Waterflooding is a common type of oil recovery techniques where water is pumped into the reservoir for increased productivity. Reservoir states change with time, as such, different injection and production settings will be required to lead the process to optimal operation which is actually a dynamic optimization problem. This could be solved through optimal control techniques which traditionally can only provide an open-loop solution. However, this solution is not appropriate for reservoir production due to numerous uncertain properties involved. Models that are updated through the current industrial practice of ‘history matching’ may fail to predict reality correctly and therefore, solutions based on history-matched models may be suboptimal or non-optimal at all. Due to its ability in counteracting the effects uncertainties, direct feedback control has been proposed recently for optimal waterflooding operations. In this work, two feedback approaches were developed for waterflooding process optimization. The first approach is based on the principle of receding horizon control (RHC) while the second is a new dynamic optimization method developed from the technique of self-optimizing control (SOC). For the SOC methodology, appropriate controlled variables (CVs) as combinations of measurement histories and manipulated variables are first derived through regression based on simulation data obtained from a nominal model. Then the optimal feedback control law was represented as a linear function of measurement histories from the CVs obtained. Based on simulation studies, the RHC approach was found to be very sensitive to uncertainties when the nominal model differed significantly from the conceived real reservoir. The SOC methodology on the other hand, was shown to achieve an operational profit with only 2% worse than the true optimal control, but 30% better than the open-loop optimal control under the same uncertainties. The simplicity of the developed SOC approach coupled with its robustness to handle uncertainties proved its potentials to real industrial applications.

Sensing rotation has been an essential topic in navigation and many other applications. Gyroscopes based on propagation of light beams over fixed distances have gained interest with the development of the laser. Since the 1970s, with the development of fiber optics these laser based gyroscopes have developed into compact devices, which can fit in the palm of your hand. In this thesis, we describe and analyze the development of a fiber-optic gyroscope. Fiber optic gyroscopes (also called fiber gyro or FOG) have been under development for different types of applications for more than 30 years all around the world. The physical basis of the fiber gyro is the Sagnac effect that was discovered in the early 1900s and is named after its discoverer. In this work, we first explain the principle of operation of the Sagnac effect and we derive the fundamental formulations in order to have an analytical understanding of the theory. Then, we examine the fiber optic gyro configuration component by component, starting with the laser diode pumped broadband light emitting Erbium-doped superfluorescent source. In addition, the principle of phase modulation, electro-optic phase modulators, fiber optic cables and fiber winding techniques, such as quadrupolar winding is explained within the context of development of the FOG. v The FOG that was assembled was based on circulation and sensing of broadband light centered around 1550nm. The fiber coil was 5km long in order to increase sensitivity in the FOG device. Since single-mode fibers were used steps were taken to ensure successful operation even with polarization dependent errors. The constructed system demonstrated a low sensitivity with a large uncertainty while sensing typical rotation rates. Reasons behind the errors and low sensitivity, as well as improvements that can be made are discussed.

本文採用數值模擬的方式，在價格具僵固性的無窮期動態Cournot模型中，探討水平式合併對廠商獲利能力及社會總福利的影響。可以發現不論市場價格調整速度的快慢，當廠商採取open-loop策略時，加入合併的廠商家數佔總產業比例必須達到某個水準，合併才會有利可圖。如果廠商使用feedback或closed-loop memoryless策略，則不論合併的廠商家數多寡，合併都能使廠商獲利。然而無論合併是否能為廠商帶來更多的利潤，只要市場上發生水平式合併，必然造成社會福利的損失。

Waterflooding is a common type of oil recovery techniques where water is pumped into the reservoir for increased productivity. Reservoir states change with time, as such, different injection and production settings will be required to lead the process to optimal operation which is actually a dynamic optimization problem. This could be solved through optimal control techniques which traditionally can only provide an open-loop solution. However, this solution is not appropriate for reservoir production due to numerous uncertain properties involved. Models that are updated through the current industrial practice of ‘history matching’ may fail to predict reality correctly and therefore, solutions based on history-matched models may be suboptimal or non-optimal at all. Due to its ability in counteracting the effects uncertainties, direct feedback control has been proposed recently for optimal waterflooding operations. In this work, two feedback approaches were developed for waterflooding process optimization. The first approach is based on the principle of receding horizon control (RHC) while the second is a new dynamic optimization method developed from the technique of self-optimizing control (SOC). For the SOC methodology, appropriate controlled variables (CVs) as combinations of measurement histories and manipulated variables are first derived through regression based on simulation data obtained from a nominal model. Then the optimal feedback control law was represented as a linear function of measurement histories from the CVs obtained. Based on simulation studies, the RHC approach was found to be very sensitive to uncertainties when the nominal model differed significantly from the conceived real reservoir. The SOC methodology on the other hand, was shown to achieve an operational profit with only 2% worse than the true optimal control, but 30% better than the open-loop optimal control under the same uncertainties. The simplicity of the developed SOC approach coupled with its robustness to handle uncertainties proved its potentials to real industrial applications.

No description available.

A robotically-controlled actuation system has been developed and built to perform active open-loop flow control experiments on transitional and turbulent backward-facing step flows in water. Control of the reattaching shear layer used hydraulic suction-and-blowing actuation emanating from 128 individual ports along the separation edge of the step. Each ports perturbation was periodic in time, but individually controlled to produce either spanwise-invariant (2D) or spanwise-varying (3D) spatial actuation profiles. An image processing system and special aqueous tuft were developed to measure the length of the recirculation bubble. Multiple images of a tuft array were time-averaged to do so. In general, 3D forcing was no more effective in reducing bubble length than 2D forcing. However, greater local spanwise reductions in reattachment length were observed for some cases of spanwise-varying forcing. Backlit dye was used to track the evolution of vorticity in the flow in video and still images.

Η αυτόματη πλοήγηση πλοίων αποτελεί ένα αντικείμενο ιδιαίτερης μελέτης κατά τα τελευταία χρόνια. Αυτό οφείλεται αφενός στην μεγάλη χρησιμότητα της εφαρμογής της, αφετέρου στην δυσκολία μαθηματικής περιγραφής και μοντελοποίησης των αντίστοιχων συστημάτων. Κατά την διπλωματική εργασία αυτή, σχεδιάστηκε ένας ελεγκτής ανοικτού βρόχου εμπροσθογενούς δράσης, με τη βοήθεια πειραματικών μετρήσεων, που σε συνδυασμό με τους συνήθεις πλέον PD και ασαφείς ελεγκτές, υλοποιήθηκε με σκοπό τον πειραματικό έλεγχο της θέσης και του προσανατολισμού του μοντέλου ενός πραγματικού πλοίου. / Control in ship navigation is a field of excess research during the last few years. The increasing need for automated ship navigation, combined with the challenging to describe and model systems of ships and other vessels, has led to many great works on this area. In this diploma thesis an open loop feedforward controller was designed using experimental data. This controller was used together with PD and Fuzzy controllers to control the position and the orientation of a model ship.

Square open loop resonator (SOLR) bandpass filters fabricated with additive manufacturing techniques are presented and studied. One filter contains novel 3D capacitive plates used to enhance resonator coupling. The filters are centered at 2.45 GHz and loaded with capacitors for miniaturization as low as 21% that of a conventional SOLR bandpass filter. The pass-band insertion loss of the filters ranges from 3.8 dB to 5.5 dB and the 3 dB bandwidth ranges from 180 MHz to 250 MHz. Also, degradation in the effective conductivity of printed ink as a function of substrate roughness is analyzed. Finally, a study of dielectric and metallic 3D printing processes that are candidates for digital manufacturing of integrated mobile phone client antennas is presented.

京都大学 / 新制・課程博士 / 博士(理学) / 甲第22973号 / 理博第4650号 / 新制||理||1668(附属図書館) / 京都大学大学院理学研究科数学・数理解析専攻 / (主査)教授 日野 正訓, 教授 泉 正己, 准教授 矢野 孝次 / 学位規則第4条第1項該当 / Doctor of Science / Kyoto University / DFAM

Le développement récent en micro- et nanotechnologies (dans des domaines tels que l’horlogerie,l’électronique, l’optique, le biomédical, . . .) a créé un fort besoin concernant des systèmes capables de manipuler et d’assembler des objets de plus en plus petits. La conception de stations robotisées, capables de manipuler des micro-objets, s’est multipliée à travers le monde, faisant intervenir des actionneurs de haute résolution adaptés au micro monde, ainsi que de nombreux capteurs.Ce mémoire ouvre une nouvelle voie pour le développement de robots de micromanipulation. Il présente la conception, la modélisation, la fabrication et la commande d’un nouveau concept de micro robot, le DiMiBot (Digital MicroroBot). Il s’agit du premier micro robot numérique – inspiré ´e de l’ ´électronique numérique – qui fait intervenir des actionneurs binaires pour générer un déplacement discret d’une grande précision sans nécessiter de capteur (en boucle ouverte). Ces actionneurs binaires extrêmement répétables et robustes (les modules bistables), assurent chacun un déplacement précis de 25 μm. Ils sont associés de manière monolithique à une architecture robotique parallèle flexible, assurant la génération d’un espace de travail discret, dont les 2N (N est le nombre de modules bistables utilisés au sein du DiMiBot) positions distincts atteignables sont parfaitement stables, répétables et robustes mécaniquement. Elles sont réparties de manière homogène dans un carré de 10,5 μm de côté La micro fabrication du premier prototype de micro robot numérique en silicium – faisant suite `a un dimensionnement minutieux en élément finis – a été réalisé au sein de la salle blanche MIMENTO de l’institut FEMTO-ST. Ce DiMiBot possédant 4 modules bistables assure une résolution de 3,5 μm pour une répétable de chacune des 16 positions atteignables de 90 nm. / With the current expansion of micro- and nano-technologies (in such domains as watchmaking, electronics,optics, biomedical, . . .), came the necessity to build systems able to manipulate and make the assembly ofsmaller and smaller objects. Design of robotic stations, able to manipulate micro-objects, expanded all over theworld, making use of high resolution actuators and numerous sensors adapted to the microworld.This thesis opens a new paradigm in the design of micromanipulation robotics. We present the design, modeling,fabrication and control of a new microrobot, the DiMiBot (Digital MicroroBot). It is the first digital microrobot— inspired from digital electronics — which makes use of binary actuators for the generation of discretedisplacements with high accuracy without any sensors (open-loop control). These highly repeatable and robustbinary actuators (bistable modules) generate an accurate displacement of 25 μm. They are monolithicallyconbined with a parallel flexible architecture, allowing the generation of a discrete workspace, in which all the2N (N is the number of bistable modules used) distinct reachable positions are perfectly stable, repeatable andmechanically robust. They are evenly spread inside a 10.5 μm length square.After dimensioning, the first digital microrobot prototype in silicon was microfabricated in MIMENTO clean-roomof FEMTO-ST institute. This DiMiBot has 4 bistable modules and generates a workspace of 3.5 μm resolutionwith 90 nm repeatability