341 |
Reliability analysis of lunar rover wheel designs - methodology and visual inspection techniqueParbatani, Karim January 2012 (has links)
This thesis presents a methodology for prediction of reliability for a lunar rover and a technique to evaluate reliability where data for operation in the lunar environment are not available in order to conduct traditional reliability analysis. This technique was applied to lunar rover wheel designs.The maximum distance travelled by the Lunar Roving Vehicle (LRV) during the Apollo 17 mission was approximately 35.9 km. Presently, McGill University, Neptec and the Canadian Space Agency are collaborating to develop a wheel set for an LRV, which is targeted to have a five-year operational service. This underlines the importance of the reliability of an LRV, which will be required to travel thousands of kilometres. The Neptec Design Group has developed a methodology to estimate the reliability of a lunar rover, i.e., the Lunar Rover Reliability Prediction (LRRP) where the project was divided into two phases. Phase 1 consisted of three milestones which included background research of the various reliability models currently available, the specific software tools that execute these models, and the ability to adapt these tools to account for lunar environmental factors (e.g., temperature, regolith, low gravity). Results demonstrated that Relex provided a strong package of fully integrated software tools to execute a group of reliability models (FMEA [Failure Modes and Effects Analysis], Fault Tree, Standards Prediction, ALT [Accelerated Life Testing]). Mission simulation software was identified to be a worthwhile addition and GoldSim was considered to be the best choice to provide mission simulation analysis. Phase 2 outlined the process for estimating the reliability of a lunar rover for a 5-year operational life on the moon based on the recommendations detailed in Phase 1. LRRP has been a success, but Neptec was only able to obtain earth bound data for use in the methodology. Hence, research on cryogenic wear was explored and the summary of relevant findings is as follows: the wear rate of steel and titanium at 293K and 77K are about the same below a sliding speed of 1m/s; the wear rate at 77K above 1m/s is higher for most materials and the coefficient of friction for steel on steel at 77K and 273 K is about the same. The information present in the literature is insufficient to evaluate LRV reliability under lunar environmental conditions. This underlines the necessity for an alternative methodology, for instance, the visual inspection technique, to evaluate reliability of systems (e.g., a lunar rover), where useful data are not available.The visual inspection technique is described and an analysis of five historical wheels and four new wheels designed by McGill University is given. The wheel designs were visually analysed according to specific criteria with respect to potential failure under lunar conditions (temperature, vacuum, regolith) considering wear resistance, thermal compatibility, excess loading, torque, etc. Potential wheel performance was quantified using a numerical scale with respect to potential failure due to events, such as temperature fluctuation, wheel-rock impact, excess payload and abrasive wear. The best designs were the Apollo 17 wire mesh wheel (historical) and the iRing wheel (McGill). A reliability checklist for lunar wheel designs is given to serve as a guideline to avoid unreliable design features. The visual inspection technique was successful and can be applied to other complex systems where experimental testing under actual conditions is not possible, although background knowledge and expert opinion are necessary to perform an evaluation. / Cette thèse présente une méthodologie pour la prédiction de la fiabilité des roues lunaires, et une technique pour évaluer la fiabilité de la conception de ces roues, pour lesquelles les données d'opération dans l'environnement lunaire ne sont pas disponibles, afin que nous puissions effectuer une analyse de fiabilité traditionnelle. La distance maximale traversée par le Rover Lunaire durant la mission Apollo 17 était d'environ 35.9 km. Aujourd'hui, l'Université de McGill, le groupe Neptec Design et l'Agence Spatiale Canadienne travaillent ensemble afin de développer une roue de Rover Lunaire pouvant être opérationnelle pendant 5 ans. Cela souligne l'importance de la fiabilité des roues, qui auront à traverser des milliers de kilomètres. Le groupe Neptec Design a réussi à développer une méthodologie afin d'estimer la fiabilité d'un rover lunaire, comme nous le montre le document Prédiction de Fiabilité de Rover Lunaire (PFRL) qui a été publié par l'entreprise. Cette méthodologie comprend deux phases. Phase 1 contient trois étapes incluant les recherches approfondies concernant les différents modèles de fiabilité disponibles, les outils logiciels spécifiques qui permettent la mise en place de ces modèles, et l'habilité à adapter ces outils à l'environnement lunaire (par exemple, la température, la présence de régolite, les radiations, l'apesanteur). La Phase 2 décrit le processus d'estimation de la fiabilité du rover lunaire pour une durée de 5 ans sur la Lune, basé sur les recommandations détaillées dans la Phase 1. Le PFRL a été un succès mais Neptec est seulement capable d'utiliser des données earthbound pour la méthodologie, ce qui est insuffisant. Donc, le domaine de l'usure à la cryogénie à été exploré à travers la littérature et les bases de données, et le résumé des éléments qui ont été relevés intéressants est le suivant : le taux d'usure de l'acier et du titane à 293K et 77K sont à peu près les mêmes, en dessous de 1m/s ; le taux d'usure à 77K en dessous de 1m/s est plus haut pour la plupart des matériaux et le coefficient de frottements pour l'acier sur l'acier à 77K et 273K est à peu près le même. Cela souligne la nécessité d'une méthodologie alternative, par exemple, la technique d'inspection visuelle, pour évaluer la fiabilité d'un système quand des données utiles ne sont pas disponible. La technique d'inspection visuelle est décrite et une analyse de 9 conceptions de roues (cinq anciennes roues et quatre nouvelles roues conçues par l'Université de McGill) est démontrée. Les conceptions des roues sont analysées visuellement en fonction de critères spécifiques, sans négliger les risques d'échec potentiels si l'on transpose aux conditions lunaires (température, aspiration, et régolites), la fiabilité mécanique, sans oublier les obstacles à la performance, la résistance à l'usure, la compatibilité thermale, la vitesse de chargement, la torsion, etc. Une échelle numérique a été créée et la performance potentielle de la roue a été quantifiée en prenant compte des évènements pouvant causer un échec comme les fluctuations de températures, le soudage à froid, les impacts de rochers, un excès de charge et une usure abrasive. Les résultats ont indiqué que la meilleure conception historique était la roue à treillis métallique d'Apollo 17 et que la meilleure conception contemporaine est la roue iRing. L'exposé donne aussi une "liste de conception pour la fiabilité", qui peut potentiellement servir de guide afin d'éviter les traits de conceptions avec des failles. La technique d'inspection visuelle a été un succès et peut être appliquée à d'autres systèmes complexes où les tests expérimentaux dans les conditions actuelles ne sont pas possibles. Cependant, des connaissances préliminaires et une expertise sont nécessaires afin de mettre en place l'évaluation.
|
342 |
Detection of cracks in aircraft structures using piezoelectric sensorsHan, Yong, 1969- January 2005 (has links)
Structural damage detection at the earliest possible stage is very important in the aerospace industry to prevent major failures. In this work, a potential cost-effective crack detection method using piezoelectric strips bonded on the structure was studied. A crack model was developed and validated. Static, modal and transient dynamic analysises are performed for the case of piezoelectric strips bonded to the structure by using a finite element method to explore the effectiveness of the crack detection method. Panel methods are used for steady flow problems of fixed wings in order to calculate the pressure distributions on the wing surface. / A flat wing with a crack subjected to aerodynamic load was simulated to predict the presence of the crack. It is found that the voltage difference between the piezoelectric strips bonded at the same location (one on the upper side and the other on the lower side) can be used to predict the presence of the crack. For wing structure crack detection, the sensitivity is limited if the steady aerodynamic load is used as an excitation.
|
343 |
The design of robust helium aerostats /Miller, Jonathan I. January 2005 (has links)
Tethered helium aerostats are receiving renewed attention in the scientific and surveillance communities. However, conventional aerostats cannot consistently survive high winds. The goal of this research was to design an aerostat that could be deployed for very long periods, thus reducing operating costs and interruptions in data acquisition. Existing designs and fabrication techniques were first reviewed and replicated in the construction of a 2.5 m diameter spherical aerostat. The constructed balloon was then flown outdoors to observe its operational qualities. A low-cost data acquisition system was assembled to characterize the balloon's dynamics. The results were used to inform a Finite Element Analysis model evaluating the critical stresses in a 10.15 m diameter balloon's envelope and its tendency to "dimple" when subjected to high wind speeds. A second model was created to appraise the performance of an aerostat with a partially hard shell, made of carbon fiber, in highly loaded areas.
|
344 |
The characterization of drilling process of woven composites using machinability maps approach /Rawat, Sanjay. January 2006 (has links)
Woven carbon fiber composites are being extensively used in aerospace, automotive and civil applications, owing to their high specific strength, higher fracture toughness, and drapeability as compared to unidirectional composites. The processing phase of composite manufacturing has a high added value to the processed laminate, so any damage during machining has a significant economical impact. However, the characteristic attributes like non-homogeneous structure, anisotropy and high abrasiveness of fibers coupled with non-optimized cutting parameters and tool wear can result in damage to the laminate. This experimental research involves quantifying the effect of processing parameters i.e. spindle speed, feed, and tool wear on hole quality during the drilling of a quasi-isotropic woven graphite fiber epoxy laminate. Drilling tests were conducted for wide range of spindle speeds (1,500 rpm to 15,000 rpm) and feed rates (20 μm/rev to 800 μm/rev) with a 5 mm diameter standard point tungsten carbide twist drill. This investigation extended the drilling approach to evaluate the benefits of high speed (12,000 rpm and 15,000 rpm) and high feed (600-800 μm/rev) regimes, which were not explored or reported in the open literature. Dependence of damage mechanisms; namely, delamination, fiber pull out, thermal damage, surface roughness deterioration, hole circularity and hole diameter errors, on cutting conditions was established. A strong correlation between different types of damages and cutting forces and temperature was also identified. High speed drilling approach produced contrasting results as spindle speed of 15,000 rpm was found to reduce delamination damage and improve surface roughness but increase the hole circularity and diameter errors. Machinability maps were designed to illustrate the affect of cutting parameters on delamination damage, hole circularity, hole diameter error and hole surface roughness. The application of machinability maps as an aid for process engineers to avoid damage and optimize the process is demonstrated. Finally, it was concluded that the process could be controlled and delamination be eliminated by increasing the spindle speed up to 15,000 rpm and feed rate up to 100 mum/rev. It results a higher productivity if a compromise on hole circularity and diameter tolerances is accepted. Having established the benefits of high speed and low feed drilling, tool wear was evaluated at 12,000 rpm and 15,000 rpm and optimal feed rate of 100 mum/rev. Chipping, abrasion and adhesion of carbon were found to be the main wear mechanisms. Abrasion at the flank face of the drills was identified to be the main wear process that controls the deterioration of the drill at high speeds. The three wear regimes i.e. primary, secondary, and tertiary, were found to strongly influence the thrust force, delamination damage, hole circularity, diameter error and hole surface roughness. Finally, the tool wear analysis revealed that a tool change strategy could be devised by monitoring the thrust force, rather than measuring the progression of flank wear.
|
345 |
Experimental characterization and simulation of a tethered aerostat with controllable tail finsHoward, Alistair January 2008 (has links)
This research investigated the use of a streamlined aerostat with movable tail fins to control the pitching motion of an aerostat in a turbulent environment. The goal of the research was to reduce the pitch fluctuations of the aerostat by actively controlling the tail fins. A streamlined aerostat with four rigid tail fins was modified to include flap type trailing edge control surfaces on the horizontal tail fins. Airborne sensors and a ground station with wind sensors were used for the control of the aerostat and data logging. In order to characterize its motion and to establish a performance baseline, the aerostat was flown numerous times without active control. The effectiveness of the tail-fin control surfaces was determined by measuring the open-loop response of the aerostat to a step change in the control surface deflection. The results from the uncontrolled and open loop testing were used to develop the closed loop control system which was then tested experimentally. An additional goal of this research was to adapt a non-linear dynamic simulation to simulate the controllable aerostat used in this experiment. The experimental aerostat's aerodynamic and geometric properties were characterized for use in the dynamic model. In general, the 'non-dynamic' quantities in the simulation showed a good match and the dynamic variables did not show as good of a match between the simulation and the experiment suggesting that the model does not correctly represent the dynamic behaviour of the aerostat. / Cette recherche a examiné l'utilisation d'un ballon à ligne profilée avec ailettes mobiles pour contrôler le tangage de ce ballon dans un environment venteux. Le but de cette recherche a été de réduire les fluctuations de l'inclinaison du ballon en contrôllant activement les ailettes arrières. Un ballon à helium à ligne profilée avec quatre ailettes rigides a été modifié en ajoutant des surfaces de contrôle du type aileron sur les ailettes arrières horizontales. Des capteurs à bord de l'aérostat et une station terrestre avec capteurs de vent ont été utilisés pour le contrôle de l'aérostat et l'entrée des données. Pour caractériser le mouvement et pour etablir une performance de base de l'aérostat, il a été volé plusieurs fois sans contrôle actif. L'efficacité des surfaces de contrôle sur les ailettes arrières a été determinée en mesurant la réponse à boucle ouverte de l'aérostat d'un changement progressif du braquage des surfaces de contrôle. Les resultats des tests non-contrôlés et de ceux contrôlés à boucle ouverte de l'aérostat ont étés utilisés pour développer le système à boucle fermée qui a ensuite été testé expérimentalement. Un but additionel de cette recherche a été d'adapter une simulation dynamique non-linéaire pour simuler l'aérostat contrôlable utilisé dans cette expérience. Les propriétés aérodynamiques et géometriques de l'aérostat experimental ont étés caractérisés pour utilisation dans ce modèle dynamique. Géneralement, les quantités 'non-dynamiques' dans la simulation étaient en bonne correspondence pendant que les variables dynamiques correspondaient moins bien entre la simulation et les données expérimentales. Ceci suggère que ce modèle ne représente pas entièrement correctement le comportement de l'aérostat.
|
346 |
Development of an intelligent injection system and its application to resin transfer moldingChu, Jingsong January 2003 (has links)
Resin Transfer Molding (RTM) is a procedure in composite material manufacturing in which resin and fibers are held apart until the last possible moment. In this way it can be contrasted with other manufacturing methods where the resin and fiber are combined prior to use. In RTM, unresinated fibers are held within a tool cavity and a differential pressure is applied to a supply of resin so that the resin flows into the reinforcement completely wetting it out. Reinforcements may be made of any fiber and the use of all forms has been reported, from unidirectional through woven or knitted cloths to needled and random mats and fully three-dimensional reinforcement preforms. The resin can be of a very wide range of chemistries and formulations, so long as the basic process requirements are met. Cure times can be from a few minutes to many hours. Resin injection machines vary widely, and production line design can be just as varied. The focus of this project will be the development of an intelligent injection system and its application to RTM. In order to achieve this, the Lab VIEW program will be employed to control pressure and flow rate—the two important parameters of RTM. This system includes an injector, computer, data acquisition boards, electric system and pneumatic system. A number of experiments have been performed to find the optimal application of this system. During the manufacturing of composite parts, all the processing parameters, such as temperature, pressure, flow rate and displacement with respect to time, can be written to a computer hard disk to be analyzed later. All the data acquired during the operation can help us to understand in depth the processing of composite materials and the relationship between process, quality and property.
|
347 |
A modular approach to the synthesis of quick-return mechanisms /Wu, Chu-Jen. January 1998 (has links)
Quick-return mechanisms are quite common in manufacturing processes. e.g in pick-and-place operations, metal-cutting, and metal-forming, where dwell is usually required. Dwell is the period where a driven link remains stationary and the tool can be replaced or the workpiece can undergo a machining operation. Unlike lower-pair mechanisms, cam mechanisms can produce dwell exactly. / In this thesis, cam-follower systems are used as building modules in the design of quick-return mechanisms. The underlying cam mechanisms are optimized to obtain a compact, quick-return mechanism. A unified method for the optimization of cam mechanisms is proposed. The optimum parameters of the cam mechanism are obtained by cam-area minimization subject to performance and geometric constraints. In addition, the power required from the motor is reduced by adding an elastic torque-compensation mechanism to the system. A Graphical User Interface is developed to aid the designer during the design process. / The modular method is applied to the synthesis and optimization of long-stroke, quick-return mechanisms. In addition, the design of a simple cam mechanism to replace the transmission mechanism of a textile machine composed of an elliptic-gear train and a four-bar linkage is also included as an application example.
|
348 |
Design for manufacturability of a composite helicopter structure made by resin transfer mouldingO'Flynn, Julian January 2008 (has links)
Resin transfer moulding (RTM) is a promising composite manufacturing method capable of producing high quality parts while meeting the rising demand for cost effectiveness. A collaborative research project involving academia, industry, and government in the Montreal area is under way to optimize RTM and help transfer this technology to local industries. As a technology demonstrator, the leading edge slat of the Bell 407 helicopter is being redesigned using composite materials and RTM. This thesis presents the mould design, part design for manufacturability, and actual production of half-length prototype parts. The mould and part designs were carried out concurrently taking into consideration results from stress analysis and flow modelling, as well as manufacturing constraints. A total of eight prototypes were manufactured. During the development of this project, important improvements were made in the way composite parts are manufactured by RTM. The lessons learned from this prototype design and manufacturing work will help to produce the full-size version of the slat as well as other components in the future. / Le moulage par injection sur renfort (resin transfer moulding, RTM) est un procédé de fabrication prometteur, produisant des pièces en matériaux composites de haute qualité tout en ayant des coûts de production avantageux. Un projet de recherche incluant des collaborateurs du secteur académique, industriel, et gouvernemental dans la région de Montréal est en cours afin d'optimiser le procédé RTM et de faciliter le transfert de cette technologie aux industriels locaux. Pour démontrer cette technologie, le bec de bord d'attaque de l'hélicoptère Bell 407 est reconçu utilisant des matériaux composites fabriqué avec le procédé RTM. Cette thèse présente la conception du moule et de la pièce en fonction de la facilité de fabrication ainsi que de la production de prototypes demi-longueurs. Le moule et la pièce ont été conçus simultanément considérant les résultats des analyses de contraintes, d'écoulement de résine dans le renfort tout en tenant compte des contraintes pouvant être encourues lors de la fabrication. Huit prototypes demi-longueurs ont été fabriqués. L'expérience acquise lors de la conception et la fabrication du moule ainsi que des prototypes aideront à la production du bec de bord d'attaque pleine longueur en plus d'être utile lors de conceptions futures.
|
349 |
Kinematic analysis of planar parallel manipulators with holonomic higher pairsHayes, Matthew John D. January 1996 (has links)
This thesis presents a detailed kinematic analysis of a 3-degree-of-freedom planar parallel manipulator with holonomic higher pairs. The manipulator consists of a circular disk which rolls without slip on the non-grounded rigid links of each of three 2R serial legs. / The first portion of the thesis is devoted to the review of the geometric and mathematical tools used in the kinematic analysis. Planar isometries and group theory are used in the developmental of the inverse kinematics (IK) algorithm. Kinematic mapping and Grobner bases are important for the forward kinematics (FK) algorithm. / After six important geometric properties of the manipulator are identified, the IK algorithm is developed. It is based on the decomposability and commutativity of planar displacements. The four step algorithm provides closed form analytic solutions. The algorithm may be used on similar parallel manipulators with any number of 2R legs, and hence, applies to a whole class of manipulators. It will be shown that there can be no more than 4$ sp{n}$ real solutions, where n is the number of 2R legs. Three numerical examples are given. / The FK problem is solved using kinematic mapping. To employ a technique from the literature, pseudo inputs must be used to specify joint parameter inputs. The resulting set of three non-linear equations in three unknowns is solved using Grobner bases theory. A numerical example is given. / Finally, velocity and acceleration analysis are performed after the determination of the Jacobian matrix.
|
350 |
Planning robotic tasks for impedance controllers in oscillating and uncertain environmentsO'Reilly, Philippe. January 1997 (has links)
This thesis presents the result of a research work done in the synthesis of an impedance controller which can execute compliant tasks in oscillating and uncertain environments. A synthesis method is proposed for finding the best control parameters, such as the impedance parameters and the desired trajectory, for a robotic manipulator which has to execute a compliant task. The control parameters are derived from an optimization method. The optimization function simulates the task execution and evaluates the performance according to the specified task goals, which are expressed as inequalities on position, velocity and force responses. These goals are built in order to obtain the control parameters that will ensure the highest robustness in oscillation and uncertainty cases. The non-experimental method builds its results from decoupled linear impedance models of the robot and the environment. The synthesis is applied to three specific tasks performed on the live line maintenance of distribution network at Hydro-Quebec, an electric utility company.
|
Page generated in 0.1 seconds