Esteban Velilla

An efficient procedure for computing lightning induced overvoltages on overhead lines

Overvoltages on electrical power or telecommunications overhead lines due to nearby lightning strokes can cause faults and disturbances in electrical and electronic devices and apparatuses connected to them. It is important to understand these induced overvoltages, because equipments with electronic devices are sensitive to such disturbances. In this paper, lightning radiated electromagnetic fields caused by the return stroke are evaluated with transmission line engineering models, representing the current through the lightning channel. Integral equations are solved by means the Trapezoidal Rule, resulting in an efficient way to compute the radiated electromagnetic field. Also, the coupling between the electromagnetic field radiated by the return stroke and the overhead line is considered by using The Method of Characteristics. In this way, transmission line equations to describe voltages and currents in the overhead line are solved in order to compute the induced overvoltages. Results are compared with others internationally published.

Using genetic algorithm and the simplex method to obtain equivalent circuits of the grounding systems

There are many methodologies that, using the frequency response of a grounding system impedance, permit to get the coefficients of a transfer function or the electrical parameters of a specific circuit, that behave as equivalents of that system. In this paper in the first part, the coefficients of the transfer function and the parameters of the circuit were obtained using the simplex method to optimize the values and minimize the square error between the objective function and the frequency response of the software Gtierras. The results of the methodologies depend on the selection of the initial values (which change for every case evaluated) to start the process. This happens because the simplex method can get local values. In the second part it is proposed a genetic algorithm (global values) to get the initial values of the simplex method and solve the problem. The result of mixing both methods is compared with the vector fitting method, showing that it is possible to obtain a circuit or a transfer function that reproduce successfully the impedance of the grounding systems in the frequency domain.

Ubicación de Generación Distribuida para Minimización de Pérdidas Usando un Algoritmo Genético Híbrido

This paper presents a methodology for optimal Distributed Generation (DG) allocation using a hybrid genetic algorithm. In this case, the objective function is the minimization of power losses. The proposed algorithm involves an artificial neural network to evaluate the fitness function and a local search subroutine that allows the algorithm to explore a wider search space, and eventually, escape from local optimal solutions. The proposed approach was tested on a 34-bus system showing that the proper allocation of DG units allows the reduction of active power losses and the improvement of voltage profile.

Flujo Óptimo Reactivo para Minimización de Pérdidas: una Nueva Propuesta de Función de Adaptación y Manejo de Restricciones

This work presents a fitness function and constraint handling approach applied to the problem of optimal reactive power dispatch. This problem aims at the reduction of power losses by deciding on voltage setpoints of generators, transformers taps and switchable capacitor and reactor banks. The fitness function consist of a product of sub-functions that represent permissible limits on system variables and that also includes a specific goal on power loss reduction. The main advantage of the proposed fitness function is the fact that the global optimum of the problem is known beforehand. A genetic algorithm was used to solve the ORPD with both fitness functions. This fitness function was implemented and compared with a traditional fitness function by means of a genetic algorithm for three study IEEE cases: 57, 118 and 300 bus systems. It was found that the proposed fitness function showed better performance than the traditional one, guaranteeing the system operation within the feasible region.

Transferencia de Impulsos y Ganancia de Tensión en Transformadores Monofásicos de Baja Potencia

The voltage impulse transferences of 1.2/50µs and 3/23µs from the primary to the secondary winding was tested in three single phase transformers used in electronic applications; two transformers of 85W and 18W, and one autotransformer of 50W were used under no load and under resistive load conditions. The autotransformer showed higher damping and lower peak value in the transferred voltage impulse than the other transformers. The maximum voltage value in the secondary winding was obtained at no load condition and the frequency response of the voltage gain of the transformers between 10Hz and 1MHz were analyzed and fitted to transfer functions to model the voltage impulse transferences. The comparison of experimental results with those obtained from transfer functions in the frequency and time domain provides valuable information about the transferred voltage impulses in single phase transformers.

Modelo de Vida Útil para Estructuras Acero-Zinc Utilizadas en la Transmisión de Energía Eléctrica en Colombia

An atmospheric corrosion degradation model for steel and zinc used in the transmission system of electric power in Colombia is developed. Three important aspects are identified: the definition of corrosion rate, the probability distribution function of useful life and the inspection/maintenance plan that favours a longer useful life. The model identifies mass loss and thickness loss functions, defining limits for degradation and fault modes which are simulated using Markov chains to evaluate the structure availability and the optimal inspection and maintenance frequency. The results allow establishing zones with lower service life and their impact on the regulatory life. It is concluded that knowledge of degradation mechanisms allows a better definition of inspection and maintenance frequencies under a regulatory life.

Análisis de Sensibilidad Computacional de la Respuesta en Frecuencia de la Impedancia de Torres de Transmisión

espanolSe determina a traves del Modelo Hibrido Electromagnetico la respuesta en frecuencia de la impedancia vista desde el conductor de guarda. El modelo considera variaciones en la geometria de dos torres; la puesta a tierra y la resistividad del suelo. La evaluacion de transitorios electromagneticos en lineas de transmision involucra diferentes elementos siendo las torres uno de los que mas podria influenciar la magnitud y forma de onda de los transitorios generados. La respuesta en frecuencia presenta un comportamiento predominantemente inductivo en el rango de frecuencia analizado, razon por la cual fue ajustada a la expresion de impedancia de un circuito serie resistivo-inductivo-capacitivo en terminos de transformadas de Laplace, obteniendose valores de inductancia en el rango de lo reportado en la literatura tecnica. EnglishThe frequency response of the impedance seen from the top of two towers by the Hybrid Electromagnetic Model is determined. The model considers variations in the geometry of the towers, geometry of grounding system, and soil resistivity. The evaluation of electromagnetic transients in transmission lines involves different components being towers the elements that can further influence the magnitude and wave shape of the generated transient. The frequency response shows a predominantly inductive behavior. For that reason, frequency response is fitted to impedance expressions of a series-inductive-capacitive circuit in terms of Laplace transform, obtaining inductance values in ranges similar to values reported in the literature.

Predicción del Precio de la Electricidad en la Bolsa mediante un Modelo Neuronal No-Lineal Autorregresivo con Entradas Exógenas

The forecasting of the monthly average stock price of Colombian electricity was carried out by means of a nonlinear autoregressive exogenous model. Such a model considers as exogenous inputs the demand, the ratio between hydraulic and thermal generation, the probability of El Nino phenomenon and the daily available energy volume; and as an autoregressive variable the monthly average stock price. For training, validation and testing, data corresponding to the period January 2003 to March 2014 were considered. Furthermore, the effectiveness of the model was tested with data for the period April 2014 to February 2015.This shows that the incorporation of historic data of prices and variables that depend on hydrologic conditions, such as El Nino phenomenon, allows reproducing the dynamics of Colombia’s energy prices, even for forthcoming periods.

Opportunities for seasonal forward contracts in the Colombian electricity market

A characteristic feature of electricity prices in countries that highly depend on their hydrology resources is their great volatility. Therefore, it is expected that market participants enroll in forward and bilateral electricity contracts to reduce their risk and hedge from price volatility. Bearing in mind the seasonality of water resources in Colombia it is sensible to think of an energy contracting mechanism that would take advantage of it. Such mechanism is the so called seasonal forward contract that would provide lower energy prices during the rainy season so that customers (such as energy intensive industries) are encouraged to increase their demand, leading to a scenario that would benefit both, generators and consumers. This paper presents a review on electricity prices and their relationship with hydrology as well as a discussion over the opportunities on seasonal forward contracting in the Colombian electricity market and its potential advantages.

ESTIMACIÓN DE LA CARGA DE TRANSFORMADORES DE POTENCIA UTILIZANDO UNA RED NEURONAL ARTIFICIAL

Resumen En este articulo se presenta una metodologia para estimar la curva de carga de transformadores de potencia utilizando redes neuronales artificiales. Para implementar la metodologia propuesta se utilizaron datos reales de dos transformadores ubicados en diferentes zonas geograficas y con diferentes condiciones de operacion. La tecnica de estimacion de carga fue implementada con datos historicos de la empresa Interconexion Electrica S.A (ISA).Para realizar la prediccion de las curvas de carga se entreno una red neuronal usando MATLAB, la cual se adapta a la curva de carga con tiempos de prediccion diaria y semanal. La metodologia propuesta permite la estimacion de la curva de carga de transformadores de potencia con un error relativo de alrededor del 10%. Este metodo puede ser utilizado en otros equipos con caracteristicas de operacion similares. Palabras clave: curva de carga, transformadores de potencia, redes neuronales artificiales Load Estimation of Power Transformers using an Artificial Neural Network