基于线性调频Z变换和小波包分解的光伏故障电弧检测方法
By combining linear frequency modulated Z-transform and wavelet packet decomposition with extreme learning machine neural network, the environmental dependence and accuracy problems of DC arc detection are solved, and high-accuracy and fast fault arc identification is achieved.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- CHONGQING UNIV OF POSTS & TELECOMM
- Filing Date
- 2023-04-04
- Publication Date
- 2026-07-17
AI Technical Summary
Existing DC arc detection methods are susceptible to environmental factors and have low accuracy. Furthermore, traditional single-frequency-band, single-index detection methods are unstable and difficult to accurately identify fault arcs.
A method based on linear frequency modulated Z-transform and wavelet packet decomposition, combined with an extreme learning machine neural network, is adopted to extract characteristic frequency bands and characteristic quantities through signal preprocessing, fast Fourier transform, linear frequency modulated Z-transform and wavelet packet decomposition, which are used for fault arc detection.
It enables rapid detection of fault arcs within 0.1 seconds, improving detection accuracy and enabling accurate differentiation between faults and normal operation in photovoltaic systems, overcoming the instability of traditional methods.
Smart Images
Figure CN116526969B_ABST