J.J. López

Time-frequency transforms comparison for power quality analysis

Voltage signal contains transient disturbances that are caused for multiple reasons. One way to improve the power quality is to analyze efficiently and to deeply understand these signals. The most typical disturbances in power quality have been generated in order to analyze by time-frequency transforms: wavelet transforms (WT) and S-transform (ST). Thus can be obtained a comparison of results of these time-frequency transforms. The voltage signals with different disturbances are simulated by Matlab code with 128 samples per cycle. The disturbances used in this work are carried out from standards, as voltage sag, voltage swell, short interruptions and harmonic distortion, in addition to other disturbances of transitory order as spikes, notches and oscillatory transients.

Classification of power quality disturbances using Wavelet and Artificial Neural Networks

An automated classification system based on Wavelet transform as a feature extraction tool in combination with Artificial Neural Network as algorithm classifier is presented. Perturbed signals generated according to mathematical models have been used to obtain experimental results in two stages, first, with a data set with simple disturbances and, later, including complex disturbances, more usual in real electrical system. In both cases noise is added to the signals from 40dB to 20dB. Two different neural networks have been used as classifier algorithm, a backpropagation and probabilistic. A data set with several disturbances, simple and complex, has been generated by simulation software based on electrical models, to test the implemented system. Evaluation results verifying the accuracy of the proposed method are presented.

Electric customer classification using Nopfield recurrent ANN

In retail power markets precise information related to electric customers is of relevant interest. For efficient tariff design and pricing it is required accurate classification and segmentation of electric customers. In this paper, it is proposed a methodology for clustering electric customers based on a recurrent Hopfield Artificial Neural Network (H-ANN). In order to reduce the size of the input set of the clustering algorithm several filtering techniques are used. The effectiveness of the proposed approach is measured using characterization indexes. Results in a set of distribution customers are presented to demonstrate de efficiency of the approach.

Application of genetic algorithms to compute the magnetic field produced by overhead transmission lines

In this paper a new procedure for estimating longitudinal and cross-sectional profiles of electromagnetic fields produced by overhead transmission lines (OTL) is presented. The method is based on genetic algorithms, and uses a limited number of field measurements produced by overhead transmission lines. The procedure has been validated for different line configurations by comparison of the measured values with the theoretical ones calculated through the representation of the magnetic field vectors with double complex numbers.

Selling price for a retailer using a comprehensive methodology with risk management

Retail electricity markets are critical in assuring that final customers receive the full benefits of competition in wholesale electricity markets. Retailers act as a service interface between the energy market and customers. In this paper, a comprehensive methodology to solve the problem that retailers face in an electricity energy market, including risk management, is proposed. The methodology consist of two main stages, at the first stage, consumers are characterized and classified using clustering techniques, and at the second stage an optimization problem for each group of consumers is solved to get the optimal energy offer. The optimization problem considers uncertainty of price from the wholesale market through a scenario tree, and the Conditional Value at Risk (CVaR) as risk measure. The problem results into a quadratic optimization problem, and it is solved using the GAMS software.

Time-Frequency Transforms for Classification of Power Quality Disturbances

The usual operations on the distribution network such as switching loads and circuits, the proliferation of power electronic equipment and non-linear loads and the distributed generation with renewable energy are several of the most common causes that are leading to an increasing polluted power system in terms of voltage signal distortion. One way of improving the power quality (PQ) parameters consists of analyzing these disturbances efficiently and understanding them deeply (Dugan, 2000) and PQ monitoring is one major task in order to achieve it. PQ monitoring is not an easy task usually involving sophisticated hardware instrumentation and software packages. Many recent approaches in PQ monitoring try to achieve it through the automated classification of different disturbances. The different approaches in this field lead their efforts in two directions, the main parts that form an automated classification as depicted in Fig. 1. The first make focus to obtain a