.png){kind=icon}
返回
MOOST: A Multiobjective Optimization-Based S-Transform for Analyzing Power Quality Disturbances
.png){kind=icon}
PRE
AI
DOI：10.1109/TIM.2025.3633302.png)
.png){kind=icon}
原文求助.png){kind=icon}
分享.png){kind=icon}
收藏
摘要
En 中文
The increasing integration of renewable energy and nonlinear loads introduces complex and nonstationary power quality disturbances (PQDs), posing significant challenges to accurate monitoring and diagnosis in modern power systems. Effective time-frequency analysis is essential for identifying these disturbances. However, existing methods often rely on manual empirical parameter tuning, exhibit limited adaptability, and suffer from poor energy concentration, particularly for high-frequency components. To address these limitations, this study proposes the multiobjective optimization-based S-transform (MOOST), which leverages the nondominated sorting genetic algorithm II (NSGA-II) to adaptively tune the Gaussian window parameters. Unlike prior methods that optimize a single criterion, MOOST employs a dual-objective framework by minimizing the mean square error (mse) to preserve signal amplitude fidelity and minimizing the energy concentration measure (ECM) to enhance high-frequency component localization. This enables a principled and data-driven tradeoff between time and frequency resolutions. Extensive experiments across various PQD scenarios, including voltage sags, harmonics, flickers, and oscillatory transients, demonstrate that MOOST outperforms both the standard S-transform (ST) and improved ST (IST). The results validate MOOST as an adaptive, accurate, and noise-resilient tool for PQD analysis, with strong potential for real-time intelligent grid monitoring.
Keyword:
Time-frequency analysis
Accuracy
Signal resolution
Energy resolution
Power quality
Optimization
Frequency conversion
Transforms
Frequency estimation
Harmonic analysis
Multiobjective optimization
nondominated sorting genetic algorithm II (NSGA-II)
power quality disturbance (PQD)
S-transform (ST)
time-frequency analysis
期刊
IF:
5.9
论文数:
2.0W
被引数:
5.8W
机构
引用论文
.png){kind=icon}
Multi-objective optimization of reliability-redundancy allocation problem with cold-standby strategy using NSGA-II基于nsga-ii的冷备策略可靠性冗余分配问题的多目标优化
PREAI
分享收藏Adversarial attack and training for deep neural network based power quality disturbance classification基于深度神经网络的电能质量扰动分类对抗攻击与训练
PREAI
分享收藏
查看更多内容