摘要
研究一种将模糊c均值聚类算法与神经网络相结合的模糊聚类神经网络(FKCNN),在分析该网络的结构和学习方法的基础上,对FKCNN的学习算法进行了一定的改进,并将改进后的模糊聚类神经网络应用于模拟电路的故障诊断,探讨了实现方法,设计了算法步骤,并举例对算法有效性进行了验证。结果表明,运用模糊聚类神经网络能够实现对具有状态可测性的模拟电路的故障诊断。
The Fuzzy Kohonen Clustering Neural Network（FKCNN）,which combines fuzzy c-means clustering algorithm and Kohonen neural network,has excellent capability in parallel processing and pattern partitioning.Based on analysis to its structure and learning method,an improvement was made to the learning method of FKCNN.Then the improved FKCNN was used in fault diagnosis of analog circuits.The approach for realizing the algorithm was designed and the validity of the algorithm was verified by using an example. The result of the experiment showed that the fault diagnosis method based on the improved FKCNN is adapted well to analog circuits when the circuits have well measurability.
出处
《电光与控制》
北大核心
2009年第11期64-66,83,共4页
Electronics Optics & Control
基金
空军武器装备军内科研项目(KJ07114)
关键词
故障诊断
模拟电路
模糊C均值
聚类
神经网络
fault diagnosis
analog circuits
fuzzy c-means
clustering
neural network
