TY - GEN
T1 - Real-Time Power Quality Disturbance Classification Using Convolutional Neural Networks
AU - Husodo, Budi Yanto
AU - Ramli, Kalamullah
AU - Ihsanto, Eko
AU - Gunawan, Teddy Surya
N1 - Funding Information:
The authors would like to express their gratitude to the Malaysian Ministry of Education (MOE), which has provided research funding through the Fundamental Research Grant, FRGS19-076-0684 (FRGS/1/2018/ICT02/UIAM/02/4). The authors would like to acknowledge as well, support from International Islamic University, University of New South Wales, Universitas Indonesia, and Universitas Mercu Buana.
Publisher Copyright:
© 2022, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.
PY - 2022
Y1 - 2022
N2 - There is a growing interest in disturbance monitoring to maintain power quality. This paper developed a real-time power quality disturbance (PQD) detection system using convolutional neural networks (CNN) due to its fast and accurate feature extraction and classification. First, 29 classes of power quality disturbance were synthetically generated around 5000 samples for each type. Second, an efficient CNN structure was developed to extract unique features. Next, the output of CNNs was then inputted into a fully connected layer with a softmax and classification layer to act as the classifier for 29 classes of PQD signals. Our proposed algorithm was then trained using 80% of the synthetic signals, while 20% of the synthetic signals were used for testing. Experimental results showed that the proposed algorithm produced a good result with the classification accuracy of 97.52% trained using 100 epochs. Furthermore, it requires only 80.96 μs to detect each 16 ms segment of PQD signals.
AB - There is a growing interest in disturbance monitoring to maintain power quality. This paper developed a real-time power quality disturbance (PQD) detection system using convolutional neural networks (CNN) due to its fast and accurate feature extraction and classification. First, 29 classes of power quality disturbance were synthetically generated around 5000 samples for each type. Second, an efficient CNN structure was developed to extract unique features. Next, the output of CNNs was then inputted into a fully connected layer with a softmax and classification layer to act as the classifier for 29 classes of PQD signals. Our proposed algorithm was then trained using 80% of the synthetic signals, while 20% of the synthetic signals were used for testing. Experimental results showed that the proposed algorithm produced a good result with the classification accuracy of 97.52% trained using 100 epochs. Furthermore, it requires only 80.96 μs to detect each 16 ms segment of PQD signals.
KW - Bidirectional long short-term memory
KW - Classification
KW - Power quality disturbance
KW - Recurrent neural network
KW - Time-frequency based feature extraction
UR - http://www.scopus.com/inward/record.url?scp=85112563674&partnerID=8YFLogxK
U2 - 10.1007/978-981-33-4597-3_64
DO - 10.1007/978-981-33-4597-3_64
M3 - Conference contribution
AN - SCOPUS:85112563674
SN - 9789813345966
T3 - Lecture Notes in Electrical Engineering
SP - 715
EP - 724
BT - Recent Trends in Mechatronics Towards Industry 4.0 - Selected Articles from iM3F 2020
A2 - Ab. Nasir, Ahmad Fakhri
A2 - Ibrahim, Ahmad Najmuddin
A2 - Ishak, Ismayuzri
A2 - Mat Yahya, Nafrizuan
A2 - Zakaria, Muhammad Aizzat
A2 - P. P. Abdul Majeed, Anwar
PB - Springer Science and Business Media Deutschland GmbH
T2 - Innovative Manufacturing, Mechatronics and Materials Forum, iM3F 2020
Y2 - 6 August 2020 through 6 August 2020
ER -