Status Monitoring for Nuclear Power Plant Using Integrated Neural Networks

摘要：

In order to improve the capacity of status monitoring for nuclear power plants (NPPs), a new intelligent monitoring method based on integrated neural networks (IANNs)is investigated in this paper. In this method, multiple neural networks (ANNs)which the types of neural networks are different were trained for the status monitoring of NPPs. The operation parameters of NPP, which having importantly impact on the nuclear safety, were selected as the input variables of neural networks. The inputs of neural networks are signals of monitored operation parameters and the outputs of neural networks axe fault patterns of NPP. The outputs of neural networks are fused using D-S evidence theory. The results of status monitoring for NPP are obtained by fusing the diagnosing results of different neural networks by two stages information fusion. The typical operation patterns of NPP were used to demonstrate the feasibility of the proposed monitoring method. The results reveal that employing integrated neural networks can improve the capacity of status monitoring of NPP.

关键词： Nuclear power plant Integrated neural networks Status monitoring Information fusion

作者: Zhou Gang Wang Xin-ye Peng Wei Yang Li

作者单位: College of Architecture and Power, Naval University of Engineering, Wuhan 430033, China Department of Science and Research, Naval University of Engineering, Wuhan 430033, China

会议类型: 国际会议

会议名称: ISSNP2008、CSEPC、ISOFIC2008(第二届21世纪和谐核电系统国际会议、第四届电厂控制中认知系统工程方法国际会议暨第三届未来核电厂仪表与控制国际会议)

会议地点: 哈尔滨

会议语种:英文

页码: 533-539

在线出版日期: 2008-09-08（万方平台首次上网日期，不代表论文的发表时间）