Hybrid Continuous Wavelet Transform and GoogLeNet Framework for Accurate Classification of Power Quality Disturbances
Journal of Advanced Engineering and Computation · 27.06.2026 · 10 (2) · 121
DOI: 10.55579/jaec.2026102.539 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of Advanced Engineering and Computation, online erschienen am 27.06.2026. Verfasst von Duy Anh Bui und 2 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Accurate classification of power quality disturbances (PQDs) is critical for maintaining grid stability amid the increasing integration of renewable energy sources. However, traditional feature extraction methods and conventional Convolutional Neural Networks (CNNs) often struggle with non-stationary signals because fixed-size convolutional kernels cannot effectively capture features across multiple temporal and spectral scales simultaneously. To overcome this limitation, this paper proposes a hybrid framework that combines the Continuous Wavelet Transform (CWT) with the GoogLeNet (Inception v1) architecture. In the proposed approach, one-dimensional voltage waveforms are transformed into two-dimensional time–frequency scalograms using CWT. These scalograms are then processed by GoogLeNet’s Inception modules, which employ parallel convolutional pathways with kernel sizes of 1×1, 3×3, and 5×5 to extract multi-scale features concurrently. Extensive experiments were conducted on a balanced dataset consisting of 2,100 simulated samples representing seven different disturbance categories. The proposed framework achieved a mean classification accuracy of 90.95% ± 1.60% over 10 independent trials, with the highest recorded accuracy reaching 93.29%. In particular, frequency-domain disturbances, including Harmonics and Oscillatory Transients, were classified with perfect accuracy (100%) and zero standard deviation across all trials. The experimental results confirm that the proposed CWT–GoogLeNet framework effectively addresses the challenge of multi-scale feature extraction and provides reliable statistical performance for automated power quality monitoring in modern smart grid environments.
Zitieren
APA
Bui, D. A., Nguyen, B. N., & Kayal, P. (2026). Hybrid Continuous Wavelet Transform and GoogLeNet Framework for Accurate Classification of Power Quality Disturbances. Journal of Advanced Engineering and Computation, 10(2), 121. https://doi.org/10.55579/jaec.2026102.539
Vancouver
Bui DA, Nguyen BN, Kayal P. Hybrid Continuous Wavelet Transform and GoogLeNet Framework for Accurate Classification of Power Quality Disturbances. Journal of Advanced Engineering and Computation. 2026;10(2):121. doi:10.55579/jaec.2026102.539
BibTeX
@article{Bui2026,
author = {Bui, Duy Anh and Nguyen, Bon Nhan and Kayal, Partha},
title = {Hybrid Continuous Wavelet Transform and GoogLeNet Framework for Accurate Classification of Power Quality Disturbances},
journal = {Journal of Advanced Engineering and Computation},
year = {2026},
volume = {10},
number = {2},
pages = {121},
doi = {10.55579/jaec.2026102.539}
}
EUVIMED