粗糙集与神经网络故障诊断组合方法应用  

Application of Rough Set-Neural Network Combinatorial Fault-diagnosing Method

作　　者：安文斗[1] 刘平 吴存良[2] 于乐泉[1] 

机构地区：[1]煤炭科学研究总院重庆分院,重庆400037 [2]攀枝花煤业集团公司,四川攀枝花617066

出　　处：《煤矿机电》2008年第1期34-36,共3页Colliery Mechanical & Electrical Technology

摘　　要：利用粗糙集理论对知识的约简能力及模糊神经网络的分类能力,构建粗糙集—神经网络故障诊断组合模型(RNN),具有良好的拓扑结构,学习速度大为提高。应用电力变压器实例验证,RNN模型诊断速度快,故障诊断正确率高。Considering the reduction ability of rough set theory and the classification ability of fuzzy neural network, a rough set - neural network combinatorial fault-diagnosing model is constructed. The model enjoys a better topological structure and greatly increased speed for learning. The practical application to power transformer verifies that the model has comparably fast and accurate diagnosing abilities.

关 键 词：粗糙集 人工神经网络 故障诊断 粗糙集-神经网络 

分 类 号：TM411[电气工程—电器]