment. The performed test matrices allowed the identification of these ranges and also validated the new optimal parameters proposed in this work. Moreover, the mechanical properties of each test were evaluated by means of hardness tests and microstructure analysis.

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior

Fundação de Amparo a Pesquisa do Estado de Minas Gerais

Conselho Nacional de Desenvolvimento Científico e Tecnológico

One of the main difficulties of Active Noise Control (ANC) in Ducts is to develop a methodology that can achieve control in a wide frequency range. However, most studies of ANC are designed to tonal signal and discrete frequencies. When studding a wide range of frequencies, some of these specific frequencies are not achieving the desired control satisfactorily. Therefore this dissertation was developed to meet the meaning of why these particular frequencies are not achieving a satisfactory level of control. Through a bench trial ANC system using an acoustic duct and a signal generator, it was possible to obtain the FRF (Frequency Response Function) and from it analyze and compare which frequencies were working in the areas of Anti-resonance (resulting in poor control) and which were acting in resonance regions (resulting in good control). Optimizing the sensors location receivers (microphones) could reach a larger number of discrete frequencies controlled satisfactorily on a single project CAR. Results, discussion and conclusions are described in this dissertation.

Conselho Nacional de Desenvolvimento Científico e Tecnológico

Conselho Nacional de Desenvolvimento Científico e Tecnológico

Conselho Nacional de Desenvolvimento Científico e Tecnológico

Conselho Nacional de Desenvolvimento Científico e Tecnológico

The main purpose of this work is the evaluation of the interaction of the spatial schemes (Traditional finite differences and DRP) with temporal schemes (Euler, Runge-Kutta 2nd and 4th order methods and the optimized methods: LDDRK and RK46-NL) used on computational aeroacoustic, on 1D (one dimension) acoustic propagation prediction. A secondary goal is to present details of the traditional finite differences construction, based on Taylor series, and the DRP, proposed by Tam and Webb (1993). To the best combination of spatial and temporal schemes will be applied non-reflexive boundary conditions on the eddy of computational domain. 2D (two dimensions) simulations will also contain radiation and outflow boundary conditions proposed by Tam and Webb (1993). The necessity to use high order methods to minimize the dispersion effects was verified. This work shows an outstanding efficiency of Runge-Kutta 2nd order over Runge-Kutta 4th order scheme, and in some cases, even superior to the optimized schemes. The simulations revealed that DRP loss efficiency over traditional finite differences scheme when grid refinement occurs. An study of non-reflexive boundary conditions construction is recommended for this work complementation.