Power Quality Disturbance Classification Using Deep BiLSTM Architectures with Exponentially Decayed Number of Nodes in the Hidden Layers

Teddy Surya Gunawan, Budi Yanto Husodo, Eko Ihsanto, Kalamullah Ramli

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

In recent years, there is growing interest in automatic power quality disturbance (PQD) classification using deep learning algorithms. In this paper, the average of instantaneous frequency and the average of spectrum entropy were used as time-frequency based feature extraction due to its discriminatory nature. Bidirectional Long Short-Term Memory (BiLSTM) architectures with exponentially decayed number of nodes in deep multilayers were utilized as Deep Recurrent Neural Network (DRNN) classifier. We experimentally generated fifteen classes of synthetic PQD signals. Each class contains 1000 samples divided randomly into training, validation, and testing. Results showed that four hidden layers of BiLSTM with exponentially decayed nodes interleaved with dropout layers provided the best classification accuracy of 99.23%.

Original languageEnglish
Title of host publicationRecent Trends in Mechatronics Towards Industry 4.0 - Selected Articles from iM3F 2020
EditorsAhmad Fakhri Ab. Nasir, Ahmad Najmuddin Ibrahim, Ismayuzri Ishak, Nafrizuan Mat Yahya, Muhammad Aizzat Zakaria, Anwar P. P. Abdul Majeed
PublisherSpringer Science and Business Media Deutschland GmbH
Pages725-736
Number of pages12
ISBN (Print)9789813345966
DOIs
Publication statusPublished - 2022
EventInnovative Manufacturing, Mechatronics and Materials Forum, iM3F 2020 - Gambang, Malaysia
Duration: 6 Aug 20206 Aug 2020

Publication series

NameLecture Notes in Electrical Engineering
Volume730
ISSN (Print)1876-1100
ISSN (Electronic)1876-1119

Conference

ConferenceInnovative Manufacturing, Mechatronics and Materials Forum, iM3F 2020
Country/TerritoryMalaysia
CityGambang
Period6/08/206/08/20

Keywords

  • Bidirectional long short-term memory
  • Classification
  • Power quality disturbance
  • Recurrent neural network
  • Time-frequency based feature extraction

Fingerprint

Dive into the research topics of 'Power Quality Disturbance Classification Using Deep BiLSTM Architectures with Exponentially Decayed Number of Nodes in the Hidden Layers'. Together they form a unique fingerprint.

Cite this