Extraction Method and Application of Traveling Wave Natural Frequencies Based on PSO and VMD

• Published in: 2020 Chinese Automation Congress (CAC) pp. 28 - 34
• Main Authors: Na, Wu, Zhengxian, Yang, Shuxian, Fan
• Format: Conference Proceeding
• Language: English
• Published: IEEE 06.11.2020
• Subjects: Data mining; Entropy; fault location; natural frequencies; PSO; Sociology; spectrum aliasing; Statistics; Time-domain analysis; Time-frequency analysis; Transient analysis; VMD
• ISSN: 2688-0938
• DOI: 10.1109/CAC51589.2020.9327146

The realization of fault location method based on traveling wave natural frequency depends on the extraction of traveling wave natural frequency. When the fault occurs on the power system with one or more nodes, there will be multiple reflections at the discontinuous wave impedance, which makes it difficult to extract the natural frequency principal component that represents the fault distance in the spectrum. Upon this foundation, a traveling wave natural frequency extraction method based on PSO and VMD is proposed in this paper. Firstly, PSO algorithm was used to calculate the envelope entropy of VMD at different parameters, and to select the optimal parameter combination corresponding to the minimum envelope entropy. Then, the fault traveling wave is decomposed by VMD based on the optimal parameter combination, and to extract principal components of natural frequency by analyzing spectrum. Finally, the simulation model of T-type transmission line is built to verify the reliability of this method. The simulation results show that this method can effectively suppress the influence of spectrum aliasing, accurately extract the natural frequency principal component including fault information, and the ranging error is small.