Bundle Extreme Learning Machine for Power Quality Analysis in Transmission Networks

Abstract

This paper presents a novel method for online power quality data analysis in transmission networks using a machine learning-based classifier. The proposed classifier has a bundle structure based on the enhanced version of the Extreme Learning Machine (ELM). Due to its fast response and easy-to-build architecture, the ELM is an appropriate machine learning model for power quality analysis. The sparse Bayesian ELM and weighted ELM have been embedded into the proposed bundle learning machine. The case study includes real field signals obtained from the Turkish electricity transmission system. Most actual events like voltage sag, voltage swell, interruption, and harmonics have been detected using the proposed algorithm. For validation purposes, the ELM algorithm is compared with state-of-the-art methods such as artificial neural network and least squares support vector machine.

Determining the Optimal Location and Number of Voltage Dip Monitoring Devices Using the Binary Bat Algorithm

“…PQ measuring and monitoring devices are used to measure and monitor power quality and they are the first step in detecting and eliminating power quality problems. The issue of power quality and PQ measuring and identification is the subject of numerous scientific studies such as [1][2][3]. The ideal case would be if PQ measuring and monitoring devices were installed on every bus in the observed system.…”

Power quality disturbances classification based on Gramian angular summation field method and convolutional neural networks

“…The work described in Reference 21 is based on processing Phasor Measurement Unit (PMU) monitoring time series in three different ways for generating picture datasets for CNN model training and testing, which would include time‐domain stacking, frequency domain stacking, and GAF stacking, in order to facilitate a powerful, image‐based CNN for pattern detection and extraction. In Reference 22, a novel approach based on optimized Bayesian CNN is presented to classify the real power quality events such as sag, swell, interruption, and harmonics which is gathered from the nation‐wide power quality monitoring system. The production of 2D grey‐scale pictures integrating temporal and spectral representations of signals is implemented in Reference 23 to detect and classify power quality disturbances.…”

“…Hence in this paper, a unique strategy is introduced that combines Gramian Angular Summation Field (GASF) with CNN to take advantage of CNN's better performance in the field of image classification. Bayesian optimization 22 is implemented to optimize CNN framework for better performance. Firstly, a 1D raw data of PQ disturbances is converted into an image data using the GASF 25 .…”