Academia.eduAcademia.edu

Harmonic detection

description10 papers
group0 followers
lightbulbAbout this topic
Harmonic detection is the process of identifying and analyzing harmonic components within a signal, typically in the context of electrical engineering and signal processing. It involves the extraction of frequency components that are integer multiples of a fundamental frequency, which is crucial for assessing signal quality and system performance.
lightbulbAbout this topic
Harmonic detection is the process of identifying and analyzing harmonic components within a signal, typically in the context of electrical engineering and signal processing. It involves the extraction of frequency components that are integer multiples of a fundamental frequency, which is crucial for assessing signal quality and system performance.

Key research themes

1. How can advanced signal processing techniques improve harmonic detection accuracy and robustness under complex real-world conditions?

This theme focuses on enhancing harmonic detection methods using optimized signal processing algorithms and improved mathematical models to address issues like noise, frequency fluctuations, interfering harmonics, and non-ideal measurement conditions. Accurate harmonic detection is critical for monitoring power quality, communication signals, and spectroscopy, and improvements here enable better system stability and diagnostics.

Key finding: Proposes a minimum side-lobe optimization window function designed by refining cosine windows to effectively suppress spectrum leakage and fence effects under power frequency fluctuations. The new window function combined... Read more
Key finding: Evaluates nine non-iterative DFT-based frequency estimators in the presence of strong quasi-harmonic interference and noise, revealing performance degradation in traditional methods. The study identifies estimators that... Read more
Key finding: Develops a detection-theoretic framework to quantify the minimum resolvable frequency difference between two closely spaced sinusoids in noise, providing an explicit formula linking resolution to SNR and desired... Read more

2. What algorithmic methods enable reliable real-time harmonic source identification and mitigation in power systems with measurement errors and dynamically varying conditions?

This theme covers techniques for locating harmonic emission sources in complex electrical grids, particularly under measurement error and transient load changes. Methods integrating robust signal processing indices, adaptive filtering, and machine learning have been proposed to deliver accurate harmonic current detection and enable active compensation in power systems. These approaches are vital for maintaining power quality and system stability in the presence of nonlinear loads and renewable integration.

Key finding: Demonstrates that current Total Harmonic Distortion (THD) serves as a reliable index for identifying multiple harmonic source locations in power systems even with ±1% uniformly distributed measurement errors. Comparisons with... Read more
Key finding: Presents and compares harmonic detection algorithms based on Extended Kalman Filters (EKF), wavelet transforms, and synchronous rotating frames for real-time estimation of fundamental current components in three-phase... Read more
Key finding: Proposes an improved p-q harmonic detection algorithm integrated with fuzzy logic to mitigate overshoot and reduce transient response time during sudden load changes in Hybrid Active Power Filters (HAPF). Simulation shows... Read more
Key finding: Implements feed-forward neural networks (FFNNs) with one and two hidden layers for harmonic detection in active power filters. Using simulated distorted waves with dominant 5th, 7th, 11th, and 13th harmonics for training, the... Read more

3. How can novel theoretical models and statistical learning approaches facilitate improved harmonic signal parameter estimation and classification?

This theme encompasses the development and application of theoretical frameworks—such as modern polarization theory, phase-based signal analysis, and adaptive linear learning—for precise harmonic parameter estimation, signal classification, and feature extraction. Such methods aid in interpreting complex harmonic phenomena across physics, audio signal processing, and electromagnetic detection, providing more reliable and interpretable harmonic characterizations.

Key finding: Develops a direct ab initio methodology based on Maximally Localized Wannier Functions and modern polarization theory to predict second harmonic generation (SHG) signals at interfaces, accounting for both local molecular... Read more
Key finding: Introduces a novel harmonic signal frequency estimation method leveraging statistical analysis of instantaneous phase probability densities and their derivatives in noisy signal mixtures, including multilevel phase and... Read more
Key finding: Reviews the application of Adaline-based adaptive linear networks for real-time estimation of harmonic amplitudes in Fourier series representations of periodic signals. The LMS learning algorithm efficiently converges to the... Read more
Key finding: Proposes a Frequency-Swept Harmonic Radar (FSHR) technique that exploits frequency diversity of harmonic re-radiation from nonlinear electronic circuits to classify devices. Statistical and Fourier feature extraction from... Read more
Key finding: Introduces a computationally efficient test that distinguishes between nuisance signals (known-frequency harmonic combinations) and signals containing additional unknown-frequency harmonics by leveraging Fourier transform... Read more

All papers in Harmonic detection

In this study, the method to apply the feed forward neural networks with two different numbers of hidden layers for harmonic detection process in active filter are described. We have simulated the distorted wave including 5th, 7th, 11th,... more
In this study, the method to apply the feed forward neural networks with two different numbers of hidden layers for harmonic detection process in active filter are described. We have simulated the distorted wave including 5th, 7th, 11th,... more
Simultaneous acquisition of the real and imaginary components in
In this study, the method to apply the feed forward neural networks with two different numbers of hidden layers for harmonic detection process in active filter are described. We have simulated the distorted wave including 5th, 7th, 11th,... more
Normally, when research on active compensation models, previous studies only assumed that the source of harmonics is nonlinear load. The nonlinear load here is fixed and balanced, the supply voltage is considered ideal, i.e. the... more