摘要
为了提高配电网故障诊断的准确性和效率,提出了将粗糙集与自适应神经网络模糊系统(adaptive neuro-fuzzy inference system,ANFIS)相结合构建粗糙集和神经网络的智能混合诊断系统,以充分利用粗糙集理论对知识的约简能力和神经网络的容错学习能力.通过粗糙集理论中的信息熵概念对诊断系统输入变量进行合理选择,即选取与故障诊断信息相关性大的参数作为输入,然后利用ANFIS进行建模和参数辨识,并通过训练样本进行学习训练,这样既减少了神经网络的学习训练时间,又提高了诊断的准确度.用该方法对某一实际配电网进行了故障诊断,结果表明:该方法计算速度快,具有良好的容错性能和在线故障诊断潜力.
In order to improve the accuracy and efficiency of fault diagnosis in distribution network, this paper puts forward an intelligent hybrid diagnosis system, which combines rough set theory with adaptive neutral fuzzy inference system theory, to make the best use of the rule reduction in rough set theory and the capabilities of fault-tolerance and learning in neutral network. Through the reasonable selection of input variables in the diagnostic system in such a way that more correlative parameters with fault diagnosis information are chosen as input variables to form the simplified rule sets, with ANFIS （ Adaptive Neuro-Fuzzy Inference System） the most simplified rule sets are called to make modeling and parameter identification, then learning training is done by training samples. So it not only reduces the learning training time, but also improves the accuracy of diagnosis. Compared with the results of fault diagnosis for a certain distribution network in conventional rough set theory, it shows that with this method it will compute faster, have better fault-tolerance and potential of on-line fault diagnosis.
出处
《重庆工学院学报（自然科学版）》
2009年第6期45-50,共6页
Journal of Chongqing Institute of Technology
基金
国家自然科学基金资助项目(50607023)
重庆工学院博士启动基金(07-60-26)
关键词
配电网
故障诊断
信息熵
ANFIS模型
distribution network
fault diagnosis
rough set theory
information entropy
ANFIS model
