Abstract
In AC/DC hybrid power system, AC system failures and commutation valve trigger pulse disorder will lead to commutation failure, which may lead to DC voltage fluctuations, power transmission interruption and other serious consequences. In order to accurately and effectively identify the specific causes of commutation failures, a double deck traceability identification method is proposed in this paper. The surface identification based on wavelet entropy and affinity propagation (AP) algorithm can distinguish internal and external faults. The deep identification uses convolution neural network which can further lock the specific cause of commutation failures. In this paper, 1) the various factors leading to commutation failures are analyzed; 2) the fault feature space consists of the wavelet analysis components of DC voltage signal, and the AP algorithm is used to identify the surface source; 3) the DC current, AC voltage and current signals are added into the sample matrix of fault time-space, and the convolution neural network is used to identify the deep traceability. Finally, the accuracy of the method is verified by using the typical HVDC model.
Highlights
Commutation failure (CF) is one of the typical faults of traditional high-voltage direct current (HVDC) system [1]
The different types of wavelet entropy are used to show the characteristics of electrical signals to construct the fault feature space in this paper, which includes the wavelet energy entropy (WEE), wavelet singular entropy (WSE), wavelet distance entropy (WDE) [21]–[23]
DC voltage is selected as the input data source signal for surface identification method
Summary
Commutation failure (CF) is one of the typical faults of traditional high-voltage direct current (HVDC) system [1]. Y. Wang et al.: Research on Double-Deck Traceability Identification Method generalized regression neural network was proposed in [17]. AC current was monitored during the CF, and wavelet packet decomposition was carried out to construct a new energy spectrum On this basis, the generalized return neural network in two modes was used to judge whether commutation failures occurs or not. The generalized return neural network in two modes was used to judge whether commutation failures occurs or not It can distinguish the specific fault cases. The decomposition of DC current and AC voltage was realized by wavelet analysis, and the specific fault types were distinguished and identified according to the wavelet energy statistics of each layer in [18].
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
