The Fault Diagnosis Research Of Voltage Sag In Distribution Network

A variety of electrical equipments and electronic devices in all walks of life have been used more and more widely in the age of the technology development rapidly. Resulting transient power quality problems are more prominent, and the voltage sag problems are the main reason of resulting transient power quality problems. When the voltage sag appeared in the power system, they may cause a massive power outage or other hazards, which results in economic and social benefits reduced greatly. So how can we detect he fault fast and accuratly is the current hot topic.Based on the current research status of power quality and voltage sag all over the world, this paper makes some analysis and summary for the characteristics and damage as well as common detection methods of voltage sag.Focusing on the voltage sag of 66kV distribution network and Researching and improving the dq transform method and ?? transform method. The improved algorithm has good effect in real-time, rapidity and accuracy.In order to make the voltage sag be token more reliable, the paper applies the method of combining wavelet packet and probabilistic neural network to detect the voltage sag. Analysis results show that the method can extract the eigenvalue effectively, and make the eigenvalue as the input of the probabilistic neural network (PNN) which can identify the distribution network or line whether appear the voltage sag or not effectively.