Structural vibration control (SVC) of large scale structures using the magnetorheological (MR) dampers are studied. Some key issues, i.e. model reduction, suppression of spillover instability, optimal placement of actuators and sensors, modeling of the MR dampers and their applications in SVC system for large scale structures, are addressed in this work. A new model reduction method minimizing the error of a modal-truncation based reduced order model (ROM) is developed. The proposed method is implemented by using a Genetic Algorithm (GA), and can be efficiently used to find a ROM for a large scale structure. The obtained ROM has a finite H2 norm and therefore can be used for H2 controller design. The mechanism of the spillover instability is studied, and a methodology to suppress the spillover instability in a SVC system is proposed. The suggested method uses pointwise actuators and sensors to construct a controller lying in an orthogonal space spanned by the several selected residual modes, such that the spillover instability caused by these residual modes can be successfully suppressed. A GA based numerical scheme used to find the optimal locations for the sensors and actuators of a SVC system is developed. The spatial H2 norm is used as the optimization index. Because the spatial H2 norm is a comprehensive index in evaluating the dynamics of a distributed system, a SVC system using the sensors and actuators located on the obtained optimal locations is able to achieve a better performance defined on a distributed domain. An improved model of MR dampers is suggested such that the model can maintain the desired hysteresis behavior when noisy data are used. For the simulation purpose, a numerical iteration technique is developed to solve the nonlinear differential equations aroused from a passive control of a structure using the MR dampers. The proposed method can be used to simulate the response of a large scale structural system with the MR dampers. The methods developed in this work are finally verified using an industrial roof structure. A passive and semi-active SVC systems are designed to attenuate the wind-induced structural vibration inside a critical area on the roof. The performances of the both SVC systems are analyzed and compared. Simulation results show that the SVC systems using the MR dampers have great potentials in reducing the structural vibration of the roof structure.

This paper considers the control of externally symmetric vector second order systems using an analytical upper bound method. The structural model is a cantilevered aluminum beam with a collocated pair of piezoceramic patches to serve as actuators and sensors. A computationally efficient method for approximating the H-infinity norm for externally symmetric systems is presented. The approximation method is then used to calculate a scalar output feedback controller to guarantee a closed-loop norm less than any user defined value. This method is tested with a finite-element representation of the beam, and then verified experimentally.

by Wai Ki Ching. / Thesis (Ph.D.)--Chinese University of Hong Kong, 1998. / Includes bibliographical references (p. 82-87). / Electronic reproduction. Hong Kong : Chinese University of Hong Kong, [2012] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Mode of access: World Wide Web. / Abstracts in English and Chinese.

In modern industrial applications, sensors are an expensive part of installed systems. Nevertheless, many system variables cannot be measured sufficiently frequently or accurately. Thus, soft sensors have been developed to estimate those variables without the expense of additional hardware. The use of a soft sensor with a bias update term has shown to perform well for disturbed systems with time delays and multirate sampling times. In industrial application, the time delay and sampling times often vary. Yet, the case of variation of the time delay and sampling time in the bias update term has not been considered in previous publications. This thesis tests a soft sensor with bias update term in simulation and gives a modification yielding better performance. It is shown that the tested method gives unstable results. Hence, a more general method with a bias update term that considers all possible sampling times in each step is proposed, giving stable results in simulation. Furthermore, the stability of the general method is proven mathematically by building a state space representation and applying / Tesis

A successful cancer treatment is a combination of early diagnosis and efficient use of anticancer drugs. There is a chance of approximately 70 - 90% of cancer patients surviving if the diagnosis is conducted early. That means if a diagnosis system is in place which can detect multiple types of cancer at an early stage, a potential cancer therapy is most likely to succeed. However, at present, the available biomedical sensors are unable to detect and differentiate between cancerous cells or tumours. They are also not able to provide continuous real-time monitoring of a patient. Additionally, oral anticancer drugs given during chemotherapy, at the moment, suffer from low bioavailability. Also, a variety of these drugs is not targeted in nature. That means the drug will potentially affect areas of the body which do not need it. The low bioavailability of the drug will not only increase the chemotherapy sessions but also makes the entire process more aggravating for the cancer patient. Therefore, there is an absolute need to have innovative and efficient anticancer drug delivery mechanisms. Finally, current biomedical sensors are primarily made up of silicon (Si) or hard substrates based materials. Even if the biomedical sensor is of a flexible material, the material is either a fragile film or flexible but not stretchable polymers such as polyimide (PI). By having a biomedical sensor which is moderately flexible or not flexible at all, a continuous on-body biomedical sensing is not possible in an efficient manner. That is because hard substrates based biomedical sensors would be difficult to be placed on a body at all times. Furthermore, the flexible biomedical sensors currently suffer from problems such as the electrode on top cracking and damaging after few uses rendering them unusable. Hence, a new fabrication process needs to be devised to solve the issues mentioned above. In this work, an attempt is made to utilise zinc oxide (ZnO) nanostructures for biomedical sensing, drug delivery and biomimetics. ZnO nanostructures are synthesised by using a low-cost wet chemistry process known as hydrothermal growth. Due to the inherent biocompatibility and unique electrical/ piezoelectric properties of ZnO, they acted as prime candidates for the applications outlined above. A high-throughput process is used to synthesise ZnO nanowires (NWs) on Si, polyimide-onsilicon (PI/Si) and directly on PI and polydimethylsiloxane (PDMS) substrates. The work utilises a variety of characterization tools. ZnO nanostructures' morphology is characterised by using a Scanning Electron Microscope (SEM), Transmission Electron Microscope (TEM) and Atomic Force Microscope (AFM). X-ray diffraction (XRD) was used to calculate the crystallite size and the crystalline orientation of the nanostructures. A novel fabrication process is developed to allow direct synthesis and direct patterning of metal electrodes on fully flexible, stretchable and bendable PDMS substrates by using standard photolithography. This novel fabrication process makes the PDMS substrates not expand when exposed to temperatures up to 110 °C. Also; the new fabrication process does not cause the PDMS to swell when exposed to various chemicals such as isopropyl alcohol (IPA) or acetone. The fabrication process has created a new paradigm shift in the field of patterning and producing devices directly on flexible and stretchable substrates. The PDMS substrate is further utilised as a sensitive bovine serum albumin (BSA) protein sensor which is capable of detecting up to femtomolar concentrations in just under 5 min of incubation time. Protein biosensing tests were carried out by measuring the change in resistance at 1V bias voltage. The PDMS based biosensor is tested as a protein sensor because proteins are important biomarkers in cancer diagnosis. Also, protein sensors are immensely useful in the detection of bacteria and viruses thereby allowing further expansion to the technology developed herewith. For the first time, ZnO NWs are used to deliver hydrophobic organic dye, Nile red, in a human body like environment. The Nile red simulates an anticancer drug as they share similar surface chemistry. There is an approximately 80% release of Nile red which shows that ZnO NWs can be used as an efficient anticancer drug delivery system with high bioavailability. For the drug delivery experiments, the dynamic dialysis based release of Nile red (Nr) from the ZnO nanowires is carried out by using UV-Visible (UV-Vis) spectroscopy. Fourier Transform Infrared (FTIR) was used to determine the coordination of Nr across the ZnO nanowires. Finally, a novel synthesis process is used to produce individual ZnO NWs on a single ZnO nanoplate (NP) which are named as ZNWNP nanostructures. ZNWNP nanostructures have high hydrophobicity without the need of any functionalization. The hydrophobicity of the hybrid ZnO nanowires on ZnO nanoplate nanostructures (ZNWNP) is characterised by using contact angle goniometry (CAG). Various contact angle theories have been used to calculate the surface free energy (SFE) of the ZNWNP nanostructures. The high hydrophobicity allows these nanostructures to be used for biomimetic applications such self-cleaning, bioinspired sensors and multimodal biosensing. Additionally, ZNWNP nanostructures can be used in biomedical sensors to create multimodal analysis. The multimodal analysis is immensely useful in cancer detection as at least three or more cancer biomarkers can be used to triangulate the diagnosis. The work presented in the thesis aims to utilise ZnO nanostructures for a variety of biomedical applications. The new fabrication process mentioned above has applications not only in biomedicine but also in the flexible electronics industry. The biomimetic nanostructures combined with the biomedical sensor gives rise to a robust multimodal analysis system which can change the course of the cancer diagnosis. That coupled with the usage of ZnO NWs as an effective anticancer drug delivery system gives an immense promise in advancing cancer therapy as a whole and making the entire treatment process less aggravating and less painful for cancer patients.

In modern industrial applications, sensors are an expensive part of installed systems. Nevertheless, many system variables cannot be measured sufficiently frequently or accurately. Thus, soft sensors have been developed to estimate those variables without the expense of additional hardware. The use of a soft sensor with a bias update term has shown to perform well for disturbed systems with time delays and multirate sampling times. In industrial application, the time delay and sampling times often vary. Yet, the case of variation of the time delay and sampling time in the bias update term has not been considered in previous publications. This thesis tests a soft sensor with bias update term in simulation and gives a modification yielding better performance. It is shown that the tested method gives unstable results. Hence, a more general method with a bias update term that considers all possible sampling times in each step is proposed, giving stable results in simulation. Furthermore, the stability of the general method is proven mathematically by building a state space representation and applying the Bauer-Premaratne-Dur´an theorem to the stability of switching systems. / Tesis

In a workplace marked by increasing change and competing commitments, business leaders require an increased understanding of alternative work programs. Utilizing spillover theory, motivator-hygiene theory, and adaptive structuration theory, this single case study was an exploration of the strategies that business leaders use for flexible work schedules and virtual work programs. The population consisted of 3 managerial business executives and 6 employees within a midwestern United States division of a global blood management solutions firm. The data collection process included a series of semistructured interviews, a focus group, and the assessment of company documentation. Methodological triangulation identified 5 emerging themes: program assessment and monitoring, standard set of virtual working hours, remote office setup, increased virtual communication, and promotion of quality of life. The transferability of this single case study remains with the reader and future researchers to determine. Future researchers may discover that the findings contribute to social change by better preparing organizations for success while simultaneously positioning individuals to attain optimum balance across life and work responsibilities.

Dans le domaine de l'électronique souple, les substrats flexibles à base de polymères sont de plus en plus utilisés. Si dans les prochaines années, les structures de propagation mises en œuvre sur ce type de substrat véhiculent une puissance, alors la connaissance de la conductivité thermique de ces matériaux est essentielle. Dans ce travail, nous mesurons la conductivité thermique de matériaux de type polymère en utilisant la méthode 3 omégas. Des mesures ont été effectuées sur du polydiméthylsiloxane (PDMS). Un procédé technologique particulier est utilisé pour la fabrication des échantillons de PDMS. De ce fait, les conducteurs métalliques sont encapsulés dans le polymère et non en surface de ce dernier. Mais cela est sans conséquence sur les valeurs de conductivité thermique mesurées. Les procédés photolithographiques utilisés traditionnellement pour réaliser les lignes métalliques sont coûteux et longs. Par conséquent, nous proposons pour ce type de matériaux une méthode alternative pour la réalisation des lignes conductrices grâce à la technologie d'impression par jet d'encre. Les conductivités thermiques du polyimide et polyétheréthercétone ont été mesurées en utilisant la méthode 3omega combinée à la technologie d'impression par jet d'encre.Des simulations numériques basées sur la méthode des éléments finis ont été développées au cours de la thèse. Les mesures expérimentales obtenues sont comparées aux résultats obtenus par une solution analytique et par notre modélisation numérique.Ainsi durant cette thèse nous montrons avec succès la possibilité d'utiliser la technologie d'impression jet d'encre pour mesurer la conductivité thermique d'un substrat souple. / The characterization of polymers is gaining a great attention as they are one of the main constituents of future flexible or organic electronics. Given the fact that thermal management is an important issue in the frame work of flexible electronics, the knowledge of the thermal conductivity of polymer materials is needed. In this work, we propose the measurement of polymer material thermal conductivity using the three omega method. This method requires heating a metallic line conductor placed on the surface of the material under test by an alternating current source. The first measurements were done on polydimethyl siloxane (PDMS) polymer material for which a special procedure that consists in embedding the metallic line conductors near the surface has been applied.In addition to the well-known limitations of photolithography process which are the cost and the process duration, a particular concern lies in the fabrication of the metallic conductors by such process which might be destructive in case of polymer materials. Consequently, we propose an alternative method for this kind of materials based on inkjet printing technology. The thermal conductivities of polyimide and polyetherether ketone have been successfully measured using the three omega method combined with inkjet printing technology for sample preparation. Numerical simulations using finite element method (FEM) are also performed. Finally, experimental measurements are compared to Cahill’s analytical solution and FEM modelling. The overall results demonstrate that the inkjet printing technology is a good candidate for the characterization of flexible materials in terms of thermal conductivity.

Depuis quelques années, nous assistons à l’essor d’une nouvelle filière d’électronique basée sur des supports flexibles. De nombreuses applications difficilement atteignables par l’électronique classique sont visées, c’est notamment le cas des tags RFID, des capteurs mobiles, des écrans flexibles…. Cette électronique est essentiellement basée sur des matériaux organiques pour lesquels la faible mobilité (<1cm2 /V.s) limite considérablement les performances hyperfréquences des composants. Dans ce contexte, l’intégration hétérogène de composants des filières GaN et graphène sur substrat flexible apparait comme une solution prometteuse pour des applications de puissance hyperfréquence où la conformabilité sur surface non plane est souhaitée. Ces travaux présentent d’une part, une méthode de transfert de composants HEMTs AlGaN/GaN sur ruban flexible et d’autre part, une technique de manipulation du substrat souple et de fabrication directe des composants à base de graphène sur celui-ci. Des HEMTs AlGaN/GaN à faible longueur de grille (LG = 100nm) ont été transférés sur ruban flexible et ont permis d’atteindre des résultats à l’état de l’art en termes de puissance hyperfréquence avec un gain de puissance linéaire (Gp) de 15,8 dB, une densité de puissance de sortie (Pout) de 420 mW / mm et une puissance ajoutée (PAE) de 29,6%. Pour les composants à base de graphène, une technique de manipulation du substrat flexible a été développée et a permis de fiabiliser le procédé technologique de fabrication. Une fréquence de coupure ft de 1GHz et une fréquence maximale d’oscillation fmax de 3 GHz ont été obtenues. / In recent years, the field of flexible electronics has been expanding. Many applications difficult to achieve by conventional electronics are targeted as RFID tags, mobile sensors, flexible screens… This field is essentially based on organic material for which the poor mobility (<1cm2 /V.s) limits considerably the device performances. In this context, the heterogeneous integration of GaN and graphene devices on a flexible substrate appears to be a promising solution for microwave power applications where conformability on a non-planar surface is needed. This work presents, on the one hand, a method to transfer AlGaN/GaN HEMTs onto flexible tape and, on the other hand, a technique for handling and manufacturing graphene-based components directly on the flexible substrate. HEMTs with short-gate length (LG = 100 nm) have been transferred onto flexible tape and showed state of the art results in terms of microwave power with a linear power gain (Gp) of 15.8 dB, an output power density (Pout) of 420 mW/ mm and an added power efficiency (PAE) of 29.6%. Concerning graphene-based devices, a flexible substrate handling technique has been developed making the manufacturing process more reliable. A cut-off frequency ft of 1 GHz and a maximum oscillation frequency fmax of 3 GHz were obtained.

Le développement de nombreuses applications nomades, souples, déformables et sur de larges surfaces nécessite la réalisation de circuits mécaniquement flexibles, intégrant des capacités d'intéraction avec l'environnement, de communication et de traitement du signal. Une part importante de ces applications proviennent actuellement de l'industrie de l'électronique organique, mais l'apparition de nouvelles méthodes de fabrication a permis la réalisation de dispositifs à la fois souples mécaniquement et électriquement performants. En outre, les techniques de report sur substrat souple présentent de nombreux avantages (flexibilité mécanique, conservation des propriétés originelles, intégration hétérogène possible). Lors de ces travaux, une procédure d'amincissement puis de transfert sur film souple (métal, verre) des composants CMOS initialement réalisés sur des tranches SOI (silicium sur isolant) a été développée. Cette solution permet la réalisation de transistors MOS flexibles et performants possédant des fréquences caractéristiques fT/fmax de 165/188 GHz. De plus, l'utilisation d'autres matériaux que le plastique permet de modifier les propriétés originelles d'un dispositif en termes de dissipations thermiques ou de distorsions harmoniques par exemple, afin d'en améliorer les performances. La réalisation de composants souples, performants et stables a donc été démontrée. / The ability to realize flexible circuits integrating sensing, signal processing, and communicating capabilities is of central importance for the development of numerous nomadic applications requiring foldable, stretchable and large area electronics. A large number of these applications currently rely on organic electronics, but new fabrication methods permitted to realize flexible mechanically and electrically efficient devices. Besides the transfert on flexible substrates offers many advantages (mechanical flexibility, preservation of original properties, possible heterogeneous integration). In this work, a solution has been developed, based on thinning and transfert onto flexible substrate (metal, glass) of high frequency (HF) CMOS devices initially patterned on conventional silicon-on-insulator (SOI) wafers. This transfer process first enables the fabrication of high performance electronics on metal, with n-MOSFETs featuring characteristic frequencies fT/fmax as high as 165/188 GHz. Secondly, the use of materials other than plastic permit to modify the original properties of a device in terms of thermal dissipation or harmonic distorsions for example, demonstrating flexibility, high performance and stability.