Oscar Duarte

An arithmetic approach for the computing with words paradigm

Computing with words applications are mostly built using rule‐based systems, which have some important deficits: First, it is not easy to deal with high dimension problems because the size of the rule base increases exponentially; second, it is not possible to concatenate two or more systems without losing information; and third, there are no ways to compute inputs from outputs. In this article we show an alternative kind of system that remedies those deficits in many applications. It is based on fuzzy arithmetic rather than fuzzy logic. We also show application examples.

Strategic bidding in Colombian electricity market using a multi-agent learning approach

In this paper, a multi-agent model of an electricity market is proposed using the agent-based computational economics (ACE) methodology. The proposed methodology for modeling the bidding price behavior of Generation Companies (GENCOs) is based on a reinforcement learning algorithm (Q-learning) that uses some soft computing techniques to face the discovery of a complex function among bidding prices, states and profits. The proposed model also comprise the power system operation of a large-scale system by simulating optimal DC power flows (DCOPF) in order to obtain real dispatches of agents and a mapping from action space (bidding strategies) to quantities dispatched. In this model, agents are provided with learning capabilities so that they learn to bid depending on market prices and their risk perception so that profits are maximized. The proposed methodology is applied on colombian power market and some results about bidding strategies dynamics are shown. In addition, a new index defined as rate of market exploitation is introduced in order to characterize the agents bidding behavior.

Algorithms to extend crisp functions and their inverse functions to fuzzy numbers

In this article we present an algorithm to extend continuous crisp functions to fuzzy numbers using the extension principle; the functions must be monotonically increasing in some of the arguments and monotonically decreasing in the others. Then, we present two different solutions to the problem of extending inverse functions that we have called possible extension and necessary extension. Finally, using these solutions we have generated a family of intermediate extensions that let us define a parameter to measure the existence of the extended inverse functions.

Agent learning methodology for generators in an electricity market

In this paper, a model of the Colombian electricity market is implemented using the agent-based computational economics (ACE) methodology. The paper propose a methodology to model the offer price behavior of generation companies upon the actual colombian market structure and the effects in market prices and agentspsila profits. This model is based on a learning algorithm that uses some soft computing techniques to face the discovery of a complex function among offer prices, power system variables and profits. In addition, this methodology allows the agents to improve their offer strategies by maximizing their own profits. Finally, the paper presents some results obtained from the model about the behavior of spot prices and agents profits.

Modeling of bidding prices using soft computing techniques

In this paper an strategy for discovering patterns over the continuous domain of bidding prices is proposed. In particular, the proposed method represents bidding functions as points in a multidimensional space where a clustering algorithm is applied. On the other hand, as a result of this method, a dramatic reduction over the search space of bidding strategies is achieved. In addition, some relations of dominance over bidding strategies are found, improving the pattern recognition process of agentpsilas bidding behavior. This method is applied on the bidding prices database for some GENCOpsilas of the colombian power market. Furthermore, an application of some data mining algorithms is presented with the purpose of quantifying some hypothesis formulated on the effect of hydrology over spot and bidding prices.

Insulation temperature increase in power transformers due to stationary disturbances: Comparison of methods

This paper presents the comparison of two methods to estimate the insulation temperature increase and lifetime reduction in power transformers due to the presence of stationary power quality disturbances. The internal insulation can suffer an accelerated aging by temporary increases of rms values of currents caused by waveform distortion (harmonic current). The recurrent presence of increase in rms values during long time periods will produce a cumulative effect that degrade the dielectric material. Models for insulation aging are employed. A discussion about the effects on insulation lifetime is presented.

Grounding system modeling in EMTP/ATP based on its frequency response

The aim of this paper is to present a grounding system modeling using a circuit network synthesized from its frequency response. This circuital representation allows simulating the grounding system in EMTP/ATP program considering its frequency dependent effects. Also, any electrical component may be included in the simulation, such as power system equipments or electronic devices. The frequency response was obtained by means of both an electromagnetic model based on the moments method (MM) and a hybrid electromagnetic model (HEM). The results were compared with previous works and field measurements, finding a good agreement.

Semi-automated academic tutor for the selection of learning paths in a curriculum: An ontology-based approach

In higher education globalization brings challenges as part of a reform process. An important aspect of such a reform has to do with the flexibility of the contents of a curriculum. Normally curricula offer students a whole set of possibilities to make decisions regarding optional courses recommended in an educational programme. In this work we present the design of a semi-automated Academic Tutor to support students in selecting learning paths (that consist of a set of courses which form the individual curricula) to achieve a particular professional profile. We designed the Academic Tutor considering the case of an Electrical Engineering Curriculum, which is represented through ontologies, a Semantic Web technology that provides a logical and formal description that can be interpreted by people and machines. The curriculum was divided into three interrelated main structures called sources of conceptualization. The sources that were transformed into ontologies were: subjects from the programme curriculums core and subjects from the variable part of the programme curriculum; skills and thematic contents. Preliminary findings present a broad and formal representation of the curriculums knowledge to be shared and reused in the field of education and engineering.

A model to compute the frequency performance of reinforced concrete structures stroked by lightning

The steel in the reinforcement concrete structures is used frequently as the natural down conductor in Lightning Protection Systems (LPS). The lightning current distribution and the associated magnetic field are interesting topics for the LPS design and devices protection. The aim of this paper is to show the model developed and the software implementation which computes the current distribution considering the concrete surrounding the steel reinforcement. The model used was the Hybrid Electromagnetic Model - HEM. The grounding system and the soil resistivity were considered into the HEM and an approach did for the HEM expressions in order to compute the coupling between to media are shown. Finally a comparison with experimental results was done and some results are presented in this paper.

Demand side management through LAB+i platform: Case study

The introduction of Smart Grids elements into electric power distribution and market systems poses new challenges and opportunities to meet demand of end users. To properly incorporate these technologies into the electricity sector in Colombia, actors need to increase research capabilities, develop and implement technologies, test them through pilot projects under local conditions and adjust the current regulatory framework. The formulation of a strategic program oriented towards Demand Side Management is thus necessary. The National Laboratory for Smart Grids, LAB+i, currently has the necessary infrastructure to develop case studies in different functionalities of Smart Grids, including Demand Side Management. This article presents a case study with real values, applied on Bogota Campus of Universidad Nacional de Colombia. An automated management system was implemented in LAB+i based on three techniques of demand response. It allows finding the optimal load curve for each building.