Year: 2022
Type: article
Abstract: Traditional high-voltage reactor monitoring and diagnosis research has problems such as high sampling demand, difficulty in noise reduction on site, many false alarms, and lack of on-site data. In order to solve the above problems, this paper proposes an acoustic–electric fusion high-voltage reactor acquisition system and defect diagnosis method based on reactor pulse current and ultrasonic detection signal. Using the envelope peak signal as the basic detection data, the sampling requirement of the system is reduced. To fill the missing data with partial discharge (PD) information, a method based on k-nearest neighbor (KNN) is proposed. An adaptive noise reduction method is carried out, and a noise threshold calculation method is given for the field sensors. A joint analysis method of acoustic and electrical signals based on correlation significance is established to determine whether a discharge event has occurred based on correlation significance. Finally, the method is applied to a UHV reactor on the spot, which proves the effectiveness of the method proposed in this paper.
Source: Energies
DOI: 10.3390/en15197196
Authors: Shuguo Gao, Chao Xing, Zhigang Zhang, Chenmeng Xiang, Haoyu Liu, Hongliang Liu, Rongbin Shi, Sihan Wang, Guoming Ma
Language: English
Subfield: Materials Chemistry
Field: Materials Science
Domain: Physical Sciences
Open Access status: gold
Estimated APC paid: $2,303
OpenAlex ID: W4303644287