RBF人工神经网络在核电厂故障诊断中的应用

留言板

尊敬的读者、作者、审稿人, 关于本刊的投稿、审稿、编辑和出版的任何问题, 您可以本页添加留言。我们将尽快给您答复。谢谢您的支持!

姓名

邮箱

手机号码

标题

留言内容

验证码

Application of RBF Artificial Neural Network to Fault Diagnose in Nuclear Power Plant

1. College of Power and Nuclear Energy, Harbin Engineering University, 150001, China;
2. Jiujiang Subsection of NO. 707 Research Institute, China Ship Building Industry Corporation, Jiujiang, Jiangxi, 332007, China

摘要: 对核电厂二回路凝给水系统常见故障原因进行了分析,结合专家知识建立了二回路凝给水系统故障知识库;在此基础上,将径向基函数(RBF)人工神经网络引入到核电厂故障诊断中。由于采用了动态RBF网络设计方法,使神经网络的规模较小,同时具有较高的泛化能力,提高了神经网络的诊断速度及准确性。并使用VC++语言建立了一个故障诊断系统。
- RBF神经网络 /
- 故障诊断 /
- 核电厂
Abstract: Some faults of condensation and feed water system in nuclear power plants are analyzed and a fault knowledge base is established based on the experts’ knowledge at the same time. RBF artificial neural network is introduced to the fault diagnose in nuclear power plants. Because of the use of dynamic designing method of RBF network, not only the scale of RBF artificial neural network is smaller, but also its generalization ability is higher, which improve the speed and accuracy of diagnose system. Finally a fault diagnose system is founded by VC++.
- RBF artificial neural network /
- Fault diagnose /
- Nuclear power plant