Fault line detection method based on the improved SVD de‐noising and ideal clustering curve for distribution networks. Issue 2 (1st March 2018)

• Record Type: Journal Article
• Authors: Wei, Xiangxiang; Wen, Boying; Yang, Dechang; Wang, Bin; Wang, Xiaowei; Yang, Xuefei; Gao, Jie
• Is Part Of: IET science, measurement & technology. Volume 12:Issue 2(2018)
• Page Start: 262
• Page End: 270
• Publication Date: 2018-03-01
• Subjects: fault diagnosis -- power distribution faults -- pattern clustering -- singular value decomposition -- fast Fourier transforms -- signal denoising -- phase measurement; 1/4 cycle damping nonperiodic component -- periodic component -- phase difference calculation -- baseband signal analysis -- fast Fourier transformation -- transient zero sequence current -- improved singular value decomposition denoising algorithm -- adaptive fault line detection method -- distribution network -- ideal clustering curve -- improved SVD denoising algorithm; Measurement -- Periodicals; Electrical engineering -- Periodicals; Electronics -- Periodicals; Nanotechnology -- Periodicals; Electromagnetism -- Periodicals; Medical instruments and apparatus -- Periodicals; 621.3
• DOI: 10.1049/iet-smt.2017.0163
• Languages: English
• ISSNs: 1751-8822

Abstract

Abstract : This study proposes an adaptive fault line detection method based on the improved singular value decomposition (SVD) de‐noising and ideal clustering curve for distribution networks. First, the improved SVD algorithm is introduced to obtain the pure transient zero sequence current. Then, the fast Fourier transformation is employed to analyse the baseband signal and calculate the phase differences. After that, if the difference values are larger than the set threshold, the detection method based on the improved SVD and ideal clustering curve of baseband components is proposed; if not, the 1/4 cycle damping non‐periodic components are rearranged, then based on the first half waveform extreme values and the rearranging damping non‐periodic components, the detection method based on the ideal clustering curve of damping components is introduced. The simulation results prove the correctness of proposed selection method. After comparing with the existing methods, the advantages of proposed method are confirmed.