Dissertations / Theses on the topic 'Shaft motion'

Condition monitoring of rotating shafts is gaining importance in industry due to the need to increase machine reliability and decrease the possible loss of production due to machine breakdown. In this work, the use of vibration signals for the detection of a crack within a shaft was investigated. The research involved the measurement of vibration signals during laboratory tests on a long rotating shaft rig. The focus of the experimental work was on the effect of cracks on the dynamics and the initiation and growth of cracks in the shaft. Measurements were taken from the shaft system both with simulated cracks (notches) cut at 45° and 90° to the shaft axis and with real propagating cracks initiated by a pre-crack cut. All defects were located at the mid- point along the shaft. The vibration responses and stresses were measured for different depths of crack. The vibration responses of the three different defects were compared using PSDs of the data to identify the change in position and magnitude of the peaks in the spectrum under each defect. Experiments to study the effect of defect depth at different shaft rotation speeds were also carried out. Finally, a shaft with a breathing crack (continuously opening and closing as the shaft rotates) was also studied experimentally, with the crack growing under normal steady state operating conditions. After completing the experiment work, the shaft was broken and the type of fracture studied. The results for both simulated and actual crack growth showed that vibration frequencies decreased as a crack progressed, indicating the possibility of using the vibration signal for crack detection. A significant relationship was found between the stage of crack growth and the vibration results. A finite element (FE) model was constructed to explore the relationship between the natural frequencies and crack depth and position along the shaft and to explain and validate the results of the experimental work. The FE model showed similar trends to the experimental results and also allowed the effect of different crack positions to be explored. The PSD data was fed into an artificial neural network after a feature extraction procedure was applied to significantly reduce the quantity of data whilst at the same time retaining the salient information. Such an approach results in a considerably reduced training time for the network due to the reduced complexity. The proposed scheme was shown to successfully identify the different defect levels. This method greatly enhances the capacity of an automated diagnostic process by linking increased capability in signal analysis to the predictive capability of the artificial neural network.

Most tilting pad journal bearing dynamic characteristics estimation methodologies assume perfect shaft tracking by the pads. In other words, they neglect pivot friction. In case of pads having point or line contact that operate under most normal load conditions, the pad tilting is due to a rocking motion which is not greatly influenced by friction. Hence this simplifying assumption might be acceptable. Heavier loading conditions, such as those typically encountered in gearboxes, demand the use of spherical pivots to avoid pivot failure. The spherical pivot is very attractive for this reason, but the tilting motion is rather a sliding action that must occur in the precision ball socket. A valid concern exists for verifying the soundness of assumed shaft tracking by the pads of such bearings. A â fixed test bearing, floating shaftâ type of test rig previously built for determining the dynamic characteristics of bearings was accordingly modified to facilitate the testing of shaft tracking for a spherical pivot bearing. This thesis describes the modifications carried out on the rig. The special instrumentation and data acquisition systems implemented to observe the minute pad motion are also discussed. Some preliminary results of the tests are presented for various loading conditions. They show excellent shaft tracking by the pads. More detailed testing and analysis of data is required to fully understand the pad motion and tracking ability of the spherical pivot design.
Master of Science

Diploma thesis focuses on measurement and analysis of shaft motion, torque, angular velocity and vibration. First part of the thesis deals with general issue of acquiring a digital signal. Next part describes suitable sensors for each measurement type, and data acquisition hardware. The last theoretical part describes methods for measurement data analysis and vibration diagnostics. Practical part of the thesis describes shaft motion and torque measurements made on laboratory equipment, and vibration measurement made on real machine system. Each measurement includes measurement data analysis and evaluation.

The dynamic behavior of a cracked shaft was studied both in run-up and run-down transient motion. A mathematical model of the shaft was developed , and experimental tests were performed in order to validate the model. A experimental testing rig was conducted, containing an horizontal flexible rotor with a rigid disc in the mid point between the bearings. Experimental and theo retical responses were compared for both cracked and uncracked shafts. All tests were done for several acceleration rates and unbalance conditions. For the theoretical study of dynamic response of the cracked shaft, the stiffness matrix for a cracked element was deduced, using the modified crack Model of Mayes. The rotor model was obtained by using the finite elements method. The equations of motion were integrated in the time domain by using Newmark method to obtain the transient response. Comparing theoretical and experimental results validated the model. For this purpose, the modified Mayes model was used to represent the dynamic response of the crached shaft. Further simulations were conducted to study the influence of acceleration rates, unbalance amplitude /orientation and crack severity on the response of a cracked shaft, running on transient mo tion.
Neste trabalho apresenta -se um estudo do comportamento dinâmico de um rotor com trinca no regime transiente, tanto na parada como na partida. Foi desenvolvido um modelo matemático -computacional de um rotor com trinca e foram também realizados testes experimentais com vistas à validação do modelo. Para o estudo experimental, foi construida uma bancada constituida por um rotor flexível horizontal com um disco rígido montado na posição média do comprimento. A resposta experimental do rotor foi comparada para os casos com e sem trinca. Para tanto, foram usados dois eixos, um com trinca e, o outro, sem trinca. Os testes experimentais foram feitos para diferentes taxas de aceleração e condições de desbalanceamento. Para o estudo teórico da resposta dinâmica do rotor com trinca, foi determinada a matriz de flexibilidade para um elemento com falha, uma vez aplicado o modelo de trinca de Mayes modificado. O modelo do rotor foi obtido através do método dos elementos finitos, e para o cálculo da resposta dinâmica no regime transiente foi utilizado o método de integração numérica de Newmark. O modelo foi validado mediante a comparação dos resultados teóricos com os experimentais, tendo sido mostrado que o Modelo de Mayes modificado representa adequadamente o comportamento dinâmico de um rotor com trinca. Foram feitas simulações para estudar a influência das taxas de aceleração, da intensidade do desbalanceamento presente no sistema e da severidade da trinca sobre o comportamento do rotor com trinca no regime transiente.
Mestre em Engenharia Mecânica

This dissertation thesis consists rotordynamics of modern turbocharger. First part begins with prediction of critical speed, prediction of onset instability of oil whirl and oil whip by XLTRC2 and comparison with measured data (Cascade diagram, shaft motion, FFT analysis). List of measurement method for the detection of the natural frequency of turbocharger (EMA). Create software for long term monitoring and recording and output data size reduction. Detection of critical speed by defined measurement methodology without using software simulation with measured data from the Cascade diagram, move the rotor shaft motion in the bearing, FFT analysis and results from measurements of natural frequencies. Comparison of predicted data (critical speed, prediction of onset instability) program XLTRC2 with the values measured by this approach measurement (waterfall diagram, shaft motion, FFT analysis)

The contributions of this dissertation are in the development of two new interrelated approaches to video data compression: 1) A level-refined motion estimation and subband compensation method for the effective motion estimation and motion compensation. 2) A shift-invariant sub-decimation decomposition method in order to overcome the deficiency of the decimation process in estimating motion due to its shift-invariant property of wavelet transform. The enormous data generated by digital videos call for an intense need of efficient video compression techniques to conserve storage space and minimize bandwidth utilization. The main idea of video compression is to reduce the interpixel redundancies inside and between the video frames by applying motion estimation and motion compensation (MEMC) in combination with spatial transform coding. To locate the global minimum of the matching criterion function reasonably, hierarchical motion estimation by coarse to fine resolution refinements using discrete wavelet transform is applied due to its intrinsic multiresolution and scalability natures.

Nondestructive methods of detecting cracks in structural components and machinery are important, both in preventing failures and in establishing maintenance procedures. This thesis considers how the vibration behavior of cracked members can be modelled mathematically and how these mathematical models may lead to advancements in crack detection procedures. Two separate cases are considered: the longitudinal vibration of a cracked bar and the coupled vibrations of a cracked rotating shaft. In the longitudinal vibration study, the equation of motion is developed for a cantilevered bar with a symmetric surface crack. Next, Galerkin's Method is used to obtain one- and two-term approximate solutions. Both forced and free vibrations of the bar are analyzed. Graphical results showing the relationships between displacement and crack size, crack position, and forcing frequency are presented and discussed. Spectral analysis is used to compare uncracked and cracked bar behavior. Finally, a sensitivity analysis of the forced vibration case is conducted to observe how the forcing frequency affects the rate of change of steady-state response at the onset of cracking. In the second part of the thesis, a similar analysis is conducted for a cracked, simply-supported Timoshenko shaft rotating at a constant angular speed. The equations of motion derived by Wauer (b) are used as the basis of the study. Again, Galerkin's Method is applied to obtain approximate solutions. Time histories and spectra are used to observe how changes in various parameters influence the vibration behavior. The effects of mass eccentricity and gravity are studied. Finally, the effect of a periodic axial impact load is considered.
Master of Science

La localisation des objets dans l'espace est l'une des fonctions centrales du système visuel. Lorsqu'un observateur ou une cible se déplace, le mouvement de l’œil ou de l'objet peut être pris en compte pour calculer la position de l'objet à n'importe quel moment. Il a été démontré à plusieurs reprises que le mouvement visuel peut fortement influencer la position perçue d'un objet. Par exemple, la position d'un stimulus contenant une texture en mouvement (De Valois & De Valois, 1991; Ramachandran & Anstis, 1990), celle d'un flash présenté sur (Cavanagh & Anstis, 2013) ou à côté (Whitney & Cavanagh, 2000) d'une texture en mouvement, et même les positions d'apparition et de disparition d'objets en mouvement (Fröhlich, 1923) sont perçus comme étant déplacées dans le sens du mouvement. Dans le cadre de cette thèse, nous avons exploré la relation entre ces changements de position provoqués par le mouvement et différentes formes d'attention visuelle: 1) l'attention spatiale temporaire, 2) l'attention globale et locale, 3) l'attention spatiale maintenue, et 4) l'attention centrée sur les objets. Dans une première série d'expériences, nous avons étudié dans quelle mesure et de quelle manière l'attention module l'effet Fröhlich (i.e la position perçue du début du mouvement est déplacée dans le sens du mouvement). Dans les expériences 1 et 2, nous avons mesuré l'effet Fröhlich dans différentes conditions de d'indiçage et nous avons montré que les indices non valides ou tardifs entraînent des déplacements perceptifs plus importants. Dans l'expérience 3, nous avons comparé les déplacements provoqués par le mouvement lorsque les sujets portaient leur attention sur un ensemble d'objets en mouvement qui formaient un groupe à quand ils portaient leur attention à un seul stimulus parmi cet ensemble. Les résultats montrent que l'effet Fröhlich n'était présent que lorsque l'attention sélectionne un seul objet et qu'il disparaît lorsque le stimulus est perçu dans un contexte de groupe (global). Ainsi, ces résultats suggèrent que l'attention sélective est à la fois à l'origine de et module l'effet Fröhlich. Ayant établi que les délais temporels de l'attention augmentent les changements de la position provoqués par le mouvement, l'étude suivante était conçue pour explorer si la distribution spatiale de l'attention a le même effet. Dans cette étude, nous avons utilisé le « flash grab » - un changement de position illusoire aperçu quand une cible est brièvement présentée brièvement sur un fond en mouvement au moment de l'inversion du sens du mouvement (Cavanagh & Anstis, 2013). Les essais étaient regroupés en différents blocks au début desquels un indice indiquait la région de l'espace dans laquelle la cible était susceptible d'apparaître. Nos résultats mettent en évidence une diminution du flash-grab quand la distribution spatiale des cibles était limitée à une plage de 90° ou moins. Dans la dernière étude, nous nous sommes demandés si l'effet du mouvement sur la position perçue affecte l'objet dans son ensemble ou si il affecte les caractéristiques distinctes d'un même objet indépendamment les uns des autres. Pour cela nous avons utilisé le paradigme flash grab en présentant brièvement une forme sur un fond en mouvement au moment de l'inversion du sens du mouvement. Les résultats indiquent que les attributs de la cible orthogonaux au mouvement du fond étaient déplacés, alors que les attributs parallèles au mouvement restaient intacts. Ceci suggère que le mouvement interagit avec la position des attributs de l'objet (et que l'attention focale sélectionne) avant que ces attributs ne soient regroupés en un objet. En conclusion, nous avons utilisé une variété de manipulations attentionnelles à des changements de position provoqués par le mouvement pour étudier le lien entre l'amplitude de l'illusion et les caractéristiques attentionnelles utilisé
Localizing objects in space is one of the central functions of the visual system. When an observer or a target is moving, the motion of the eye or the object can be taken into account to compute the current object locations. It has been shown many times that visual motion can strongly influence the perceived position of an object. For example, a stationary patch containing moving texture (De Valois & De Valois, 199; Ramachandran & Anstis, 1990), a flash presented on (Cavanagh & Anstis, 2013) or next to (Whitney & Cavanagh, 2000) a moving texture, and even the onset and offset positions of the moving targets (Fröhlich, 1923) are perceived as shifted in the direction of motion. In this thesis we explore the relationship between these motion-induced position shifts and visual attention in the following forms: 1) transient spatial attention, 2) global and local attention, 3) sustained spatial attention, and 4) object-based attention. In the first series of experiments we looked at whether and how attention modulates the shift in localization of motion onset (Fröhlich effect). In Experiments 1 and 2 we measured Fröhlich effect under different cueing conditions and established that invalid or late cues produced larger perceptual shifts. In Experiment 3 we compare the motion-induced shifts when the subjects attended to a set of moving stimuli as a group and when they attended to an orientation singleton. We showed that the Fröhlich effect was only present when the target was individuated and disappeared when the stimulus was perceived globally. Thus, the Fröhlich effect appeared to be both produced and modulated by focal attention. Having established that temporal delays of attention increase motion-induced position shifts, the next study explored if spatial distribution of attention has a similar effect. In this study we used flash grab - an illusory position shift seen when a target is briefly flashed on top of a moving background that abruptly changes direction (Cavanagh & Anstis, 2013). Trials were presented in blocks and before each block a cue indicated a range of possible target locations. We found that the flash grab was reduced if the spatial distribution of targets was limited to a range of 90° or less. The final study asked whether motion shifts the perceived position of an object as a whole or if separate features of a single object are shifted independently. To test this we used the flash grab paradigm and briefly presented a shape on top of a moving background at the moment it changed direction. The results showed that the features of the target that were orthogonal to the background motion were shifted, whereas the features parallel to the motion were intact. This suggests that motion interacts with the position of the object's features (and focal attention selects them) before they are bound together into an object. In conclusion, we applied a variety of attentional manipulations to motion-induced position shifts, and examined the link between the strength of the illusion and the characteristics of attention used in a particular task. We found that 1) motion-induced position shifts require focused attention and the possibility to track an individual motion trajectory; 2) allowing attention to be allocated more efficiently in space reduces the illusion; and 3) motion-induced shifts operate on the feature-based and not object-based level

The involvement of dynamical fluctuations in driving enzymatic processes is widely accepted. With respect to NQM tunnelling enzymes, the role of promoting motions in facilitating hydrogenic transfers is well studied. Few studies have however, specifically attributed, dedicated dynamical fluctuations characterised by their timescales and magnitudes, as a function of a reaction coordinate, to specific groups in a protein system. An effectively full suite of backbone resonance assignments were obtained for PETNR and on relevant ligand complexes. This provided an essential platform on which residue specific, backbone amide fluctuations were assessed. This thesis documents the application of pressure up to 1500 bar, in tandem with high resolution TROSY based NMR analysis, as a means of studying residue specific, conformer exchange perturbations. Residue specific amide compression profiles of the PETNR:FMN free enzyme system, and complexes with progesterone and tetrahydropyridine dinucleotides have been obtained. The binding of progesterone appears to induce conformational tightening of residues within the active site vicinity. The complexation of PETNR:FMN with tetrahydropyridine dinucleotides, appears to stimulate conformational shifts towards intermediate, and in some cases, slow exchange regimes in multiple residues about the active site vicinity. This is evidenced by extensive intensity attenuation of 1H-15N TROSY resonances, on the binding of tetrahydropyridine dinucleotides at 1 bar pressure, and on going from 1 bar to 1500 bar pressure. Multiple regions of sequence, spatially clustering about the active site vicinity within a 10 Å sphere of the FMN binding pocket, display appreciable sensitivity to ligand binding. Differential responses of residues to the application of high pressure between complexes was noted within segments of these regions. A region of sequence, named the β-hairpin flap displays significant differential compression profiles between the PETNR:FMN free enzyme system, and associated progesterone and tetrahydropyridine dinucleotide complexes. A role in mediating ligand engagement is proposed for R130 and R142 in the β-hairpin flap. A central hydrogen bonding network, perhaps constituting a putative proton wire in the active site of the PETNR:FMN:Progesterone complex, has been identified that could enable the shuttling of protons following catalytic protonation of oxidative substrate. The resonance response behaviour of G185 acts as a sensitive reporter on the formation of these interactions, revealed by an interrogation of the differences in chemical shift changes on progesterone binding, and in response to high pressure. The recruitment of high resolution crystallographic data sets readily supported a structural and dynamical interpretation of the observed chemical shift responses to ligand binding at 1 bar pressure, and on the application high pressure. A definitive atomistic identification of fast motion contribution to activation barrier compression was not obtained. Nevertheless, detailed, residue specific amide compression profiles, and shifts in backbone amide conformational exchange regimes in response to ground state ligand binding, and at high pressure, have been catalogued in the PETNR:FMN free enzyme system. These dynamical profiles in the free enzyme are contrasted against comparative, residue specific observations in analogue complexes of the oxidative and reductive half reactions of PETNR.

This thesis proposes and implements a system for movement analysis of vehicles on crossroads. It detects and tracks the movement of vehicles in the video, gained from the stationary video camera, which has the view of some crossroad. The trajectories are stored and their number and directions are analysed. The detection was made using cascade classifier. A dataset of 10500 positive and 10500 negative samples has been created to train the classifier. Vehicles are tracked using KCF method. For trajectory clustering, needed by analysis, the Mean Shift method is used. Testing showed, that the overall success of vehicle movement analysis is 92.77%.

Pour les technologies avancées, la lithographie optique par immersion utilisant des sources 193nm atteint ses limites en termes de résolutions. Les nouvelles techniques de lithographie telles que l’Extrême UV ou l’écriture à faisceaux multiples n’étant pas encore au point d’un point de vue industriel, un ensemble de techniques d’améliorations de la résolution (RET) en évolution constante permet de continuer à travailler avec les équipements à immersion. C’est dans ce contexte que s’inscrit cette thèse en se focalisant sur un des éléments principaux : le réticule. La première partie regroupe l’état de l’art de la photolithographie optique et des techniques d’amélioration de la résolution, ainsi qu’un ensemble de notions utile à la compréhension des parties B et C. La partie B concerne la réduction des effets de focalisation, plus connu sous l’appellation anglaise ‘’Best focus shift mitigation’’, observés sur des niveaux denses de type ‘’Metal’’ pour les technologies avancées. Cette partie permet de comprendre l’influence des organes physiques sur le trajet optique et propose comme solution au Best focus Shift un changement de réticule, validé d’abord via simulation puis par voie expérimentale sur plaquette. La partie C de ce manuscrit présente la prise en compte des effets inhérents à la fabrication des réticules dans l’étape de correction des effets de proximité optique, de la description des effets impliqués à la calibration et l’utilisation d’un modèle dédié
For advanced technologies nodes, immersion optical lithography using 193nm sources reaches its limits in terms of resolutions. Since new lithography techniques such as Extreme UV or multi-beam writing are not yet ready from an industrial point of view, a continuous evolving set of tools known as Resolution Enhancement Technics (RET) allows to continue working with immersion equipment, pushing the resolution limits as much as possible. With the increasing design complexity, this task is more and more challenging. Within this frame this dissertation is addressed to improve the lithographic process variability by focusing on one of the main elements: the reticles.The first part brings together the state of the art of optical photolithography and resolution improvement techniques, as well as a set of concepts useful for parts B and C understanding. Part B deals with the reduction of focusing effects, better known as "Best focus shift mitigation", observed on dense levels such as "Metal" for advanced technologies. This part helps to understand the influence of the physical organs on the optical path and proposes as a solution to the Best focus Shift a reticle change, firstly validated via simulation and then experimentally at wafer level. Part C presents how the inherent effects of reticle fabrication can be taken into account in the Optical Proximity effects Corrections steps, from the description of the effects involved to the calibration and the use of a dedicated mask model

Thesis (Ph. D.)--Physics, Georgia Institute of Technology, 2007.
You Li, Committee Chair ; Citrin David, Committee Member ; Kuzmich Alex, Committee Member ; Fox Ronald, Committee Member ; Chapman Michael, Committee Member.

Cette thèse tente d’analyser comment les œuvres de Violette Leduc ont remis en cause, de façon avant-gardiste, les normes de sexe, de genre et de sexualité dans l’écriture de fiction. Différentes méthodes critiques sollicitées durant le travail permettent de mettre au jour le lien intime et corporel qu’entretient Violette Leduc avec ses œuvres. Ces méthodes permettent également d’entrevoir une poétique décalée, décentrée, qui déconstruit les genres littéraires et offre de nouveaux points de vue sur le monde. Il s’agit par conséquent de montrer dans ce travail que c’est à travers la poétique et non la rhétorique que l’auteure façonne, à la façon de l’artisan, une résistance à la norme. Trois concepts majeurs servent d’outils critiques à l’étude des textes : la performativité, la déconstruction et la limite toujours floue entre réalité et fiction. Ces outils, confrontés aux trois thèmes principaux de la corporéité, de la sexualité et de l’infra-ordinaire qui sont analysés de près dans ces recherches, remettent en question les désignations de « sexe », de « genre » et de « sexualité ». La position du lecteur se voit elle-même décentrée, afin qu’il puisse accéder aux « trésors à prendre » du texte. Il en résulte, conséquemment, la nécessité de reconsidérer les processus de la poétique et les impacts philosophiques, sociologiques et politiques de ces textes
In this thesis we have tried to analyze how the works of Violette Leduc challenged, in an avant-garde manner, the norms of sex, gender and sexuality in fictional writing. Various critical methods have helped reveal the physical and intimate relationships between Violette Leduc and her own body of work. They have also helped foresee a shifting, off-centered poetics that deconstructs the literary genres and offers alternative views of the world. Hence we have aimed at displaying how the writer crafted, crafts(wo)manlike, a resistance to norms through poetics rather than rhetoric. Three main concepts have been used as critical tools for text analysis : performativity, deconstruction, and an ever-blurred limit between reality and fiction. Confronted to the three main themes of corporeity, sexuality and infra-ordinary – which are thoroughly examined in this work – these tools help questioning the very notions of sex, gender and sexuality. The reader himself is urged to shift off-center in order to get access to the texts’ « treasures to-be-taken ». It has appeared necessary, then, to reconsider the process of poetics and the philosophical, sociological and political impact of these texts

Etude de la dynamique atomique interne; possibilité de séparer la contribution des fluctuations du vide et de la réaction du rayonnement pour differents processus radiatifs (émission spontanée, déplacement de Lamb, etc. ). Etude de la dynamique atomique externe en partant de l'analogie entre ce problème et celui du mouvement brownien; établissement d'une équation de Fokker-Planck-Kramers pour l'évolution de la fonction de Wigner atomique, avec termes de force stationnaire, force de frottement etc. Etude détaillée du mouvement atomique dans une onde laser stationnaire intense: refroidissement, piégeage.

碩士
國立交通大學
工學院碩士在職專班精密與自動化工程組
98
This research proposed a practical and useful method which apply Finite Element Analysis to investigate the difference of stress distribution between the constant velocity joint of the BJ type（6 ball CV Joint）and EBJ type（8 ball CV Joint）. One motivation of the research is to confirm the possibility of cost saving in material and production of C.V.Joint, and to get the better superiority in design at the Automobile Chassis space without reducing the strength and durability. This research adopted SolidWorks cooperating with Finite Element Aanalysis system（COSMOS Works）to build the geometric model and solve the stressed distribution of the components of Ball Joint assembly on the driveshaft. The C.V. Joint models in the study are similar models as BJ and EBJ type of NTN product. On behalf of the iterative parameter optimized process, we can conclude that the model of EBJ with less weight could be a more competitive C.V. Joint product than BJ type. As shown in the analysis result, we have investigated the difference of stress distribution between BJ type and EBJ type, and the weight in EBJ type is 20% smaller than BJ type. The Stress at the contact point between EBJ ball and housing is smaller than that of BJ ball, so we can hypothesize that the temperature rise on the housing of the EBJ type will be smaller than that of BJ type, and the conclusion is similar with NTN tested result. The research proposes a model-analysis investigation concurrent design process and the new EBJ type of C. V. Joint has been evaluated and can be concluded that it could be a more competitive product with much more benefit on less weight, smaller size and less temperature rise than that of BJ type.

博士
國立成功大學
系統及船舶機電工程學系
103
Abstract The vibration analyses of the ship propulsion system, including axial vibration, torsional vibration, transverse vibration and whirling motion etc., are the critical issues for its design, and all the ship-design engineers must face this problem carefully. In which, both the torsional vibration and whirling motion are associated with the shaft rotation, it is obvious that the gyroscopic moment on the propeller, flywheel and flanges etc. will vary periodically, when a torsional vibration occurs in the ship propulsion system, and any initial-displacement-induced non-synchronous precession could be critical. In the existing literature, the dynamic characteristics of whirling motions of the shaft-disk systems are usually analyzed by using the TMM or FEM. The results of the last methods are the approximate solutions and the exact solution concerned is rare. For this reason, this thesis presents an analytical method for yielding the exact whirling speeds and mode shapes of a distributed-mass shaft mounted by arbitrary rigid disks. In theory, the whirling motion of a shaft-disk system is three-dimensional, however, if the transverse displacement in the xy-plane and that in the xz-plane are combined by using a complex number, and the effects of each rigid disk on the shaft are replaced by a lumped mass together with a frequency-dependent equivalent mass moment of inertia, then the whirling motion of a rotating shaft-disk system will be similar to the two-dimensional transverse free vibration of a stationary beam. Therefore, after the foregoing manipulations, some techniques for the free vibrations of a stationary beam carrying various concentrated elements will be available for the whirling motions of a rotating shaft mounted by arbitrary rigid disks. Based on the above theory, some practical examples are illustrated. Numerical results reveal that when a shaft-disk system is put into a spinning motion about its longitudinal axis, each of its natural frequencies of stationary transverse free vibrations will split into forward whirling speed and backward one, furthermore, the characteristics of whirling motions are significantly dependent on the slopes of the associated natural mode shapes at the positions where the rigid disks located. In order to confirm the reliability of the presented theory and the developed computer programs, most of the results obtained from the presented method are compared with those obtained from the existing literature or the FEM and good agreements are achieved.

碩士
淡江大學
資訊工程學系碩士班
96
We propose a novel motion analysis algorithm by using the mean-shift segmentation and motion estimation technique. Mean shift algorithm is frequently used to extract objects from video according to its efficiency and robustness of non-rigid object tracking. For diminishing the computational complexity in searching process, an efficient block matching algorithm: cross-diamond-hexagonal search algorithm was used. In the motion analysis procedure, the stick figure of object obtained by thinning process is treated as guidance to gather the statistics of motion information. The experimental results show that the proposed method provides precise description of the behavior of object in several video sequences.

碩士
國立東華大學
資訊工程學系
96
This study proposes a novel model-based algorithm, dynamic kernel based progressive particle filter, for 3D unconstraint human body tracking. The articulated full body motion tracking is the large number of degrees of freedom to be estimated. The aim of this study is to reduce the computational cost and improve the accuracy. The progressive particle filter consists of three principal approaches: the multiple predictions technique of particle filter, the iterative mode seeking algorithm of mean shift and hierarchical searching approach. The hierarchical searching approach decomposes the high dimensional space into three subspaces, global motion layer, local inside motion layer and local outside motion layer, for decreasing the searching range. Natural human motion is usually non-linear and non-Gaussian, the system applies the random sampling technique of particle filter to predict the unconstraint motion of human body. The mean shift trackers are embedded into each particle to improve the accuracy by iterative mode seeking. Moreover, the study proposes a dynamic kernel model, which can automatically adjust the kernel bandwidth of mean shift trackers according to the probability of each state by improving the searching effectively. This study combines the progressive particle filter and dynamic kernel model for reducing the computational cost and improving the accuracy simultaneously.

碩士
國立中興大學
機械工程學系所
98
The objective of this thesis is to study the dynamic responses of the flexible spinning-shaft system including the rigid-body motion effect. The spinning-shaft system being considered include the rigid disk, the flexible shaft, and bearing supports which are modeled as springs and viscous dampers. To derive, the equations of motion of the spinning-shaft system. First, the kinetic energy and the strain energy of the spinning shaft, the kinetic energy of the rigid disk, and the work done by support forces of the bearings and centrifugal force of the disk are found. Then by employing the Hamilton’s principle together with the finite element method, the equations of motion of the spinning-shaft system including both the rigid-body motion and flexible deformation effects of the shaft are obtained. In the study of dynamic responses, in order to simplify the analysis, the influence of the flexible deformation of the shaft on the rigid-body motion of system is neglected. The equations of motion then can be divided into two groups corresponding to that of rigid-body motion and that of flexible motion affected by the rigid-body motion. The nonlinear equations for rigid-body motion is further simplified to linear ones for comparison. In the numerical analysis, a higher-order Runge-Kutta method is used to solve the equations of motion describing the rigid-body motion of the system. With above obtained results equations of flexible motion affected by rigid-body motion are then analyzed by the Newmark’s method. The influences of the parameters such as rotational viscous damping coefficient, the locations of bearings and the disk, and the motor’s acceleration on the dynamic responses are investigated. Comparisons are also made with those of the spinning shaft system without considering the rigid-body motion effect.

碩士
國立臺灣科技大學
電機工程系
95
Motion detection and object tracking are very popular research in the field of computer vision all the time. The achievements of the research can be widely applied to some specific field. However, most of digital image process systems are based on personal computer. Image materials should be captured by the CPU. Moreover, the CPU has to run Windows Operating System as well. These procedures often cause huge loading of the CPU and lower efficiency of the system. Because there are many benefits when using Digital Signal Processors (DSP), not only are DSP-based systems highly accurate as well as efficient in the processing of high volume and complex algorithms but can also meet most systems’ requirement for real-time performance. We combine the TI TMS320DM642 EVM with CCS software system to be the research developing platform. At beginning we use EDMA controller to create DMA gateway then it directly stores image materials in the capture buffer to wait for process. Then the position and width of the moving object can be found by Moving Edge Detection Method. We will capture target model from current frame. When we have the target model, we can track moving object according to the target model based on Mean Shift algorithm. Finally the processed images will be exported on the screen by EDMA controller. The pan/tilt plate will be controlled by GPIO to keep moving object in the central screen at the same time. We complete the automatic motion tracking system in the end.

碩士
國立陽明大學
生醫光電研究所
100
The primary purpose of the research is to monitor the cell membrane motion and get the Elasticity or Young's Modulus of the cell membrane. The velocity of cell membrane motion is about nano scale, so we need a measurement system with non-invasive and high sensitivity. Then we combine Laser Doppler velocimetry and confocal fluorescence microscope which make the measurement system not only have the advantage of non-invasive and high sensitivity but real-time measurement and high axial resolution. Moreover we use fluorescence techniques to locate the horizontal plane which sample is on. In order to have the ability to select the horizontal plane we detect. We use motorized linear stage to change the focal plane of confocal microscope system. Because the noise caused by environment would influence the detection from Laser Doppler velocimetry. So the setup we design have same effect with common optical path design. By Using the polarize optic, the system could effectively eliminate the noise caused by environment. We have verified the system can detect the signal of reflection light from piezo. In signal analysis, the method we use is Hilbert Huang Transform We can use it to analyze the Doppler signal and get the real-time velocity and displacement.

碩士
國立臺灣大學
機械工程學研究所
91
The aim of this thesis is to develop a non-contact error motion measurement system for small shafts in high speed. Laser diodes and linear photodiode arrays are built up as detectors for measuring radial and axial error motions. Starting with literature surveying, different types of error motions and measuring methods were studied. Our measuring system was designed and developed based on the determined error motions. This system’s configuration consists of laser ray generators and photodiode arrays; the signal processor includes driver and encoding circuits for radial and axial error motions measurement; and for automatic data acquisition and processing, DAQ Card and LabView software are applied. The shadow principle is employed to measure the radial error motion, and the axial error motion is detected by the reflective principle. The photodiodes detect the motions of laser rays and transfer them into electronic charges. Besides the analog signals processing, digital processing circuits are developed to decrease the interference noise and to increase sensing accuracy. This thesis also explores the influence of signal processing methods on resolution, linearity and sensitivity thoroughly; and by applying the digital processing method the dynamic behavior of the developed system is also empirically studied.