This work details the development and implementation of a numerical model capable of solving strongly-coupled fluid-structure interaction problems involving long thin structures, which are common multi-physics problems encountered in many applications. In most fluid-structure interaction problems the deformation of the slender elastic bodies is significant and cannot be described by a purely linear analysis. We present a new formulation to model these larger displacements. By extending the standard modal decomposition technique for linear structural analysis, the governing equations and boundary conditions are updated to account for the leading-order non-linear terms and a new modal formulation with quadratic modes is derived. The quadratic modal approach is tested on standard benchmark problems of increasing complexity and compared with analytical and full non-linear numerical solutions. Two computational fluid-structure interaction approaches are then implemented in a partitioned manner: a finite volume method for discretisation of both the fluid and solid domains and the quadratic modal formulation for the structure coupled with a finite volume fluid solver. Strong-coupling is achieved by means of a fixed-point solver with dynamic relaxation. The fluid-structure interaction approaches are validated and compared on benchmark problems of increasing complexity and strength of coupling between the fluid and solid domains. Fluid-structure interaction systems may become unstable due to the interaction between the fluid-induced pressure and structural rigidity. A thorough stability analysis of finite elastic plates in uniform flow is conducted by varying the structural length and flow velocity showing that these are critical parameters. Validation of the results with those from analytical methods is done. An analysis of the dynamic interactions between multiple finite plates in various configurations is also conducted.

Made available in DSpace on 2014-06-12T17:38:35Z (GMT). No. of bitstreams: 2 arquivo3894_1.pdf: 2490981 bytes, checksum: ae74a5ee01e3de583e61526f3090adf2 (MD5) license.txt: 1748 bytes, checksum: 8a4605be74aa9ea9d79846c1fba20a33 (MD5) Previous issue date: 2009 / Este trabalho estuda o sistema de amplificação de onda de uma máquina de usinagem por abrasão ultrassonora por meio de modelos de elementos finitos implementados nos programas computacionais MSC Patran e ANSYS. O sistema é constituído por um amplificador fixado ao gerador de ondas de 20 kHz, um amplificador intercambiável de perfil exponencial (sonotrodo) e uma ferramenta que atua nas partículas abrasivas contra a superfície do material a ser usinado. Neste estudo, considera-se uma ferramenta de casca e miolo cilíndricos, com aproximadamente 60 g e 25 mm de comprimento, que permite a obtenção de estruturas monolíticas de quartzo usadas como sensores de pressão em ambientes severos. O comportamento do sistema de amplificação foi analisado em relação às frequências naturais de vibração (análise modal) e à amplitude na extremidade da ferramenta (análise harmônica). Para permitir agilidade e rapidez no processamento dos modelos, mostrou-se inicialmente que a simetria axial do conjunto sonotrodo-ferramenta poderia ser empregada na concepção dos modelos numéricos. A consistência dos modelos numéricos do sistema de amplificação foi verificada por meio da determinação analítica e numérica das frequências naturais dos modos longitudinal, torcional e transversal de uma viga cilíndrica modelizada na condição de contorno engastado-livre. Optou-se também por otimizar o número de elementos dos modelos e de restringir os modos de vibração indesejáveis. A partir do cálculo das primeiras frequências naturais do modo longitudinal de modelos concebidos com diferentes ligas de titânio, verificou-se que a frequência do terceiro harmônico de um dos modelos (19.74 kHz) encontra-se 0,5 % acima da frequência de operação do sistema de amplificação do equipamento. As análises harmônicas mostraram que o sistema de amplificação composto pelo sonotrodo de perfil exponencial opera no terceiro harmônico do modo longitudinal com um fator de amplificação de oito vezes. Diante desses resultados, conclui-se que os tempos excessivos de usinagem anteriormente vivenciados na obtenção de protótipos dos sensores de pressão, com o sistema de amplificação em estudo, não podem ser atribuídos à amplificação da onda mecânica na extremidade da ferramenta. Os longos períodos de usinagem podem estar relacionados à defasagem observada entre a frequência nominal do equipamento e a frequência de operação do sistema (19.65 kHz), causada pela grande massa da ferramenta. Em função desses resultados, um novo perfil de sonotrodo foi proposto de tal forma que o sistema de amplificação modelizado apresentasse frequência natural de 20 kHz e com mesmo nível de amplificação. A análise harmônica deste novo modelo, constituído por um sonotrodo de perfil cônico composto, também serviu para verificar que a distribuição das tensões ao longo do sistema de amplificação não excede, segundo o critério de Von Mises, a tensão de escoamento da liga de titânio considerada quando o mesmo é excitado com uma carga estática de 25 N

Dissertação (mestrado)—Universidade de Brasília, Faculdade de Tecnologia, Departamento de Engenharia Mecânica, 2018. / Em busca de melhor aproveitamento da energia eólica disponível, a demanda em relação a projetos de torres eólicas é por torres cada vez mais altas e pás maiores. Pás de torres eólicas são estruturas flexíveis que estão sujeitas a altas velocidades do vento e restritas pela movimentação da torre. As principais forças aplicadas no sistema são do tipo inercial, gravitacional e aerodinâmica, que são responsáveis pelas vibrações da pá. Além disso, o acoplamento das vibrações da pá e da torre leva à um sistema com elevado grau de complexidade. A torre, em função de suas dimensões, pode ser modelada como uma viga esbelta e pode estar sujeita a vibrações excessivas, que podem causar acidentes. A primeira frequência natural da torre está contida na faixa de atuação das pás, então um estudo do seu acoplamento com as pás é necessário. Este trabalho foca no estudo da resposta dinâmica da interação torre-pá através de modelos dinâmicos discretos. O modelo discreto considera efeitos de peso-próprio e de força centrífuga devido à rotação das pás. Evidencia-se a relação entre posição do rotor e frequência de resposta do sistema e, também, entre velocidade de rotação e frequência de resposta do sistema. / To better exploit available wind energy, the demand for wind tower design is for increasingly higher towers and larger blades. Wind tower blades are flexible structures that are subject to high forcing velocities and restricted by tower movement. The main forces applied in the system are inertial, gravitational and aerodynamic, which are responsible for blade vibrations. In addition, coupling vibrations of the blade and the tower leads to a system with a high degree of complexity. The tower, depending on its size, is considered a slender beam and may be subject to excessive vibrations, which may cause accidents. The first natural frequency of the tower is contained in the operating range of the blades, so a study of their coupling with the blades is necessary. This work presents a study on dynamic response of tower-blade interaction through discrete dynamical models. The discrete model considers self-weight effects and centrifugal force due to blade rotation. It shows the relationship between rotor position and system frequency response and also between rotation speed and frequency response of the system.

Dissertation presentée pour obtenir le grade de Docteur en Sciences Techniques, à l'École Polytechnique Fédérale de Lausanne

Tese de doutoramento. Engenharia Civil. Faculdade de Engenharia. Universidade do Porto. 2010

Tese de doutoramento. Engenharia Civil. Universidade do Porto. Faculdade de Engenharia. 2011

In April of 2008, the Federal Highway Administration (FHWA) launched the Long Term Bridge Performance (LTBP) program. The program was established to collect scientific quality data from a number of bridges across the nation over a period of 20 years. The data will be used to provide a better picture of bridge health and structural performance. Utah Department of Transportation (UDOT) structure number 1F 205, located 2.4 km (1.5 mi) west of Perry, Utah, was selected as one of the LTBP pilot bridges (this bridge will also be referred to as the Cannery Street Overpass). UDOT performs regular maintenance on this bridge and in April of 2011 they began a rehabilitation project over a 13-km (8-mi) section of I-15 that included the Cannery Street Overpass. The main purpose of this rehabilitation was to improve pavement conditions. As part of this work, in the fall of 2011 UDOT removed all of the asphalt from the bridge deck, performed deck repairs, and placed a new asphalt layer. A unique opportunity presented itself to better understand the relationship between the mass and resonant vibration frequencies of the structure. This relationship is understood by (omega_n)^2=k/m, where omega_n=resonant frequency; k=stiffness; and m=mass. A decrease in mass should yield an increase in resonant frequency. Dynamic testing was done on the bridge to obtain its resonant frequencies. This testing included measuring the velocity response of the structure at different points on the bridge due to ambient vibrations (mainly from traffic). Three tests were performed before, during, and after UDOT's scheduled maintenance. These testing states include: State 1. Original asphalt on bridge deck State 2. No asphalt on bridge deck State 3. New asphalt on bridge deck These three states represent three different mass states of the bridge. The original asphalt layer was substantially heavier than the new asphalt layer. The data obtained from all three tests was processed in order to extract modal properties of the bridge. The changes in modal properties were analyzed and the results of the testing proved to be insightful at defining the relationship between mass and resonant frequency.

Building smart home environments which automatically or semi-automatically assist and comfort occupants is an important topic in the pervasive computing field, especially with the coming of cheap, easy-to-install sensors. This has given rise to the indispensable need for human activity recognition from ubiquitous sensors whose purpose is to observe and understand what occupants are trying to do from sensory data. The main approach to the problem of human activity recognition is a probabilistic one so as to handle the complication of uncertainty, the overlapping of human behaviours and environmental noise. This thesis develops a probabilistic model as a framework for human activity recognition using multiple multi-modal sensors in complex pervasive environments. The probabilistic model to be developed is adapted and based on the abstract hidden Markov model (AHMM) with one layer to fuse multiple sensors. The concept of factored state representation is employed in the model to parsimoniously represent the state transitions for reducing the number of required parameters. The exact method is used in learning the model’s parameters and performing inference. To be able to incorporate a large number of sensors, several more parsimonious representations including the mixtures of smaller multinomials and sigmoid functions are investigated to model the state transitions, resulting in a reduction of the number of parameters and time required for training. / We examine the approximate variational method to significantly reduce the time required for training the model instead of using the exact method. A system of fixed point equations is derived to iteratively update the free variational parameters. We also present the factored model in the case where all variables are continuous with the use of the conditional Gaussian distribution to model state transitions. The variational method is still employed in this case to speed up the model’s training process. The developed model is implemented and applied in recognizing daily activity in our smart home and the Nokia lab from multiple sensors. The experimental results show that the model is appropriate for fusing multiple sensors in activity recognition with a reasonable recognition performance.

The convenience of Operational Modal Analysis (OMA), over conventional Experimental Modal Analysis (EMA), has seen to its increasing popularity over the last decade for the purpose of evaluating dynamic properties of structures. OMA features an advantage of requiring only output information, which is in tandem with its main drawback of lacking scaled modeshape information. While correctly scaled modeshapes can be assumed under a restrictive assumption of spectrally white inputs, in reality, input spectra are at best broadband in nature. In this thesis, an OMA method for Multiple-Input Multiple-Output (MIMO) applications in mechanical structures is developed. The aim is to separate MIMO responses into a collection of Single-Input Single-Output (SISO) processes (matrix FRF) using cepstral-based methods, under less restrictive and hence more realistic coloured broadband excitation. Existing cepstral curve-fitting techniques can be subsequently applied to give regenerated FRFs with correct relative scaling. This cepstral-based method is based on the matrix Mean Differential Cepstrum (MDC) and operates in the frequency domain. Application of the matrix MDC onto MIMO responses leads to a matrix differential equation which together with the use of finite differences, directly solves or identifies the matrix FRF in a propagative manner. An alternative approach based on whitened MIMO responses can be similarly formulated for the indirect solution of the matrix FRF. Both the direct and indirect approaches can be modified with a Taylor series approximation to give a total of four propagative solution sequences. The method is developed using relatively simple simulated and experimental systems, involving both impulsive and burst random excitations. Detailed analysis of the results is performed using more complicated Single-Input Multiple-Output (SIMO) and MIMO systems, involving both driving and non-driving point measurements. The use of the matrix MDC method together with existing cepstral curve-fitting technique to give correct relative scaling is demonstrated on a simulated MIMO system with coloured inputs. Accurate representation of the actual FRFs is achieved by the matrix MDC technique for SIMO set-ups. In MIMO scenarios, excellent identification was obtained for the case of simulated impulsive input while the experimental and burst random input cases were less favourable. The results show that the matrix MDC technique works in MIMO scenarios, but possible noise-related issues need to be addressed in both experimental and burst random input cases for a more satisfactory identification outcome.

This thesis presents a research work based on an integrated multi-modal approach for sports video indexing and retrieval. By combining specific features extractable from multiple (audio-visual) modalities, generic structure and specific events can be detected and classified. During browsing and retrieval, users will benefit from the integration of high-level semantic and some descriptive mid-level features such as whistle and close-up view of player(s). The main objective is to contribute to the three major components of sports video indexing systems. The first component is a set of powerful techniques to extract audio-visual features and semantic contents automatically. The main purposes are to reduce manual annotations and to summarize the lengthy contents into a compact, meaningful and more enjoyable presentation. The second component is an expressive and flexible indexing technique that supports gradual index construction. Indexing scheme is essential to determine the methods by which users can access a video database. The third and last component is a query language that can generate dynamic video summaries for smart browsing and support user-oriented retrievals.