L’éolien est un secteur industriel en pleine expansion, qui joue un rôle fondamental dans le développement des énergies renouvelables. Cependant ces machines sont performantes sur une plage de fonctionnement étroite. Afin d’adapter l’éolienne aux changements de vent, une solution actuellement mise en place sur certaines éoliennes commerciales consiste à faire varier l’angle de calage (ie l’inclinaison) des pales au cours de son fonctionnement. Cette méthode de contrôle actif élargit la plage de hauts rendements ainsi que la plage de fonctionnement global, et améliore le démarrage de l’éolienne, mais elle n’augmente pas le rendement maximal atteint par une éolienne à angle de calage optimal fixé. Cependant la complexité́ de cette méthode ainsi que ses coûts de conception, de construction et de maintenance la rende inaccessible pour beaucoup d’éoliennes, en particulier celles de petite taille. Récemment des recherches se sont orientées vers un contrôle passif de l’angle de calage. Dans cette thèse nous examinons expérimentalement et théoriquement l’intérêt d’utiliser des pales flexibles suivant la corde sur une éolienne à axe horizontal. L’étude se concentre sur deux questions : - comprendre le mécanisme de reconfiguration de la pale flexible bio-inspirée : la déformation de la pale est due à la compétition entre les forces aérodynamiques, qui augmentent l’angle de calage moyen, et la force centrifuge qui le diminue. Ces effets sont gouvernés par deux nombres adimensionnés, respectivement le nombre de Cauchy et le nombre centrifuge. - qualifier et quantifier le gain en performances de l’éolienne : une flexibilité́ de pale modérée élargit la plage de fonctionnement, et augmente significativement le rendement de l’éolienne, expérimentalement jusqu’à 35% sur la plage de hauts rendements. Une procédure d’optimisation visant à déterminer le matériau optimal de la pale flexible est présentée. Ces gains obtenus en régime stationnaire sont conservés expérimentalement en moyenne en régime instationnaire. Deux temps caractéristiques sont identifiés : le temps de reconfiguration de la pale flexible et le temps de variation de la fréquence de rotation de l’éolienne / Wind energy is a rapidly growing branch of industry, playing a significant role in the development of renewable energies. However these machines are efficient only on a narrow working range. In order to adapt wind turbines to wind changes, some commercial machines are pitch controlled during rotation. This active control method extends the high-efficiency range and the total working range, and improves the starting phase, but it does not increase the maximum efficiency reached by a wind turbine with the fixed optimal pitch angle. However this method is complex and costly (design, construction, maintenance). Thus it becomes cost-effective only for large wind turbines. Research recently focused on passive pitch control. In this thesis, the contribution of chord wise flexible blades is studied both experimentally and theoretically. The thesis concentrates on: - the reconfiguration mechanism of the bio-inspired flexible blade : the deformation is the result of the competition between aerodynamic forces, which increase the pitch angle, and the centrifugal force, which reduces it. These two effects are governed by two dimensionless numbers, respectively the Cauchy number and the centrifugal number. - how to qualify and quantify the efficiency gains : a moderate flexibility extends the working range, and significantly increases wind turbine efficiency, up to 35% on the high-efficiency working range. An optimization procedure is presented, which aims at determining the optimal material to construct the blade. These improvements measured in steady regime are maintained on average when rotational speed is unsteady. Two characteristic times are identified: the reconfiguration time of the flexible blade and the time of variation of the rotational speed of the wind turbine

The stomatal network in plants is a well-characterized biological system that hypothetically solves the constrained optimization problem of maximizing CO2 uptake from the air while constraining evaporative water loss during the process of photosynthesis. There are numerous such constrained optimization problems present in the real world as well as in computer science. This thesis work attempts to solve one such constrained optimization problem in a distributed manner by taking a cue from the dynamics of stomatal networks. The problem considered here is Dynamic Thermal Management (DTM) in a multi-processing element system in computing. There have been several approaches in the past that tried to solve the problem of DTM by varying the frequency of operation of blocks in the computing system. The selection of frequencies for DTM such that overall performance is maximized while temperature is constrained is a non-deterministic polynomial-time (NP) hard problem. In this thesis, a distributed approach to solve the problem of DTM using a cellular neural network is proposed. A cellular neural network is used to mimic the stomatal network with slight variations based on the problem considered.

Electronic noses combine an array of cross-selective gas sensors with a pattern recognition engine to identify odors. Pattern recognition of multivariate gas sensor response is usually performed using existing statistical and chemometric techniques. An alternative solution involves developing novel algorithms inspired by information processing in the biological olfactory system. The objective of this dissertation is to develop a neuromorphic architecture for pattern recognition for a chemosensor array inspired by key signal processing mechanisms in the olfactory system. Our approach can be summarized as follows. First, a high-dimensional odor signal is generated from a chemical sensor array. Three approaches have been proposed to generate this combinatorial and high dimensional odor signal: temperature-modulation of a metal-oxide chemoresistor, a large population of optical microbead sensors, and infrared spectroscopy. The resulting high-dimensional odor signals are subject to dimensionality reduction using a self-organizing model of chemotopic convergence. This convergence transforms the initial combinatorial high-dimensional code into an organized spatial pattern (i.e., an odor image), which decouples odor identity from intensity. Two lateral inhibitory circuits subsequently process the highly overlapping odor images obtained after convergence. The first shunting lateral inhibition circuits perform gain control enabling identification of the odorant across a wide range of concentration. This shunting lateral inhibition is followed by an additive lateral inhibition circuit with center-surround connections. These circuits improve contrast between odor images leading to more sparse and orthogonal patterns than the one available at the input. The sharpened odor image is stored in a neurodynamic model of a cortex. Finally, anti-Hebbian/ Hebbian inhibitory feedback from the cortical circuits to the contrast enhancement circuits performs mixture segmentation and weaker odor/background suppression, respectively. We validate the models using experimental datasets and show our results are consistent with recent neurobiological findings.

This dissertation discusses an implementation of a design, control and motion planning for a novel extendable modular redundant robotic manipulator in space constraints, which robots may encounter for completing required tasks in small and constrained environment. The design intent is to facilitate the movement of the proposed robotic manipulator in constrained environments, such as rubble piles. The proposed robotic manipulator with multi Degree of Freedom (m-DOF) links is capable of elongating by 25% of its nominal length. In this context, a design optimization problem with multiple objectives is also considered. In order to identify the benefits of the proposed design strategy, the reachable workspace of the proposed manipulator is compared with that of the Jet Propulsion Laboratory (JPL) serpentine robot. The simulation results show that the proposed manipulator has a relatively efficient reachable workspace, needed in constrained environments. The singularity and manipulability of the designed manipulator are investigated. In this study, we investigate the number of links that produces the optimal design architecture of the proposed robotic manipulator. The total number of links decided by a design optimization can be useful distinction in practice. Also, we have considered a novel robust bio-inspired Sliding Mode Control (SMC) to achieve favorable tracking performance for a class of robotic manipulators with uncertainties. To eliminate the chattering problem of the conventional sliding mode control, we apply the Brain Emotional Learning Based Intelligent Control (BELBIC) to adaptively adjust the control input law in sliding mode control. The on-line computed parameters achieve favorable system robustness in process of parameter uncertainties and external disturbances. The simulation results demonstrate that our control strategy is effective in tracking high speed trajectories with less chattering, as compared to the conventional sliding mode control. The learning process of BLS is shown to enhance the performance of a new robust controller. Lastly, we consider the potential field methodology to generate a desired trajectory in small and constrained environments. Also, Obstacle Collision Avoidance (OCA) is applied to obtain an inverse kinematic solution of a redundant robotic manipulator.

The thesis focuses on special space--time computational techniquesintroduced recently for computational aerodynamics modeling of flapping wings of an actual locust. These techniques complement the Deforming-Spatial-Domain/Stabilized Space--Time (DSD/SST) formulation, which is the core computational technique. The DSD/SST formulation was developed for flows with moving interfaces, and the version used in the computations is "DST/SST-VMST," which is the space--time version of the residual-based variational multiscale (VMS) method. The special space--time techniques are based on using NURBS basis functions for the temporal representation of the motion of the locust wings. The motion data is extracted from the high-speed video recordings of a locust in a wind tunnel. In addition, temporal NURBS basis functions are used in representation of the motion of the volume meshes computed and also in remeshing. These ingredients provide an accurate and e fficient way of dealing with the wind tunnel data and the mesh. The thesis includes a detailed study on how the spatial and temporal resolutions influence the quality of the numerical solution.

No description available.

Wireless sensor networks (WSNs) have emerged in strategic applications such as target detection, localization, and tracking in battlefields, where the large-scale na- ture renders centralized control prohibitive. In addition, the finite batteries in sensor nodes demand energy-aware network control. In this thesis, we propose an energy- efficient topology management model inspired by biological inter-cellular signaling schemes. The model allows sensor nodes to cluster around imminent targets in a purely distributed and autonomous fashion. In particular, nodes in the target vicinity collaborate to form clusters based on their relative observation quality values. Sub- sequently, the clustered sensor nodes compete based on their energy levels until some of them gain active status while the rest remain idle, again according to a distributed algorithm based on biological processes. A final phase of the model has the active cluster members compete until one of them becomes the clusterhead. We examine the behavior of such a model in both finite-size and infinite-size networks. Specifically, we show that the proposed model is inherently stable and achieves superior energy efficiency against reference protocols for networks of finite size. Furthermore, we dis- cuss the behavior of the model in the asymptotic case when the number of nodes goes to infinity. In this setting, we study the average number of cluster members.

Electronic noses combine an array of cross-selective gas sensors with a pattern recognition engine to identify odors. Pattern recognition of multivariate gas sensor response is usually performed using existing statistical and chemometric techniques. An alternative solution involves developing novel algorithms inspired by information processing in the biological olfactory system. The objective of this dissertation is to develop a neuromorphic architecture for pattern recognition for a chemosensor array inspired by key signal processing mechanisms in the olfactory system. Our approach can be summarized as follows. First, a high-dimensional odor signal is generated from a chemical sensor array. Three approaches have been proposed to generate this combinatorial and high dimensional odor signal: temperature-modulation of a metal-oxide chemoresistor, a large population of optical microbead sensors, and infrared spectroscopy. The resulting high-dimensional odor signals are subject to dimensionality reduction using a self-organizing model of chemotopic convergence. This convergence transforms the initial combinatorial high-dimensional code into an organized spatial pattern (i.e., an odor image), which decouples odor identity from intensity. Two lateral inhibitory circuits subsequently process the highly overlapping odor images obtained after convergence. The first shunting lateral inhibition circuits perform gain control enabling identification of the odorant across a wide range of concentration. This shunting lateral inhibition is followed by an additive lateral inhibition circuit with center-surround connections. These circuits improve contrast between odor images leading to more sparse and orthogonal patterns than the one available at the input. The sharpened odor image is stored in a neurodynamic model of a cortex. Finally, anti-Hebbian/ Hebbian inhibitory feedback from the cortical circuits to the contrast enhancement circuits performs mixture segmentation and weaker odor/background suppression, respectively. We validate the models using experimental datasets and show our results are consistent with recent neurobiological findings.

This work details CMOS, bio-inspired, bio-compatible circuits which were used as synapses between an artificial neuron and a living neuron and between two living neurons. An intracellular signal from a living neuron was amplified, an integrate-and-fire neuron was used as a simple processing element to detect the spikes, and an artificial synapse was used to send outputs to another living neuron. The key structure is an electronic synapse which is based around a floating-gate pFET. The charge on the floating-gate is analogous to the synaptic weight and can be modified. This modification can be viewed as similar to long-term potentiation and long-term depression. The modification can either be programmed (supervised learning) or can adapt to the inputs (unsupervised learning). Since the technology to change the floating-gate weight has greatly improved, these weights can be set quickly and accurately. Intrinsic floating-gate learning rules were explored and the ability to change the synaptic weight was shown.

No description available.