•  
  •  
 

Keywords

distribution transformer; improved Prony algorithm; second harmonic content; multi-feature discrimination system

Abstract

The magnetizing inrush current will be generated when the distribution transformer is closed with no load. If its value is greater than the setting value of line protection, the circuit breaker will fail to close, Therefore, it is necessary to identify the magnetizing inrush current and the general short-circuit current so that the circuit breaker can avoid the magnetizing inrush current and achieve normal closing. A comprehensive multi-feature identification method of magnetizing inrush current in distribution transformers based on an improved Prony algorithm is proposed, and the characteristics of magnetizing inrush current of one and multiple distribution transformers are analyzed. Backward-prediction link is introduced to improve the traditional Prony algorithm and improve the accuracy of magnetizing inrush current identification. The advantages, disadvantages, and application scope of the fundamental wave attenuation factor method, the second harmonic discrimination method, and the wavelet decomposition identification method of the magnetizing inrush current are comprehensively compared. A comprehensive identification method of the magnetizing inrush current and fault current of the distribution transformers based on a multi-feature discrimination system is put forward to improve the identification accuracy of the magnetizing inrush current. The recorded samples of field fault current and magnetizing inrush current are taken as identification objects, and the proposed algorithm is compared with the single feature method and wavelet decomposition identification algorithm, verifying its effectiveness and engineering feasibility.

DOI

10.19781/j.issn.1673-9140.2024.06.008

First Page

69

Last Page

78

Download

Share

COinS