High-voltage equipment condition monitoring and diagnosis system based on information fusion
- 159 Downloads
As high-voltage electric equipment has complex structure and works in harsh environments, fiber Bragg grating (FBG) sensors are applied to realize the real-time monitoring of some parameters in which temperature is the main parameter. Using FBG sensors to monitor temperature of high-voltage electric equipment can overcome the disadvantages of harsh monitoring environment such as high-voltage, big current, strong electromagnetic interference and so on. The fault of high-voltage electric equipment is difficult to be distinguished as there may be many different reasons. The traditional or simple methods cannot totally meet the demand of fault diagnosis of high-voltage electric equipment. First, taking neural network as a classifier to distinguish different fault types from complex fault information in the feature layer can supply a good foundation to final information fusion diagnosis. Second, Dempster–Shafer evidence theory is used to make a comprehensive diagnosis of fault information in the decision layer. All the uses above can increase the speed and accuracy of diagnosis and have practical significance. The fault diagnosis system shows good results and provides an effective way to realize the real-time condition monitoring and more accurate fault diagnosis of high-voltage electric equipment.
KeywordsInformation fusion Condition monitoring Fault diagnosis FBG sensing system Neural network
This article was supported by the Natural Science Foundation of China (Grant No. 60674107), the Natural Science Foundation of Hebei Province (Grant No. F2006000343), and the Research and Development Plan of Science and Technology of Shijiazhuang (No. 06713026A).
- 2.Tenbohlen S, Figel F (2000) On-line condition monitoring of power transformers. Power Engineering Society Winter Meeting. IEEE 3:2211–2216Google Scholar
- 4.Shenghe Sun (2002) The fiber measurement and sensing technology. Harbin Poly Technical University Press, HarbinGoogle Scholar
- 8.Booth C, McDonald JR (1998) The use of artificial neural networks for condition monitoring of electrical power transformers. Neural Comput 23:97–109Google Scholar
- 15.Yanjing S, Shen Z, Changxin M (2007). Improved BP Neural Netw Transformer Fault Diagn China Univ Min Technol 17:138–142Google Scholar