Classification of Multiple and Single Power Quality Disturbances Using a Decision Tree-Based Approach
- PDF / 716,520 Bytes
- 11 Pages / 595.276 x 790.866 pts Page_size
- 92 Downloads / 238 Views
Classification of Multiple and Single Power Quality Disturbances Using a Decision Tree-Based Approach Bruno Henrique Groenner Barbosa · Danton Diego Ferreira
Received: 5 July 2012 / Revised: 22 May 2013 / Accepted: 9 June 2013 © Brazilian Society for Automatics–SBA 2013
Abstract This paper presents a new approach for power quality (PQ) disturbance classification. In order to classify both single and multiple disturbances, the principle of divideand-conquer is employed and a tree structure is constructed by means of simple classifiers: Perceptrons and a Bayesian classifier. Aiming at reducing its computational cost, only six parameters extracted from the filtered electrical signal are used by the final classifier. Such parameters are the secondorder cumulants and the RMS value. Results show that the proposed approach can classify many types of PQ disturbances with good accuracy even for different values of signalto-noise ratio and for real data. Keywords Power quality · Disturbance · Multiple · Decision tree · Classification
1 Introduction Power Quality (PQ) refers to a wide variety of electromagnetic phenomena that characterize voltage and current at a given time and given location on the power system (I. S. C. C. on Power Quality 1995). PQ has become an important area of research given the growing concern of providing distortion-free power supply to consumers. Proliferation of nonlinear loads, such as personal computers, AC drives [variable-frequency drives (VFD)], and distributed power generation causes changes in patterns of voltage and current, B. H. G. Barbosa · D. D. Ferreira (B) Engineering Department, Federal University of Lavras, Lavras, Minas Gerais, Brazil e-mail: [email protected] B. H. G. Barbosa e-mail: [email protected]
leading to PQ harmonic disturbances (Morsi and El-Hawary 2008). It should also take into account the new scenario of the electric power system (EPS) which is characterized by deregulation and high penetration of distributed output. In this new scenario, reaching a high degree of reliability will only be possible by integrating new technologies in the modern Smart Grid infrastructure (Ochoa et al. 2011; Zhang and Chow 2012). Thus, developing more efficient tools for monitoring PQ becomes extremely important. PQ disturbances such as voltage sags, swells, impulsive and oscillatory transients, notches, and harmonics cause failure or malfunction of electrical equipment, which can consequently reduce its useful life and cause considerable losses. In order to improve PQ, the sources and causes of the disturbances should be well known before taking appropriate measures. However, determining causes and sources of disturbances requires the ability to detect and classify them. Recently, some researchers have attempted to use appropriate signal processing and computational intelligence for detection and classification of PQ disturbances. The study (Bollen et al. 2009) presents a good review of the main tools of signal processing applied to PQ disturbance. Although there are currently a rea
Data Loading...
