OLTC fault diagnosis method based on instantaneous energy entropy and SVM

Abstract

The invention discloses an OLTC fault diagnosis method based on an instantaneous energy entropy and an SVM. The method comprises the following steps of: 1, attaching a vibration detection probe to thetop end of the box wall of an on-load tap-changer, and acquiring 80 groups of vibration signals generated in the action process of the on-load tap-changer in different states, such as normal state, loose state, wear state, burning state and the like; 2, firstly, wavelet packet noise reduction is carried out on the vibration signal, and then EMD decomposition is carried out on the de-noised signal; 3, IMF instantaneous energy change of each unit after decomposition is solved, and entropy values of the instantaneous energy change are taken to form feature vectors; and 4, forming a feature vector by using the entropy as an input of a support vector machine to realize OLTC fault classification. The working state of the transformer on-load tap-changer can be monitored in real time, and the requirement for real-time fault diagnosis of the transformer on-load tap-changer is met. Data support and theoretical basis are provided for targeted maintenance, and manpower, material resources and time are prevented from being wasted.

Description

technical field [0001] The invention relates to an OLTC fault diagnosis method based on instantaneous energy entropy and SVM, and belongs to the technical field of power equipment signal monitoring. Background technique [0002] On-load tap-changer (OLTC) is an important part of power transformer, and its operating status is directly related to the stability and safety of the transformer and system. OLTC is one of the components with the highest failure rate in transformers. Its failure not only directly affects the operation of the transformer, but also affects the quality and operation of the power grid. According to the statistics in the ditch, the accidents caused by OLTC faults account for about 28% of the total accidents of transformers, and the fault types are basically mechanical faults, such as loose contacts, falling off contacts, mechanism jamming, slipping, and refusal to move. Mechanical failure will directly damage the OLTC and the transformer itself, and the...

AI Technical Summary

Problems solved by technology

The power outage maintenance period of on-load tap-changers is often long, and it is difficult to detect early mechanical failures in time. Failures and damages often occur before power outage maintenance, and power outage maintenance affects the normal operation of the transformer, which requires a lot of manpower, material and financial resources

[0004] On-line monitoring methods mainly include thermal noise-based diagnosis method and vibration-based on-line monitoring, etc. The thermal noise-based diagnosis is that the thermal noise generated by the heat generation after the transformer tap-changer breaks down spreads to the outside of the transformer, and is detected by installing a noise sensor on the transformer shell. To carry out tap changer fault diagnosis, but when the thermal noise is transmitted to the sensor, the energy loss is too large, and various noises interfere with the engineering application, which is difficult to realize

Images