Artículos de conferencia

Permanent URI for this collection

https://repositorio.uc.cl/handle/11534/23156

Browse

Now showing 1 - 20 of 34

A Comparison of Path Planning Algorithms for Omni-Directional Robots in Dynamic Environments
(IEEE, 2006) Haro, F.; Torres Torriti, Miguel Attilio
The scope of this paper is to analyze and compare three path planning methods for omni-directional robots, which are based on a) the bug algorithm, b) the potential fields algorithm, and c) the A* algorithm for minimum cost path with multiresolution grids. The approaches are compared in terms of computational costs and the resulting path lengths. Results obtained indicate that the bug algorithm is a suitable choice for this type of application as its computational cost is lower than that of the other methods. Furthermore, minor modifications of the standard bug algorithm, such as the tangent following modification, allow the path planner to handle well the situations encountered in typical multi-robot environments
A DAC reconstruction filter for narrowband long distance Communications
(IEEE, 2014) Aldunate Varela, Joaquín Pablo; Feres Ducci, Carlos Jose; Vlahoyiannatos, Spyros; Oberli Graf, Christian Robert; de Villers-Grandchamps Zaldivar, Jean-Paul Idesbald; Guarini Hermann, Marcelo Walter
There are applications that require long-range, low data rate and low power consumption communications that could be implemented with wireless sensor networks equipped with multiple antennas. We are implementing a transceiver for that purpose. The transceiver requires a specific reconstruction filter which isn't commercially available for our target technology, XFAB XH018. An integrated active DAC reconstruction filter was designed and implemented on the target technology using Synopsys design tools. This document present the design, implementation and testing of the reconstruction filter. The filter achieved the design requirements.
A Digital Memristor Emulator for FPGA-Based Artificial Neural Networks
(IEEE, 2016) Vourkas, Ioannis; Abusleme Hoffman, Ángel Christian; Ntinas, Vasileios; Sirakoulis, Georgios C.; Rubio, Antonio
FPGAs are reconfigurable electronic platforms, well-suited to implement complex artificial neural networks (ANNs). To this end, the compact hardware (HW) implementation of artificial synapses is an important step to obtain human brain-like functionalities at circuit-level. In this context, the memristor has been proposed as the electronic analogue of biological synapses, but the price of commercially available samples still remains high, hence motivating the development of HW emulators. In this work we present the first digital memristor emulator based upon a voltage-controlled threshold-type bipolar memristor model. We validate its functionality in low-cost yet powerful FPGA families. We test its suitability for complex memristive circuits and prove its synaptic properties in a small associative memory via a perceptron ANN.
A model of internal control may improve the response time of an automatic arterial pressure controller
(IEEE, 2000) Urzua Urzua, Jorge; Sauma Santis, Enzo Enrique; Cipriano, Aldo; Guarini Hermann, Marcelo Walter
A simplified model for the arterial pressure control system was implemented on a personal computer using Matlab Simulink(R). Model responses to variations of systemic vascular resistance were comparable to those predicted by physiology. Computer simulation suggested that including this model of the internal pressure control system within the design of an external controller would achieve better arterial pressure control and faster response than previous systems.
A Novel Three-Port NPC Converter for Grid-Tied Photovoltaic Systems with Integrated Battery Energy Storage
(IEEE, 2020) Neira Castillo, Sebastian Felipe; Lizana Gajardo, Alonso Javier; Pereda Torres, Javier Eduardo
The variable behaviour of the photovoltaic (PV) generation arises as the main drawback of this technology. Therefore, Battery Energy Storage (BES) units appear as a promising solution to overcome the inherent intermittent generation profile of PV systems. BES can be implemented separated from the PV generation units or integrated within a single power system. The latter option gives a better dynamic response, as the variable generation can be directly complemented with the BES, controlling the power flow between both elements with the grid. Thus, there is a need for power converters with capability of interface both PV and BES with the grid in a reliable and efficient way. This paper proposes a Three-Port NPC converter to connect a PV array and a BES unit with the ac grid using a single power processing stage. The converter is based in the H-NPC inverter and allows the connection of a low voltage battery to act as a power buffer ensuring a firm energy dispatch to the grid. Furthermore, simulation results show that the proposed topology maintains the good performance of the H-NPC in terms of efficiency, voltage harmonic distortion and reduced leakage current.
A proposal for mixture of experts with entropic regularization
(IEEE, 2017) Peralta, Billy; Saavedra Domínguez, Ariel Saavedra; Caro, Luis
In these days, there are a growing interest in pattern recognition for tasks as prediction of weather events, recommendation of the best route, intrusion detection or face detection. Each of these tasks can be modelled as classification problem, where a common alternative is to use an ensemble model of classification. A well-known example is given by Mixture-of-Experts model, which represents a probabilistic artificial neural network consisting of local experts classifiers weighted by a gate network, and whose combination creates an environment of competition among experts seeking to obtain patterns of the data source. We observe that this architecture assume that one gate influence only one data point, consequently the training can be misguided in real datasets where the data is better explained by multiple experts. In this work, we present a variant of regular Mixture-of-Experts model, which consists of maximizing of the entropy of gate network in addition to classification cost minimization. The results show the advantage of our approach in multiple datasets in terms of accuracy metric. As a future work, we plan to apply this idea to the Mixture-of-Experts with embedded feature selection.
A series active power filter scheme for current harmonic compensation
(IEEE, 2008) Figueroa, Diego; Moran, Luis; Ruminot, Pedro; Dixon, Juan
A series active power filter to eliminate current harmonic components and compensate reactive power is presented and analyzed. The proposed active compensation technique is based in a series active filter composed by two single-phase inverters sharing the same DC bus, and is suitable for current harmonics and reactive power compensation generated by static converters. The proposed approach allows for low cost, simplicity and part count reduction in comparison with previous compensation solutions (passive and active). The paper analyzes the proposed series active power filter in terms of principles of operation, power circuit topology, control design and implementation. Experimental results obtained in a laboratory DSP based controlled prototype confirm the viability of the proposed compensation technique.
Approaches to transmission planning: A transmission expansion game
(IEEE, 2012) Molina Castro, Juan David; Contreras, Javier; Rudnick Van de Wyngard, Hugh
Defining and making decisions about network investments has become a hard task in a competitive environment. This work defines a methodology to look for expansion alternatives in a transmission system. We propose a Transmission Expansion Game model that consists of four main elements: i) generating transmission expansion plan scenarios, ii) valuation of a project based on the design of a linear contract, bargaining solutions and hidden actions, iii) optimal value of a risky investor's portfolio made up of several projects, and iv) transmission cost allocation with wind energy assets. The results obtained show that the model developed is efficient to solve the combinatorial problem. A principal-agent model obtains the real costs of the bidders and creates incentives for disclosure of information. The private optimal portfolio and bargaining for allocation cost gives evidence for the central planner to adjust the project assignment process and to carry out the proposed expansion plan efficiently. To test the methodology we analyze the Chilean Central Interconnected System.
Auctioning adequacy in south america through long-term contracts and options: From classic pay-as-bid to multi-item dynamic auctions
(IEEE, 2009) Moreno Vieyra, Rodrigo Andrés; Bezerra, Bernardo; Barroso, Luis A.; Mocarquer, S.; Rudnick van de Wyngard, Hugh
The adequacy problem in electricity market is becoming a very important issue as there is neither theoretical proof nor practical evidence of correct delivering of sufficient and timely generation capacity when it is needed in a real (imperfect) environment. In contrast, classical market design seems to fail when facing high demand growth and/or large hydro share as seen in several Latin American countries such as Chile, Brazil, Colombia and Peru among others. Currently, various mechanisms have arisen across this region with the intention of stimulating energy procurement and new investment. These are mainly based on long-term contract and options obligations, which are allocated through auctions. Auction theory then becomes very important to ensure optimal allocation and efficient prices for both the new generation and the end user. However, difficulties arise when applying pure auction theory because basic hypotheses are not met by most electricity markets. The objective of this paper is to address and discuss the Latin American experience with auction design for long-term contracts focusing on practical design and theory. The different mechanisms and auctions for ensuring supply adequacy are listed along with theoretical justification as part of the potential solution for the adequacy problem that different economies have proposed.
Building integral equation methods with the open-source library BEM++
(IEEE, 2016) Van't Wout, Elwin; Betcke, Timo; Scroggs, Matthew
Surface Integral Equations are often used to model electromagnetic scattering phenomena. Large-scale problems can efficiently be solved with Boundary Element Methods (BEM) of which the Method of Moments (MoM) has found widespread use in the computational electromagnetics community. The framework of integral equations allows for the design of many different formulations for a wide range of scattering problems. Most of them are a clever combination of the basic electric and magnetic field integral operators. In this paper, the open-source library BEM++ will be used as a powerful tool to build different integral equation formulations and preconditioners.
CNC control of a laser cutting machine
(IEEE, 1994) Chiang Sánchez, Luciano Eduardo; Ramos Grez, Jorge
This article describes the design and construction of a low power laser cutting machine with three degrees of freedom for applications such as wood, textile fabrics and the like. An available 10 Watt laser gun is used to provide the cutting mechanism. The results show that such a machine is feasible at a reasonable cost for applications where speed and repeatability are more important than high precision and power.<>
Considerations of a high repitition Capillary Discharge operated in nitrogen as a water-window X-ray microscope source
(IEEE, 2009) Wyndham, Edmund; Favre Domínguez, Mario; Valdivia, M. P.; Valenzuela Ahumada, Julio César
The capillary discharge is a very bright radiation source. When operated as a ns discharge at a peak current of order 10 kA and at a high repetition rate of order 200 Hz a very bright source at 13.5 nm is obtained. We present a series of observations that will extend the use of the discharge as a source of He-like nitrogen emission at 2.89 nm as a possible source for a water window soft X-ray microscope. A successful source depends on intense axial electron beams generated by the transient hollow cathode mechanism. These e-beams greatly enhance the plasma X-ray emission above that of a quasi Maxwellian distribution. Crucial to the practical realization of such a source is extremely low inductance geometry, effective heat removal and a ceramic wall and electrode heat loading that avoids ablation and impurities in the plasma. Observations of the time resolved optical spectrum from both ends of the plasma together with filtered X-ray diodes and a Faraday cup permit the verification of model parameters and also verify whether the wall loading is not evaporating the surface. Four capillary lengths and internal diameters are explored. Furthermore the electrical circuit based on low cost IGBT's is presented and a drive configuration that minimizes transformer magnetization losses that in other configurations is deposited in the plasma causing wall evaporation and contamination due to damped current oscillations over a period well after the main plasma discharge.
Cost effective vision system for mobile robots
(IEEE, 2002) Martin Aguirre, Juan Francisco; Chiang Sánchez, Luciano Eduardo
A Webcam-based vision system for performing image processing in real time in a restricted environment is presented. A color based image segmentation algorithm is the fundamental part of the system. Algorithm efficiency was very important due to hardware limitations. This led to a simple and cost-effective vision system that is able to detect and identify several mobile robots through successive frames. The main achievements of the presented system can be described as robust object detection under relative constant illumination and sufficient refresh rate at image processing to evaluate position and direction of movement of target objects with a rough idea of speed.
Energy planning using MESSAGE: The effect of large energy blocks in the Chilean system
(IEEE, 2010) Watts Casimis, David; Martínez Aranza, Víctor Julio
The evolution of the Chilean energy matrix into a more sustainable system with high levels of energy sufficiency and security of supply in the long term brings with it important challenges in the way that investors and the regulator make long term decisions and commitments. This translates into a paradigm change in the way in which the system is traditionally managed and planned. This change not only affects the operation and the dispatch of the system but it also affects the costs associated with the operation, economic planning, and financing of it. Long term energy planning is now fundamental for studying changes in the future national energy supply. This article contributes in this way by developing an energy model of the main national system and evaluating the effect of large injections of electrical energy on the structure of the matrix of electricity generation. Utilizing the planning tool MESSAGE, the following are studied: the impact of hydroelectricity generation in the south of the country, nuclear energy generation, and the effect of large blocks of energy that are dispersed and at times intermittent, as renewable energies are, in particular wind power.
Exploring differences in how learners navigate in MOOCs based on self-regulated learning and learning styles: A process mining approach
(IEEE, 2016) Maldonado Mahauad, Jorge Javier; Palta, René; Vázquez, Jorge; Bermeo, Jorge L.; Pérez Sanagustín, Mar; Munoz Gama, Jorge
Study in a Massive Open and Online Courses (MOOCs) is challenging, since participants take the course without the support of a teacher. Taking a MOOC require the students to have the ability to self-regulate their learning. However, every person has its own learning style and the way each one interacts and self-regulate in a MOOC varies. In this work we present an exploratory study from a process-oriented perspective to study whether students with different learning styles and SRL profiles show differences in navigating through a MOOC. Specifically, we investigate using Process Mining Techniques to analyze log files recording the course behavior of 99 learners across an Open edX MOOC combined with data from self-reported surveys. Our findings show that learners with different SRL profiles follow similar navigation paths, but there are differences when differentiating students by their learning styles.
Flexible transmission planning considering growing uncertainties from Renewable energy integration
(IEEE, 2012) Ren, Hui; Fan, Xiaozhou; Watts Casimis, David; Lv, Xingchen
Power systems are set to undergo dramatic changes driven by several factors ranging from climate change to technological developments. It is expected that networks will become more flexible to deal with increasing uncertainties, in those coming from future generation technologies and their locations as well as those from system operation practices. These developments will require fundamental changes in the way power systems are planned. This paper introduces a method of flexible transmission network planning with the traditional corrective control action assumed in the course of network planning. A two-stage solution algorithm is proposed combing Genetic Algorithm and Monte Carlo simulation. Candidate planning schemes are decided by Genetic Algorithm. Monte Carlo simulation and sensitivity method are used to find the most vulnerable part of the network, and then decides the amended control devices needed to be installed to achieve the optimum objective function and satisfy operational constraints (for each tested planning scheme). Discrete constraints associated with voltage control devices are included to make the simulation more realistic. The most flexible planning scheme is then defined as the one which needs the least investment on extra control devices. The proposed approach is implemented on an 18- bus test system and its feasibility is demonstrated.
High power machine drive, based on three-stage connection of "H" converters, and active front end rectifiers
(IEEE, 2003) Dixon Rojas, Juan; Breton, Alberto A.; Rios, Felipe E.; Moran, Luis
A three-stage inverter using "H" converters is being analyzed for high power machine drive applications. The great advantage of this kind of converter is the minimum harmonic distortion obtained at the machine side. The drawbacks are the isolated power supplies required for each one of the three stages of the multiconverter. In this paper this problem has been overcome in two ways: 1) by using independent windings for each phase of the motor, or 2) by using independent input transformers. Special configurations and combinations of passive rectifiers and active front end rectifiers for one of the stages of the drive are used to eliminate all input harmonics. The topology can also keep unity power factor at the input terminals. Simulation results are shown and some experiments with small four-stage prototypes are displayed. The control of this multi-converter is being implemented using DSP controllers, which give flexibility to the system.
Hybrid modeling and predictive control for hydrometallurgical processes
(IEEE, 2014) Karelovic, P.; Cipriano, Aldo
Model-based control strategies rely heavily on precise models to make accurate predictions. In mineral processing, particularly, due to complexities such as strong nonlinearities, variable coupling, time varying parameters, etc.; the development of accurate process models becomes an increasingly difficult task. Additionally, the ability to simulate process behavior without having to disturb plant operation prevents the loss of man-hours and production.
Impacts of optimal allocation of DFACTS on the penetration capacity of distributed generation
(IEEE, 2010) Ren, Hui.; Yu, X.; Teng, Weijun; Watts Casimis, David
Distributed generation (DG) with sometimes undesirable or troublesome location and capacity has adverse impacts on the transmission losses of the distribution network, on voltage stability and network reliability. With the growing application of renewable and unconventional energy sources, the penetration capacity of DG is increasing, therefore affecting power flows in the distribution network. Distributed FACTS (DFACTS) devices, with their capability of power flow control, could be a possible solution to this issue. In this paper, Monte Carlo simulation is used to simulate the commission and operation of DG. A sensitivity method is used to select the install location of DFACTS. How DFACTS help to release restrictions on the location and maximum penetration capacity of DG units is analyzed. Simulation results show that with the application of DFACTS, limitations of the distributed power's position and capacity can be effectively reduced. Simulation results provide useful information on DFACTS application in distribution system with DGs.
Improving the initial response process in urban emergencies
(IEEE, 2012) Monares, A.; Ochoa, S.; Pino, J.; Herskovic, Valeria
Growing population and service interconnection make urban areas increasingly vulnerable to emergency situations. Although the firefighting response process plays a key role in most urban emergencies, it has remained essentially unchanged in the past decades; therefore it has several limitations to coordinate effective and efficient response activities. This paper studies the firefighting initial response process in order to measure, understand and improve it. The study involved a literature review, interviews with firefighting personnel and the analysis of 66 urban emergencies in the city of Santiago (Chile). Based on the obtained results this article proposes an emergency coordination system to improve the initial response process in urban emergencies. In order to realize the impact of the proposed system on the initial response process, a simulation based on past emergencies is presented and discussed.

Browse

Browsing Artículos de conferencia by browse.metadata.fuente "IEEE"

Results Per Page

Sort Options