Fault Recognition of On-Load Tap-Changer Based on Improved BP Neural Network
Download as PDF
DOI: 10.23977/iccsc.2017.1005
Corresponding Author
Chunbing Jiang
ABSTRACT
On-load tap-changer (OLTC) is an important part in power transformer, which is
used to change the transformation ratio by switching contact form one winding tap to
another without interrupting the load. It is necessary to make sure the effective running of
OLTC. Thus, a fault recognition method for OLTC based on improved BP neural network is
proposed in this paper, according to the working characteristicsof OLTC. The feature of this
improved neural network is that the learning rate is variable. Experimental results on 2000
test samples from offline AC test device have showed the effectiveness of the proposed
fault recognition method.
KEYWORDS
Fault recognition, On-load tap-changer, Neural network, Variable learning rate.