Integrated neural network fault diagnosis technique and applications

摘要：

For maximum improving neural network fault diagnosis rate of equipment, the characteristics and existing problems of single sub-NN fault diagnosis are analyzed; Integrated neural network based on the signal types and various eigenvector combinations is proposed from the view of information fusion. A three-layer-series NN is constructed by building information distribution network, sub-NN and decision-making information fusion network. To use the neural network in transformer fault diagnosis has obtained better effect by effective combination of input signal and fault diagnosis from different side.

关键词： integrated neural net work fault diagnosis transformer sub-NN information fusion

作者: Hongqi Wu Xiaobin Li

作者单位: Department of Electron and Electricity Engineering Baoji University of Arts and Sciences Baoji, Shanxi Province, China

会议类型: 国际会议

会议名称: 2010 Second Asia-Pacific Conference on Information Processing(2010年第二届亚太地区信息处理国际会议 APCIP 2010)

会议地点: 南昌

会议语种:英文

页码: 47-49

在线出版日期: 2010-09-17（万方平台首次上网日期，不代表论文的发表时间）