Article: A fast reconstruction of power harmonic signals based on empirical wavelet transform Journal: International Journal of Systems, Control and Communications (IJSCC) 2026 Vol.17 No.2 pp.176 - 196 Abstract: In order to overcome the problems of poor performance, low signal-to-noise ratio, and long reconstruction time of traditional methods, a fast reconstruction method of power harmonic signals based on empirical wavelet transform is proposed. Using sensors, data acquisition cards, and other devices to collect power harmonic signals, the collected signals are input into an adaptive feature enhancement neural network based on self attention to obtain denoised signals. The empirical wavelet transform is used to reconstruct the denoised signals, and its multi-scale analysis ability helps to preserve useful information in the signals while removing noise interference. The experimental results show that the proposed method has good reconstruction effect on power harmonic signals, with an average signal-to-noise ratio of 57.45 dB and an average reconstruction time of 80.21 ms. It has relatively high application value and makes important contributions to improving the accuracy of power quality monitoring and optimising harmonic control strategies. Inderscience Publishers - linking academia, business and industry through research

Title: A fast reconstruction of power harmonic signals based on empirical wavelet transform

Authors: Zufang Yang; Lian Xue

Addresses: College of Artificial Intelligence, Wuhan Technology and Business University, Wuhan, 430056, China; Institute of Information and Intelligent Engineering Applications, Wuhan Technology and Business University, Wuhan, 430056, China ' College of Artificial Intelligence, Wuhan Technology and Business University, Wuhan, 430056, China; Institute of Information and Intelligent Engineering Applications, Wuhan Technology and Business University, Wuhan, 430056, China

Abstract: In order to overcome the problems of poor performance, low signal-to-noise ratio, and long reconstruction time of traditional methods, a fast reconstruction method of power harmonic signals based on empirical wavelet transform is proposed. Using sensors, data acquisition cards, and other devices to collect power harmonic signals, the collected signals are input into an adaptive feature enhancement neural network based on self attention to obtain denoised signals. The empirical wavelet transform is used to reconstruct the denoised signals, and its multi-scale analysis ability helps to preserve useful information in the signals while removing noise interference. The experimental results show that the proposed method has good reconstruction effect on power harmonic signals, with an average signal-to-noise ratio of 57.45 dB and an average reconstruction time of 80.21 ms. It has relatively high application value and makes important contributions to improving the accuracy of power quality monitoring and optimising harmonic control strategies.

Keywords: empirical wavelet transform; power harmonic signal; fast reconstruction; sensors; data acquisition card.

DOI: 10.1504/IJSCC.2026.152972

International Journal of Systems, Control and Communications, 2026 Vol.17 No.2, pp.176 - 196

Received: 09 Apr 2025
Accepted: 06 Jun 2025
Published online: 17 Apr 2026 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article

Title: A fast reconstruction of power harmonic signals based on empirical wavelet transform

Keep up-to-date