TESIS de Postgrado de acceso ABIERTO

El problema de asignación de salidas de trenes (DMP por sus siglas en inglés)es un problema de planificación de trenes, donde el objetivo es encontrar la mejormanera de asignar trenes a salidas programadas, sujetas a un conjunto de restriccionesrelacionadas con la gestión ferroviaria. Este es un problema que surge delrolling stock units management on railway sites problem presentado en el DesafíoROADEF/EURO 2014, el cual se suele tratar como dos sub-problemas: El DMP yun problema de planificación de rutas. En este trabajo se presenta formalmente elDMP, indicando su modelo, los estudios realizados a la fecha sobre este problema yotros problemas de gestión de trenes relacionados, para luego presentar la propuestade un algoritmo tipo GRASP (G-DMP) y un Algoritmo Evolutivo (AE-DMP)para resolver este problema y luego evaluar su desempeño por medio de diferentesexperimentos.

The Adaptive Optics technology, implemented in dierent optical applications, is anelementary component in modern ground-based astronomical systems. It provides correctionof aberrations induced by atmospheric turbulence on the beam of light from a celestial body.Thus, it is possible to improve the resolution of scienti c images.The object of study of this Thesis is the MagAO system (6.5m Magellan Telescope AdaptiveOptics), implemented on the Clay Telescope, located at Las Campanas Observatory.Unfortunately, the data obtained by adaptive optics systems, such as MagAO, is altered notonly by atmospheric turbulence, but also by inherent vibrations of the system and additionalvibrations caused by instrumentation elements.Currently, the inuence of vibrations in the images obtained by MagAO is small and isusually neglected. However, in future astronomical projects, it is essential to identify thesevibrations, its nature, magnitude, and mitigation.In this Thesis, power spectral densities are obtained from data captured during astronomicalobservation sessions, to learn how the MagAO system is aected by vibrations. Thisanalysis allows to identify and measure the inuence of the system instruments operatingin MagAO.Based on the spectral analysis of captured data, mathematical models are obtained, todescribe the system and its vibrations. For this process, a standard AO loop is used, takinginto account perturbations induced by turbulence and vibrations, to obtain a linear modelfor MagAO.Finally, the model obtained allows us to propose improvements for the AO loop, throughthe design of control alternatives. The system performance for these controllers is analysedthrough simulations.

Se realizaron mediciones de isotermas de solubilidad de 2-(3,4-Dimethoxyphenyl)-5,6,7,8-tetramethoxychromen-4-one (nobiletina) y 2-methyl-1,4-naphthoquinone (menadiona) en dióxido de carbono supercrítico a temperaturas de (313, 323 y 333) K y presiones de (9 a 32) MPa usando una metodología analítica con recirculación, con determinación directa de la composición molar de la fase rica en dióxido de carbono usando cromatografía líquida de alto rendimiento. Los resultados indican que el rango de la solubilidad de nobiletina medida fue de 107 · 10-6 mol · mol-1 a 333 K y 18.35 MPa a 182 · 10-6 mol · mol-1 a 333 K y 31.40 MPa. Similarmente, la solubilidad de menadiona medida varió de 630 · 10-6 mol · mol-1 a 333 K y 9.72 MPa a 1057 · 10-6 mol · mol-1 a 333 K y 13.43 MPa. La validación de los datos de solubilidad experimentales se realizó usando tres enfoques, a saber, estimación de la incertidumbre combinada expandida para cada punto de solubilidad a partir de valores de parámetros experimentales (=23 · 10-6 mol · mol-1 y =111 · 10-6 mol · mol-1 para nobiletina y menadiona, respectivamente); consistencia termodinámica, verificada utilizando un test adaptado de herramientas basadas en la ecuación de Gibbs-Duhem y datos de solubilidad obtenidos de modelación; y auto-consistencia, demostrada mediante la correlación de los datos de solubilidad con un modelo semi-empírico como una función de la temperatura, presión y densidad de CO2 puro.

This work presents and experimentally validates a new control scheme for electricaldrive systems, named cascaded predictive speed and current control (PSCC). This newstrategy uses the model predictive control concept. It has a cascaded structure like thatfound in field-oriented control or direct torque control. Therefore the control strategy hastwo loops, external and internal, both implemented with model predictive control. Theexternal loop controls the speed, while the inner loop controls the stator currents. Theinner control loop is based on Finite Control Set Model Predictive Control (FS-MPC),and the external loop uses MPC deadbeat, making full use of the inner loop's highlydynamic response. Experimental results show that the proposed strategy is overshootfreeand has a high dynamic response.On the other hand, FS-MPC has many advantages, such as fast dynamic response andintuitive implementation. For these reasons it has been thoroughly researched in thelast decade. However, the spread switching frequency spectrum of FS-MPC remains amajor disadvantage of the strategy. The second part of this research discusses a modulatedmodel predictive control that guarantees a fixed spectrum switching frequency inthe linear modulation range and extend its optimized response to the over-modulationregion. Due to the equivalent high gain of the predictive control, and due to the limiton the voltage actuation of the power converter, it is expected for the actuation voltageto enter into the over-modulation region during large reference change or in response toload impacts. For this situation an optimized over-modulation strategy, that convergesto FS-MPC response for large tracking errors, is proposed. This technique seamlesslycombines the good steady-state switching performance of PWM with the high dynamicresponse of FS-MPC during large transients. Experimental results that show thegood steady-state switching performance, the FS-MPC like transient response and theseamless transition between modes of operation are presented for a permanent magnetsynchronous machine drive to test the proposed modulated predictive control strategy.