摘要
提出了基于粗糙集和神经网络的故障诊断方法。采用Kohonen网络对连续属性值进行离散化,应用粗糙集理论对特征参数进行属性约简,并把约简结果生成规则作为BP网络的输入。仿真结果表明,经粗糙集理论优化后的样本集进行神经网络训练,提高了神经网络的学习速度和故障诊断正确率,减少了训练时间。
A fault diagnosis method based on rough set and neural network is introduced. The continuous attributes are discretized with a Kohonen neural network, rough sets theory is used to simplify the attribute parameters. The reduction results are transformed into rules, which are used as input of the BP neural network.The simulations show that the learning speed and diagnosis correctness are greatly improved after the training data is processed by rough sets, and the computation time is decreased by using rough set theory.
出处
《科技信息》
2009年第29期116-117,110,共3页
Science & Technology Information
关键词
粗糙集
属性约简
神经网络
故障诊断
Rough set
Attribute reduction
Neural network
Fauh diagnosis
