Facultad de Ingeniería y Tecnologías

Our aim was to investigate whether hot water extract (CLW) of Curcuma longa L. could prevent non-alcoholic fatty liver disease (NAFLD). HepG2 cells were treated with free fatty acid (FFA) mixture (oleic acid: palmitic acid, 2:1) for 24 h to stimulate in vitro fatty liver. In addition, C57BL/6 mice were fed 60 kcal% high-fat (HF) diet for eight weeks to induce fatty liver in vivo. Intracellular reactive oxygen species (ROS) and malondialdehyde (MDA) productions were increased by FFA and HF-diet, but supplementation with CLW significantly decreased these levels. CLW treatment ameliorated antioxidant activities that were suppressed by exposure to the FFA and HF-diet. Cluster of differentiation 36 (CD36) and fatty acid transport proteins (FATP2 and FATP5) were increased in HF-diet groups, while CLW suppressed their expression levels. Moreover, sterol regulatory element-binding protein-1c (SREBP-1c), acetyl-coenzyme A carboxylase (ACC), and fatty acid synthase (FAS) expression levels were down-regulated in the CLW groups compared to HF-diet groups. On the other hand, 5 adenosine monophosphate-activated protein kinase (AMPK), Peroxisome proliferator-activated receptor alpha (PPAR-α), and carnitine palmitoyltransferase 1 (CPT-1) expressions were up-regulated in the CLW groups. HF-diet fed mice showed high hepatic triglycerides (TG) content compared to the normal diet mice. However, the administration of CLW restored the hepatic TG level, indicating an inhibitory effect against lipid accumulation by CLW. These results suggest that CLW could be a potentially useful agent for the prevention of NAFLD through modulating fatty acid uptake.

An electrical double-layer capacitor (EDLC) is based on the physical adsorption/desorption of electrolyte ions onto the surface of electrodes. Due to its high surface area and other properties, such as electrochemical stability and high electrical conductivity, carbon materials are the most widely used materials for EDLC electrodes. In this work, we study an activated carbon felt obtained from sheep wool felt (ACF'f) as a supercapacitor electrode. The ACF'f was characterized by elemental analysis, scanning electron microscopy (SEM), textural analysis, and X-ray photoelectron spectroscopy (XPS). The electrochemical behaviour of the ACF'f was tested in a two-electrode Swagelok ®-type, using acidic and basic aqueous electrolytes. At low current densities, the maximum specific capacitance determined from the charge-discharge curves were 163 F•g −1 and 152 F•g −1 , in acidic and basic electrolytes, respectively. The capacitance retention at higher current densities was better in acidic electrolyte while, for both electrolytes, the voltammogram of the sample presents a typical capacitive behaviour, being in accordance with the electrochemical results.

Carbon material is the largest material used as electrode on advanced energy storage devices. The modern lifestyle requires more energy, consequently, more smart energy use and efficient devices are needed. The constant evolution of materials technologies looking for green material and renewable raw material, that have minimal impact on the environment, is one of the most important subjects studied in recent years. The scientific and industry community are paying more attention to new forms of carbon such as nanotubes, graphene, and activated carbon fiber. The purpose of this work is to convert human hair into a hollow carbon filament to be used as a supercapacitor electrode. The human hair needs 3 stages to be converted into carbon filament: textile manufacture, oxidation, and carbonization. The electrochemical behavior was analyzed in a threeelectrode electrochemical cell system with 2 M of H 2 SO 4 electrolyte medium. The behavior of the electrode was characterized electrochemically by galvanostatic charge/discharge curves, cyclic voltammetry, and electrochemical impedance spectroscopy, showing 163 F g -1 of a maximum value of specific capacitance.

Natural fibres are abundant and renewable sources that constitute interesting alternative precursors to reduce the cost and environmental impact in the manufacturing of carbon fibres. However, they are prone to fuse and degrade upon thermal treatment, which strongly affects the morphology and stability of the prepared fibres. This study has analysed the chemical, structural and morphological transformations of wool fibres during the oxidative stabilization in air, to investigate their potential use as natural precursors for the production of carbon fibres. A thorough analysis of the fibres stabilized at mild temperatures has been carried out to follow the evolution of the O-, N-and Smoieties of the wool fibres during the oxidative stabilization. Below 400 ºC, the structure and morphology of the fibres was preserved, with important changes in the composition. Sulphur and oxygen contents decreased with the stabilization temperature due to decomposition and oxidation of S-and O-groups (e.g., cleavage of sulphur bridges, oxidative crosslinking). Nitrogen content remained rather constant regardless of the stabilization temperature and time, indicating a rearrangement of N-groups most likely through the transformation of amides in carbodiimide groups. At 300 ºC, dehydration reactions involving adjacent surface groups take place, causing the loss of the secondary structure of the fibres, while the polypeptide bonds remain unaffected. Data showed that the oxidative at 300 ºC can render materials with a fibrous morphology and high carbon yield, susceptible of undergoing further carbonization at higher temperatures to obtain carbon fibres.

In multi-phase buck converters the efficiency at light load can be substantially improved by deactivating some phases and reducing the switching losses associated. This paper analyzes the perturbations introduced by the phase-enable logic and proposes a correction term to mitigate the effect of the transitions on the output voltage. The strategy proposed is verified using simulations.

The estimation of the vertical in humans is important in everyday life although the mechanisms involved are not completely understood yet. This paper presents two sets of experiments with normal subjects, using the same virtual reality setup, aiming to help in this understanding. First, a steady state experiment is presented, which is used to determine the gravitational vertical precision while the second, a dynamical transient response experiment, is used to find dynamic models of each subject response. Results show that the dynamic models are able to reproduce the results of the steady state experiment while having the benefits that a dynamic model brings to evaluate subjects performance.

The Gyrover is a single wheel gyroscopically stabilized mobile robot developed at Carnegie Mellon University. An internal pendulum serves as a counter weight for a drive motor that causes fore/aft motion, while a tilt-mechanism on a large gyroscope provides a mechanism for lateral actuation. In this paper, we develop a detailed dynamic model for the Gyrover, and use this model in an extended Kalman filter to estimate the complete state. A linearized version of the model is used to develop a state feedback controller. The design methodology is based on a semidefinite programming procedure which optimize the stability region subject to a set of Linear Matrix Inequalities that capture stability and pole placement constraints. Finally, the controller design combined with the extended Kalman filter are verified on the prototype.

This paper presents a method for real-time switching among a pool of controllers to achieve desired performance. The algorithm uses a set of neural networks to evaluate the future performance of each controller to make the selection. Results are shown in simulation cases. Theoretical issues are discussed and future lines of research are stated.

This paper presents new results on switching control using neural networks. Given a set of candidate controllers, a pair of neural networks is trained to identify the stability region and estimate the closed-loop performance for each controller. The neural network outputs are used in the on-line switching rule to select the controller output to be applied to the system during each control period. The paper presents architectures and training procedures for the neural networks and sufcient conditions for stability of the closed-loop system using the proposed switching strategy. The neural-network-based switching strategy is applied to generate the switching strategy embeded in the simplex architecture, a real-time infrastructure for soft on-line control system upgrades. Results are shown for the real-time level control of a submerged vessel.

This paper presents new results on switching control using neural networks. Given a set of candidate controllers, a pair of neural networks is trained to identify the stability region and estimate the closed-loop performance for each controller. The neural network outputs are used in the on-line switching rule to select the controller output to be applied to the system during each control period. The paper presents architectures and training procedures for the neural networks and sufcient conditions for stability of the closed-loop system using the proposed switching strategy. The neural-network-based switching strategy is applied to generate the switching strategy embeded in the simplex architecture, a real-time infrastructure for soft on-line control system upgrades. Results are shown for the real-time level control of a submerged vessel.

The Gyrover is a single wheel gyroscopically stabilized mobile robot developed at Carnegie Mellon University. An internal pendulum serves as a counter weight for a drive motor that causes fore/aft motion, while a tilt-mechanism on a large gyroscope provides a mechanism for lateral actuation. In this paper, we develop a detailed dynamic model for the Gyrover, and use this model in an extended Kalman filter to estimate the complete state. A linearized version of the model is used to develop a state feedback controller. The design methodology is based on a semidefinite programming procedure which optimize the stability region subject to a set of Linear Matrix Inequalities that capture stability and pole placement constraints. Finally, the controller design combined with the extended Kalman filter are verified on the prototype.

A research on the application of Neural Networks in Control was done. The plant proposed was the Inverted Pendulum as an unstable system with non-linearities mainly due to friction force. A CMAC neural network, simulated in a PC AT compatible computer, was used in two different structures. In the first structure the CMAC learned to control the plant copying an existing linear controller (CRE). In the second one the CMAC learned the inverted transference of the plant to be controlled. Then control actions were generated based on a trajectory planner. Both structures gave very good results as described herein.

This paper presents a two-level control strategy for the trajectory tracking of the quadcopter. The external loop is designed to move the translational dynamics of the robot to follow smooth parametric functions using an appropriate variables substitution. The outer loop generates the desired trajectories of the orientation angles, which are controlled in the inner loop. The parameter gains are calculated using linear optimal control in order to minimize the energy consumption of the rotors. The control architecture is designed to be compatible with standard electronic platforms of commercial quadcopters and incorporates the purpose of maximizing flight time, an important requirement in real applications of multi-rotor aerial vehicles. Some numerical simulations show the performance of the control approach.

Memoria de grado presentada como requisito parcial para la obtención del grado de Ingeniero en Electrónica -Telecomunicaciones Montevideo, 21 de Junio de 2013 Resumen Universidad Católica del Uruguay. Facultad de Ingeniería y Tecnologías. [Ferreira Patricio -Zerpa Rodrigo] 2