L’objectif du présent travail a été de concevoir et de fabriquer des dispositifs sur silicium pour convertir de l’énergie thermique en une énergie électrique en utilisant le changement de phase liquide-gaz dans le but de générer une variation subite de pression suivie d’une conversion d’énergie mécanique vers une énergie électrique à l’aide d’un piézoélectrique. La construction des dispositifs a dû rester simple, avec des matériaux courants et en respectant des limites dimensionnelles. Empreinte inférieure à un diamètre de 20 mm et une épaisseur en dessous des 2 mm.Les prototypes fabriqués sont composés de 3 plaques en silicium, contenant une chambre d’évaporation, une chambre de condensation et un canal réunissant les deux. Un transducteur piézoélectrique a été reporté sur la chambre de condensation et assure l’étanchéité ainsi que la génération d’énergie électrique.Le processus de conception inclut plusieurs étapes, dont la définition de la géométrie et du type de fluide de travail utilisé en tant qu’agent thermique. Le travail effectué a permis de sélectionner le type de piézoélectrique, sa taille ainsi que sa méthode d’intégration. Une étude a également été conduite pour déterminer la méthode optimale d’assemblage des plaques en silicium.La réalisation pratique des dispositifs a été orientée vers la sélection des meilleurs procédés technologiques pour la fabrication des structures. Toutes les expériences ont été conduites en salle blanche avec utilisation de l’oxydation humide, la photolithographie, la gravure KOH, ainsi que d’une technique d’assemblage des plaques silicium avec utilisation de la résine SU-8 comme couche intermédiaire. En plus, quelques outils spécifiques ont été conçus lors du présent travail, pour faciliter la fabrication des dispositifs, dont un système sous vide dédié à l’assemblage des plaques en silicium.Les dispositifs ont été testés afin d’établir leur mécanisme d’oscillation thermique ainsi que leurs propriétés électriques. L’influence tu taux de remplissage et de la température de surface chaude sur le signal en sortie ont également été étudiées. Le calcul de l’énergie générée a aussi été effectué. Dans la dernière partie de l’étude, des étapes d’optimisation pour les dispositifs développés dans le présent travail sont proposées. / The goal of the present work was to design and fabricate a fully silicon oscillating device that converts thermal energy into electricity, applying phenomena of liquid to gas phase-change and piezoelectricity. It should be characterized by simplicity of construction, small size, and ease of manufacture. The diameter should not exceed 2 cm, while the thickness should be within 2 mm.The device was composed of three Si wafers comprising evaporation and condensing chambers, and the channel connecting these two elements. A PZT-based transducer mounted on top of the structure was applied to ensure energy conversion.The design process included the establishment of the device geometry, the type of the working fluid enclosed inside the system, a type, size and assembly technique of a piezoelectric element, as well as a bonding method of several silicon elements of the device.The practical realization of the designed prototypes was aimed at selecting the most suitable technological processes for structure fabrication. All the experiments had been performed in a clean room environment and employed wet oxidation, photolithography, a well-known, easily available wet chemical etching in KOH solution, and a silicon bonding technique with the use of SU-8 photoresist as an intermediate layer. Additionally, during the practical work a few tools have been designed and developed to enhance the device fabrication, amongst which a vacuum pump dedicated to bond the three silicon wafers as structural elements of the prototypesThe fabricated prototypes were tested in terms of oscillation mechanism and electrical properties. The influence of the filling ratio and the hot temperature value on the generated signal was established. Additionally, the power range of the prototypes has been evaluated. In the last part of the study, optimization steps for the devices developed in the present work have been proposed.

Transdutores baseados em cascas piezocompósitas têm uma vasta aplicação no campo de estruturas inteligentes, principalmente como atuadores, sensores e coletores de energia. Essas estruturas piezocompósitas são geralmente compostas por dois ou mais tipos de materiais, como por exemplo materiais piezelétricos, ortotrópicos elásticos (possuem fibras de reforçamento) e isotrópicos (materiais homogêneos). Vários fatores devem ser considerados no projeto de transdutores baseados em cascas piezocompósitas, como o tamanho, a forma, a localização e a polarização do material piezelétrico, bem como a orientação das fibras do material ortotrópico. O projeto desses transdutores é complexo e trabalhos anteriores envolvendo esses tipos de materiais sugerem utilizar Método de Otimização Topológica (MOT) para aprimorar o desempenho dos transdutores distribuindo o material piezelétrico sobre substratos fixos de materiais isotrópicos e ortotrópicos, ou otimizar a orientação das fibras dos materiais ortotrópicos com material piezelétrico com tamanho, forma e localização previamente estabelecidos. Assim, nesta tese, propõe-se o desenvolvimento de uma metodologia baseada no MOT para projetar transdutores piezocompósitos de casca considerando, simultaneamente, a otimização da distribuição e do sentido de polarização do material piezelétrico, e também a otimização da orientação das fibras de materiais ortotrópicos, que é livre para assumir valores diferentes ao longo da mesma camada compósita. Utilizando essa metodologia, são obtidos resultados numéricos para atuadores e sensores em regime estático e para coletores de energia com circuito elétrico acoplado, em regime dinâmico amortecido. Para os casos dos sensores e dos coletores de energia, também são consideradas as tensões mecânicas na estrutura, as quais devem obedecer os critérios de von Mises (para materiais isotrópicos) e de Tsai-Wu (para materiais ortotrópicos) para que não haja falhas na estrutura, que está sujeita a esforços mecânicos. / Transducers based on laminated piezocomposite shell structures have a wide application in the field of smart structures, especially as actuators, sensors and energy harvesting devices. These piezocomposite structures are generally composed by two or more kinds of materials, such as piezoelectric, isotropic, and elastic orthotropic (fiber reinforcement) materials. Several factors must be considered in the design of piezocomposite transducers, such as size, shape, location and polarization of the piezoelectric material and the fiber orientation of the orthotropic material. The design of these transducers is complex and previous studies involving these types of materials suggest using \"Topology Optimization Method\" (TOM) to enhance the performance of piezoelectric transducers by distributing piezoelectric material over fixed isotropic and orthotropic substrate or to optimize the fiber orientation of orthotropic materials with piezoelectric patches previously established. Thus, this thesis proposes the development of a methodology based on the TOM to design laminated piezocomposite shell transducers by considering simultaneously the optimization of distribution and the polarization direction of the piezoelectric material, and also the optimization of the fiber orientation orthotropic material, which is free to assume different values along the same composite layer. By using this methodology, numerical results are obtained for actuators and sensors under static response, and energy harvesting devices with an electrical circuit coupled, in dynamic damped analysis. In the case of sensors and energy harvesting devices, which are subjected to mechanical loads, the mechanical stresses in the structure are also considered, which must satisfy two stress criteria to prevent failure: von Mises for isotropic materials and Tsai-Wu for orthotropic materials.

A natureza vibratória, oriunda da transferência energética, manifesta-se em todos os sistemas e estruturas. Na engenharia, essa transferência energética se revela como um fenômeno vibratório indesejável ou desejável. Em sistemas mecânicos, o controle dinâmico para mitigação de vibrações indesejáveis se realiza através de diversas técnicas e configurações, entre os mais usuais, o Amortecedor de Massa Sintonizável (AMS). No entanto, sabe-se que seu desempenho é suscetível a alterações nas frequências de operação e na natureza das excitações. Em sistemas elétricos, as vibrações do ambiente podem ser desejáveis e convertidas em energia elétrica útil para a realimentação de rede de sensores sem fio e computação pervasiva. A fim de combinar estes fenômenos e reduzir custos de operação, é necessário projetar dispositivos sintonizáveis robustos capazes de operar eficientemente em uma banda larga de frequências. Portanto, este trabalho visa projetar e analisar experimentalmente um atenuador de vibrações eletromagnético sintonizável (semi-ativo) com captação energética (AEMSCE) através da introdução deliberada de não linearidades. O AEMSCE consiste em um sistema massa-mola-amortecedor não linear com um ímã oscilante central orientado sob forças repulsivas magnéticas e uma bobina instalada, sendo capaz de dissipar as vibrações da estrutura e convertê-las em energia elétrica útil. Os recursos e parâmetros do AEMSCE são apresentados e identificados. O fator de transdução eletromagnético que acopla o sistema mecânico ao elétrico é quantificado. Mostra-se que a variação da distância entre ímãs promove ao sistema ressonância ajustável e que a força de restauração magnética resultante apresenta uma faixa de operação linear. No trabalho, verifica-se que o campo de máxima captação energética está contido na faixa de operação linear confirmando a relevância deste campo linear. O comportamento do sistema é analisado considerando as influências da força restauradora magnética, das forças amortecidas e da força de atrito. A partir dessas análises e das aproximações realizadas, apresenta-se estratégias de controle passivas e técnicas de otimização para mitigação, cuja resultante é um campo de atenuação ótimo, assim como desenvolve-se métodos de otimalidade para maximizar a conversão energética do AEMSCE, cuja resultante é um amortecimento elétrico ótimo ou um amortecimento admissível ótimo (deslocamento máximo). Métodos analíticos e simulações numéricas são desenvolvidos em todo o trabalho com diferentes configurações para analisar a robustez e eficiência do dispositivo, através do comportamento dinâmico vibratório à resposta transiente e estacionária induzido por excitação de base harmônica. De maneira geral, os resultados mostram que os parâmetros de sintonia e amortecimento do AEMSCE podem ser combinados e ajustados para ampliar o controle de vibrações da estrutura e maximizar a captação energética, principalmente na ressonância. Verifica-se que existe uma relação de importância da tensão induzida e do amortecimento elétrico, através da variação da resistência de carga no resistor, com a atenuação e captação energética. Por fim, este trabalho buscou apresentar os melhores métodos e resultados de parâmetros de amortecimento a fim de obter informações como guia de projeto para otimizar os dispositivos futuros e para a proposição de incorporar um controle semiativo ao AEMSCE. Como forma de melhorar o desempenho em aplicações futuras, é possível combinar as propriedades ótimas resultantes e ajustá-las através de estratégias de controle semiativa, explorando a dinâmica linear e não linear do sistema. / Vibratory nature, derived from the energy transfer, manifests itself in all systems and structures. In engineering, this energy transfer is revealed as an undesirable or desirable vibrational phenomenon. In mechanical systems, the dynamic control to mitigate undesirable vibrations is achieved through several techniques and configurations, among the most usual, the Tunable Mass Damper (TMD). However, it is known that their performance is susceptible to changes in the operating frequencies and the nature of the excitations. In electrical systems, ambient vibrations may be desirable and converted into useful electrical energy for the feedback of wireless sensors network and pervasive computing. In order to combine these phenomena and reduce operating costs, it is necessary to design robust tunable devices capable of operating efficiently over a wide frequency band. Therefore, this work aims to design and experimentally analyze a tunable electromagnetic vibrations absorver (semi-active) with energy harvesting (TEMAEH) through the deliberated introduction of non-linearities. TEMAEH consists of a non-linear mass-spring-damper system with a central oscillating magnet oriented under magnetic repulsive forces and a coil installed, being able to dissipate vibrations of the structure and convert them into useful electrical energy. The TEMAEH features and parameters are presented and identified. Electromagnetic transduction factor that couples the mechanical to electrical system is quantified. It is shown that the variation of the distance between magnets provides adjustable resonance to the system and that the resulting magnetic restoring force has a linear operating range. In the work, it is verified that the field of maximum energy harvesting is contained in the linear operating range confirming the relevance of this linear field. The behavior of the system is analyzed considering the influences of magnetic restoring force, damped forces and frictional force. From these analyzes and the approximations performed, passive control strategies and optimization techniques for mitigation are presented, resulting in an optimum attenuation field, as well as optimization methods to maximize the energy conversion of the TEMAEH, resulting in an optimum electric damping or optimum permissible damping (maximum displacement). Analytical methods and numerical simulations are developed throughout the work with different configurations to analyze the robustness and efficiency of the device through the dynamic behavior of vibration to the transient and stationary response induced by harmonic based excitation. In general, the results show that the tuning and damping parameters of the TEMAEH can be combined and adjusted to increase the vibration control of the structure and to maximize energy harvesting, especially in resonance. It is verified that there is a relation of importance of the induced voltage and the electrical damping, through the variation of the load resistance in the resistor, with the attenuation and power generation. Finally, this work sought to present the best methods and results of damping parameters in order to obtain information as a project guide to optimize future devices and for the proposition to incorporate a semiative control to TEMAEH. As a way to improve performance in future applications, it is possible to combine the resulting optimal properties and adjust them through semiative control strategies, exploring the linear and non-linear dynamics of the system.

The design and implementation of self-powered, low power implant microcontroller, with wireless data transmitter system that captures data as subcutaneous bio-sensing platform has been achieved with glucose fuel cell (GFC) energy harvesting power solution. Data transfer is unidirectional, implant to reader and is initiated by a single transmission from the external reader. The implant's memory contents are transmitted as a stream of wireless pulses to the reader. This work explored two different approaches on current technologies used for designing self-powered bio medical devices (BMDs) and active implantable medical devices (IMDs), their processing, sampling data, transmission of data and energy hasting powering techniques with a view to identifying state-of-the art technologies and methods to improve the long-term powering and recharging of IMDs via a highly safe, efficient and convenient way. The designed low power implant microcontroller, with wireless data transmitter system combines glucose energy harvesting technique by using materials with efficient catalyst capabilities based on platinum nanoparticles supported on Vulcan carbon cloth (PtVCC) as a cathode electrode for GFC configuration, while plain Platinum (Pt) mesh/sheet acted as anode. The PtVCC and Pt electro-reaction, catalytic activities and stability resulted in a design of a direct GFC with high output voltage and current, >0.4V and >300μA respectively per cell, and increased this voltage to value >4V, to power the implant system, by using a voltage booster; direct current to direct current (DC-DC) converter circuit, and a rechargeable battery. The innovative self-powered bio-sensing platform integrating GFC design, meets the self-powered IMDs expectations in terms of simplified fabrication and materials that allows one-compartment design that can directly be placed on the surface of medical implant to provide sufficient output power boosted by DC-DC converter to produced higher output voltage ten times greater than the input value, enough to power most efficient electronic devices. This research therefore proposes the practicability and potential of designing and implementing a wireless bio sensor system powered by an energy harvesting solution, based on GFC to produce a proof-of-concept design system and integration, including power management and data communication (sampling and transmission) platform suitable for self-low-powered periodically-activated IMD.

La croissance formidable des dispositifs sans fils et autonomes (réseaux de capteurs, objets connectés…) voit actuellement son développement limité par les batteries qui présente une durée de vie limitée et ainsi soulève des problèmes de maintenance. Afin de palier à cette limitation, l’utilisation de l’énergie directement disponible dans l’environnement immédiat du dispositif, conduisant au concept de « récupération d’énergie », est une voie fortement explorée depuis une dizaine d’années. Ainsi, l’objectif de cette thèse a été de développer de nouvelles techniques et/ou d’utiliser de nouveaux principes de conversion afin de proposer des alternatives aux techniques de récupération d’énergie classiques. Dans un premier temps, l’optimisation de récupérateurs électrostatiques a été étudiée. Les performances de ces systèmes étant fortement liées à la variation de capacité, une structure fractale, permettant un accroissement important des surfaces en regard entre deux électrodes (et donc de la capacité) lorsque ces dernières sont proches, a été proposée et modélisée. Il est ainsi montrer un accroissement significatif des possibilités de récupération d’énergie ; ces dernières étant étroitement liées à l’amplitude de vibration du système. Le second axe de recherche de cette thèse s’est attelé à développer un modèle haut niveau simple mais précis pour les structure utilisant des polymères électrostrictifs fonctionnant en flexion. Une analyse énergétique a permis de mettre en place un modèle électromécanique masse-ressort-amortisseur couplé avec une source de courant contrôlée par les excitations mécaniques et électriques du système, permettant ainsi une conception plus aisée du microgénérateur. Enfin, la dernière partie de cette thèse s’est intéressée à la conversion d’énergie thermique utilisant la variation de perméabilité des matériaux ferromagnétiques, ouvrant de nouvelles possibilités de conversion de l’énergie. En particulier, une technique simple et autonome consiste à créer un champ magnétique de polarisation à l’aide d’un aimant, permettant une variation du flux magnétique lors d’un changement de température, qui peut être converti sous forme électrique à l’aide d’un bobinage. / This thesis has been devoted to electrostatic mechanical energy harvesting based on capacitors inspired by fractal geometry, to mechanical energy harvesting based on beams with electrostrictive polymers, and to thermal energy harvesting based on ferromagnetic materials. For electrostatic energy harvesting without electrets, interdigitated capacitors are usually applied as in-plane overlap varying and in-plane gap closing electrostatic generators. In consideration of the limit of aspect ratio for fingers in the capacitor, we would like to improve the capacitor configuration by taking advantage of self-similarity patterns. The concept is to gradually add fingers of smaller widths between original ones to form a mountain-shape capacitor. According to the different width ranges of capacitors, they are classified as of different orders whose performances vary with the vibration amplitude. Harvested energy over one cycle for capacitors of order 1, 2 and 3 has been demonstrated by theoretical and FEM results. In application, the order of capacitor needs to be properly chosen to maximize the harvested energy. Electrostrictive polymer (polyurethane) has been utilized along with a beam to perform mechanical energy harvesting. Two models have been analyzed: clamped-free beam with a polymer film attached at the clamped end, clamped-free bimorph beam. The simple model for electrostrictive devices under flexural solicitation is set up on the base of analysis of energy conversion and it shows that the electrostrictive system can be reduced to a simple spring-mass-damper system with a quadratic dependence with the applied voltage on the mechanical side and to a current source controlled by the applied voltage with a capacitive internal impedance on the electrical side. Experiments based on the clamped-free beam with a polymer film attached to the clamped end have been carried out to evaluate the mechanical to electrical conversion. The thermal energy generator is based on a ferromagnetic material, a magnet and a coil. As the magnetic permeability of ferromagnetic materials encounters drastic variation around the Curie temperature, the concept of the generator is to take advantage of the permeability variation caused by temperature decrease to generate sharp variation in magnetic flux which induces a current in the coil. According to theoretical results, the generated current is closely related to the temperature variation and the variation velocity. Experiments have been carried out on Ni30Fe of which the Curie temperature is 55 ºC. When the temperature decreases from 20.5 ºC to -42.4 ºC, the maximum power is about 4×10^(-7)W with the load to be 2 Ω.

Les matériaux actifs, tels que les matériaux piézoélectriques et électrostrictifs, sont couramment utilisés dans la conception de dispositifs exploitant leurs propriétés respectives. La propriété principale de ces matériaux réside dans le fort couplage entre les comportements électrique et mécanique (piézoélectricité). Dans la majorité des cas, ces matériaux sont utilisés séparément. L’utilisation combinée de ces matériaux permet la réalisation de dispositifs innovants basés sur l’effet électrostrictifs: l’apparition d’une polarisation électrique induite par une contrainte mécanique et réciproquement l’apparition d’une déformation mécanique sous l’action d’un champ électrique. Les applications « support » concernent les capteurs et les actionneurs. L’étude de ce couplage passe par la caractérisation de ces matériaux, puis par la mise en place de modèles décrivant finement leurs comportements et enfin par le développement d’outils pour la conception. L’objectif de la thèse est de remplacer le matériau céramique, rigide et à faible déformation, par un film polymère nanocomposite électroactifs, présentant des grandes déformations et forces d'actionnement sous champ électrique modéré grâce à l'incorporation dans la matrice polymère de micro et nano-objets (charge) conducteurs ou semi-conducteurs. De plus, pour des applications plus spécifiques de la récupération d’énergie, la charge du film polymère par des micro et nano-objets conducteurs sera également étudiée. Idéalement, il serait très intéressant de réaliser un matériau multifonctionnel, sensible à la fois à une stimulation mécanique (propriétés de détection et/ou de récupération d’énergie par couplage électromécanique). / In the last decades, direct energy conversion devices for medium and low grades waste heat have received significant attention due to the necessity to develop more energy efficient engineering systems. A great deal of research has in recent years been carried out on harvesting energy using piezoelectric, electrostatic, electromagnetic , and thermoelectric ,transduction, with the aim of harvesting enough energy to enable data transmission. For this purpose, piezoelectric elements have been extensively used in the past; however they present high rigidity and limited mechanical strain abilities as well as delicate manufacturing process for complex shapes, making them unsuitable in many applications. Thus, recent trends in both industrial and research fields have focused on electrostrictive polymers for electromechanical energy conversion. This interest is explained by many advantages such as high productivity, flexibility, and processability. Hence, electrostrictive polymer films are much more suitable for energy harvesting devices requiring high flexibilities, such as systems in smart textiles and mobile or autonomous devices. Electrostrictive polymers can also be obtained in many different shapes and over large surfaces. . In the last years, electrostrictive polymers have been investigated as electroactive materials for energy harvesting. However for scavenging energy a static field is necessary, since this material is isotope, there is no permanent polarization compare to piezoelectric material. A solution for avoid this problem; concern the hybridization of electrostrictive polymer with electret. Finally, the implementation of electrostrictive materials is much simpler for small-scale systems (MEMS). Hence, several studies have analyzed the energy conversion performance of electrostrictive polymers, both in terms of actuation and energy harvesting.

Ces travaux de thèse se veulent exploratoires et visent à proposer d’une part une nouvelle approche hybride piézoélectrique-polymère pour le développement de récupérateur d’énergie et d’autre part à développer des méthodes d’actionnement destinée au contrôle de forme structurelle. L’utilisation de polymères diélectriques dans les dispositifs de récupération d’énergie permet de concilier une densité énergétique élevée avec une faible rigidité et une simplicité de mise en œuvre. L’approche hybride proposée dans le cadre de ces travaux permet de s’affranchir de la source de polarisation externe en la substituant par un dispositif piézoélectrique apte à générer, sous l’action d’une contrainte mécanique, le champ électrique nécessaire au fonctionnement du dispositif. L’optimisation du transfert d’énergie entre deux systèmes supposé quelconque a d’abord été généralisée. Les travaux ont ensuite été orientés vers l’investigation de la faisabilité de l’approche hybride piezo-polymère. Les performances des dispositifs hybrides ont été évaluées expérimentalement et comparées à celles obtenues avec des récupérateurs piézoélectriques simples et avec des récupérateurs polymères à polarisation externe. Il est montré que l’utilisation d’un dispositif hybride permet de concilier les qualités des deux approches simples, à savoir la faculté d’initier le processus de conversion d’énergie grâce au champ électrique généré par l’élément piézoélectrique et d’exploiter la haute densité d’énergie des polymères. La deuxième partie de ces travaux porte sur l’utilisation de matériaux électroactifs comme dispositifs d’actionnement destinés au contrôle de forme de structures minces. L’application finale visée ici est le développement de matériau de type peau active. Différents types de polymères ont d’abord été testés et leur performances ont été comparées avec des modèles théoriques spécifiquement développés. Des structures originales ont été proposées pour solutionner certains problèmes liés à l’actionnement par polymères diélectriques. Des prototypes simples ont permis de valider le principe du contrôle de forme des structures à l’aide de polymères diélectriques. / Dielectric polymers have seen their importance grow in the field of electroactive materials because of their undeniable advantages, particularly for potential applications such as energy harvesting, actuation or sensors. The work done in this thesis is exploratory and aims primarily to provide on one hand a new piezoelectric-polymer hybrid approach for the development of energy harvesting systems and secondly to develop operating methods shape control of structures using electroactive polymers. The use of dielectric polymers in energy harvesting devices reconciles a high energy density with low rigidity and simplicity of integration. The main problem which is characteristic of such devices is that they necessarily require the use of an external bias high voltage supply (> 1kV) to achieve significant energy densities. The hybrid approach proposed in the context of this work eliminates this source of external energy by using a piezoelectric device capable of generating, under the action of a mechanical stress, the electric field required to operate the device. The problem of optimization of energy transfer between any two systems was also studied. The work was then directed towards the investigation of the feasibility of the piezo-polymer hybrid approach. Various configurations have been proposed and evaluated in order to deduce their optimal parameters. The performance of hybrid devices was experimentally evaluated and compared with that obtained with simple piezoelectric or electrostatic (using polymers) systems. The second part of this work focuses on the use of electroactive materials as actuators for shape control of thin structures. The final application aimed here is the development of an active skin type material allowing reconfiguration of orbiting satellite antennas. Different types of polymers were first tested and their performance has been compared with the theoretical models developed specifically in this context. Original structures have been proposed to solve some problems related to the actuation using dielectric polymers. Simple prototypes have validated the principle of the structural shape control. If their use brings undeniable advantages over conventional operating techniques, the fact remains that certain specific characteristics of electroactive polymers limit their performance. For example, the square law characteristic of the electroactive polymer control requires the use of particular geometries in order to obtain a symmetrical two-way displacement. This fact complicates the control of such actuators but allows in the end to add new features. Thus the use of a sectored network of polymer actuator is required to obtain a symmetrical movement on a single type of structure blocked blocked-beam, but allows to consider different deformation profiles. Other similar problems have been addressed using different original structures.

Le développement des systèmes de récupération d'énergie est liée à l'émergence des applications de type Internet des objets (IoT) plus spécifiquement à la prolifération des réseaux de capteurs autonomes. Les progrès réalisés ces dernières années dans le domaine des technologies de l’information et de la communication ont permis de lever certains verrous technologiques au développement de ces réseaux de capteurs intelligents et autonomes, notamment grâce à l’amélioration des performances intrinsèques des composants microélectroniques (vitesse, consommation), la conception de circuits plus économes en énergie, ou bien la mise en place de standards de communications radio adaptés à ces contraintes énergétiques. Etant donné l’ubiquité des sources d’énergie, la fabrication de générateurs permettant d’alimenter directement ces capteurs et les rendre autonomes en énergie à partir de ces sources représente une alternative viable à l’utilisation de batteries pour prolonger la durée de vie de ces capteurs communicants. Diverses technologies de générateurs ont ainsi été proposées pour s’adapter aux différentes formes que peut prendre l’énergie, qu’elle soit d’origine thermique, mécanique ou solaire. Le présent travail est une contribution à l'élaboration d’un récupérateur d’énergie thermique à base de bilames thermiques et de matériaux piézoélectriques. Ce type de générateurs, proposé et développé au sein de STMicroelectronics à Crolles, se veut être une alternative fiable et bas coût à l’utilisation de matériaux thermoélectriques exploitant l’effet Seebeck pour générer de l’énergie électrique. Des preuves de concept de tels systèmes ont déjà été développées aussi bien à macro-échelle qu’a micro-échelle. Ce travail s’inscrit dans la continuité du développement d’un récupérateur d’énergie macroscopique reposant sur ce principe-là. L’objectif de cette thèse est dans un premier temps d’optimiser cette structure pour atteindre des niveaux de puissances plus élevés que la première preuve de concept puis dans un second temps, de réaliser son intégration afin d’effectuer des démonstrations de capteur autonome et confirmer la viabilité de la technologie développée pour de telles applications. / The development of energy harvesting systems is linked to the emergence of the Internet of Things (IoT) more specifically to the proliferation of Wireless Sensor Networks that should respond to the growing needs for monitoring data in domains as diverse as the industry, the urban environments, the home or even the human body. Recent progress in the CMOS technology have enabled to remove some of the technical obstacles to the deployment of these smart and autonomous devices, specifically thanks to the improvements of the performances of microelectronic components, the design of ultra-low-power circuits and even the creation of wireless communication standards well adapted to the needs of wireless sensors. Given the availability of ambient energy sources like mechanical, thermal, light etc., energy harvesters are becoming reliable alternatives to batteries in order to extend the autonomy of these sensors. Consequently, various technologies of generators have been developed to harvest different kind of energies in function of their availability. The present work is a contribution to the development of a thermal energy harvester based on bimetallic strips heat engine and piezoelectric membranes. This type of technology developed by STMicroelectronics is intended to be a low cost alternative to thermoelectric generators exploiting the seebeck effect to convert heat into electricity. Based on this working principle, many harvesters both at the micro and macro scale have been fabricated. This thesis deals with the development of macroscopic energy harvesters whose first proofs of concept were established in a previous thesis. An important part of this manuscript deals with the thermal optimization of this energy harvester both in static and dynamic modes. Once the thermal properties improved, various piezoelectric materials were tested and compared to find the most adapted ones to our application and the same work is realized to choose the best device’s architecture. The integration of the energy harvester is then realized and wireless sensor node applications are demonstrated using various communication protocols and sensors. SPICE modeling of the system is also made and coupled with simulations of power management circuits developed by CEA’s design team. Finally, alternative ways to exploit wasted heat and vibrations are proposed through the development of piezoelectric bimetals and dual energy harvesters able to harvest thermal energy and mechanical energy at the same time: piezoelectric bimetals are realized either by direct deposition of piezoelectric composites or piezoelectric thin films onto bimetals. In the case of the dual energy harvester, piezoelectric cantilever beams were designed and simulated to vibrate at low frequencies (between 50Hz and 125Hz).

In telecommunications, distributed cooperative communications refer to techniques which allow different users in a wireless network to share or combine their information in order to increase diversity gain or power gain. Unlike conventional point-to-point communications maximizing the performance of the individual link, distributed cooperative communications enable multiple users to collaborate with each other to achieve an overall improvement in performance, e.g., improved range and data rates. The first part of this dissertation focuses the problem of jointly decoding binary messages from a single distant transmitter to a cooperative receive cluster. The outage probability of distributed reception with binary hard decision exchanges is compared with the outage probability of ideal receive beamforming with unquantized observation exchanges. Low- dimensional analysis and numerical results show, via two simple but surprisingly good approximations, that the outage probability performance of distributed reception with hard decision exchanges is well-predicted by the SNR of ideal receive beamforming after subtracting a hard decision penalty of slightly less than 2 dB. These results, developed in non-asymptotic regimes, are consistent with prior asymptotic results (for a large number of nodes and low per-node SNR) on hard decisions in binary communication systems. We next consider the problem of estimating and tracking channels in a distributed transmission system with multiple transmitters and multiple receivers. In order to track and predict the effective channel between each transmit node and each receive node to facilitate coherent transmission, a linear time-invariant state- space model is developed and is shown to be observable but nonstabilizable. To quantify the steady-state performance of a Kalman filter channel tracker, two methods are developed to efficiently compute the steady-state prediction covariance. An asymptotic analysis is also presented for the homogenous oscillator case for systems with a large number of transmit and receive nodes with closed-form results for all of the elements in the asymptotic prediction covariance as a function of the carrier frequency, oscillator parameters, and channel measurement period. Numeric results confirm the analysis and demonstrate the effect of the oscillator parameters on the ability of the distributed transmission system to achieve coherent transmission. In recent years, the development of efficient radio frequency (RF) radiation wireless power transfer (WPT) systems has become an active research area, motivated by the widespread use of low-power devices that can be charged wirelessly. In this dissertation, we next consider a time division multiple access scenario where a wireless access point transmits to a group of users which harvest the energy and then use this energy to transmit back to the access point. Past approaches have found the optimal time allocation to maximize sum throughput under the assumption that the users must use all of their harvested power in each block of the "harvest-then-transmit" protocol. This dissertation considers optimal time and energy allocation to maximize the sum throughput for the case when the nodes can save energy for later blocks. To maximize the sum throughput over a finite horizon, the initial optimization problem is separated into two sub-problems and finally can be formulated into a standard box- constrained optimization problem, which can be solved efficiently. A tight upper bound is derived by relaxing the energy harvesting causality. A disadvantage of RF-radiation based WPT is that path loss effects can significantly reduce the amount of power received by energy harvesting devices. To overcome this problem, recent investigations have considered the use of distributed transmit beamforming (DTB) in wireless communication systems where two or more individual transmit nodes pool their antenna resources to emulate a virtual antenna array. In order to take the advantages of the DTB in the WPT, in this dissertation, we study the optimization of the feedback rate to maximize the energy efficiency in the WPT system. Since periodic feedback improves the beamforming gain but requires the receivers to expend energy, there is a fundamental tradeoff between the feedback period and the efficiency of the WPT system. We develop a new model to combine WPT and DTB and explicitly account for independent oscillator dynamics and the cost of feedback energy from the receive nodes. We then formulate a "Normalized Weighted Mean Energy Harvesting Rate" (NWMEHR) maximization problem to select the feedback period to maximize the weighted averaged amount of net energy harvested by the receive nodes per unit of time as a function of the oscillator parameters. We develop an explicit method to numerically calculate the globally optimal feedback period.

Avec le développement des circuits intégrés à très faible consommation d'énergie, la nécessité de réduire les coûts d'exploitation des dispositifs électroniques embarqués et l'utilisation des piles usagées constituant une menace pour l'environnement, le concept de récupération d'énergie a acquis un nouvel intérêt. La récupération d'énergie couvre le piégeage de nombreuses sources d'énergie ambiantes perdues et leur conversion en énergie électrique. Une large gamme de dispositifs de récupération d'énergie des vibrations mécaniques a été développée. Une configuration commune consiste en un système de masse-ressort avec un matériau piézoélectrique en parallèle avec le ressort pour convertir une partie de l'énergie mécanique pendant les oscillations en énergie électrique. Jusqu'à présent, le matériau le plus utilisé pour la récupération d'énergie piézoélectrique est le titano-zirconate de plomb (PbZr1-xTixO3) (PZT). Le PZT est le matériau de référence pour les applications microsystème électromécanique-MEMS (MechanoElectroMechanicalSystems) dans le domaine de la récupération d'énergie. Les matériaux piézoélectriques à base de plomb tels que le PZT et niobate-titanate de plomb-magnésium (PMN-PT) offrent des facteurs de couplage piézoélectriques supérieurs à ceux d'autres matériaux. Cependant, malgré ses excellentes propriétés électriques (diélectriques, ferroélectriques et piézoélectriques), le PZT et d'autres matériaux à base de plomb devraient bientôt être remplacés par des composés sans plomb, à cause des problèmes environnementaux. Notre travail vise à développer des matériaux sans plomb de haute performance pour la récupération d'énergie par vibration mécanique. Nous nous sommes intéressés à la fabrication et la caractérisation des dispositifs MEMS pour la récupération d'énergie en utilisant les matériaux piézoélectriques sans plomb tels que le nitrure d'aluminium (AIN), le titanate de baryum BaTiO3 (BTO) et la ferrite de bismuth BiFeO3 (BFO). Les matériaux piézoélectriques PZT (utilisé comme référence à cause ses coefficients piézoélectriques élevés), BTO, BFO et AIN ont été déposés en utilisant des méthodes de dépôt telles que la pulvérisation cathodique et le dépôt sol-gel, conduisant à des films minces à grande échelle, homogènes et de haute densité, avec une épaisseur contrôlée avec précision. Le dépôt de films de 300 nm d'épaisseur par pulvérisation cathodique ou par Sol-Gel a été réalisé sur du substrat de SrTiO3 (STO) recouvert d'une électrode inférieure de SrRuO3 (SRO), qui est le substrat de référence pour les oxydes fonctionnels (PZT, BTO et BFO), et sur un substrat de silicium recouvert de platine, qui est le modèle industriel classique. Quels que soient les matériaux piézoélectriques, nous avons obtenu des films épitaxiés sur substrat de STO et texturés sur substrat de silicium. Des mesures structurales, électriques et piézoélectriques sur les films de BTO, AIN et PZT montrent qu'ils ont de bonnes propriétés physiques en accord avec la littérature. / With the development of ultra-low-power integrated circuits, the need to reduce operating costs for embedded electronic devices, and since used batteries pose a threat to the environment, the concept of energy harvesting has gained a new relevance. Energy harvesting covers the scavenging of many lost ambient energy sources and their conversion into electrical energy. A broad range of energy harvesting devices has been developed to scavenge energy from mechanical vibrations. A common configuration consists of a spring-mass system with a piezoelectric material in parallel with the spring to convert some of the mechanical energy during oscillations into electrical power. So far the most used material for piezoelectric energy harvesting is the Lead Zirconate Titanate (PbZr1-xTixO3) (PZT). PZT is the reference material for MEMS (MechanoElectroMechanicalSystems) applications in the field of energy harvesting. Lead-based piezoelectric materials such as PZT and lead magnesium niobate-lead titanate (PMN-PT) offer incomparable piezoelectric coupling factors to other materials. However, despite its excellent electrical properties (dielectric, ferroelectric and piezoelectric), PZT and other Lead based materials should be replaced shortly by leadfree compounds, due to environmental issues. Our work aims at developing lead-free high performance vibration energy-harvesting. We focus on the fabrication and characterization of aluminum nitride (AlN), Barium titanate BaTiO3 (BTO) and Bismuth ferrite BiFeO3 (BFO) devices for energy harvesting. PZT (as a reference because it’s high piezoelectric coefficients), BTO, BFO and AlN have been deposited using sputtering methods, leading to high homogeneous, large scale thin films with a precisely controlled thickness. The deposition of 300nm-thick films by sputtering or spin coating was performed on SrTiO3 (STO) substrate with SrRuO3 (SRO) bottom electrode, which is the reference substrate for the functional oxides (PZT, BTO and BFO), and platinum coated silicon substrate, which is the classic industrial template. Whatever the piezoelectric materials, we obtained epitaxial films on STO substrate and textured films on silicon substrate. Structural, electrical and piezoelectric measurements on the BTO, AlN and PZT films show that they have good physical properties in agreement with the literature.