摘要
为了提高不完备信息系统故障诊断的正确性与效率,提出一种基于粗糙集理论、蚁群优化算法和RBF神经网络相结合的故障智能诊断方法;该方法首先利用"条件组合补齐算法"对不完备的数据进行完备化处理,再利用粗糙集对条件属性进行知识约简,得到具有最大完备度的最小规则集,接着用蚁群算法优化RBF神经网络的权值,并将最小规则集用于训练RBF神经网络模型,获得故障智能诊断模型;通过实际工程数据验证故障智能诊断模型的有效性,结果表明提出的方法能有效实现系统故障的诊断。
In order to improve fault diagnosis correctness and efficiency of incomplete information system,an intelligent diagnosis method of fault based on rough set（RS）,ant colony optimization（ACO）algorithm and radial basis function（RBF）neural network is proposed.In this intelligent diagnosis method,the combination and condition supplement algorithm is used to deal with the incomplete data with the maximum completeness.The RS as a new mathematical tool is used to remove redundant information in order to obtain the minimum rule set.Then the ACO algorithm is directly used to optimize the weights of RBF neural network in order to establish an optimized RBF neural network model,then the minimum rule set is inputted the optimized RBF neural network model in order to obtain an intelligent diagnosis model.The actual data are used to verify the effectiveness of intelligent diagnosis model.The experiment results show that the proposed intelligent diagnosis method can effectively diagnose the faults of system.
作者
周頔
Zhou Di(Industrial Technological Institute of Intelligent Manufacturing, Sichuan University of Arts and Science, Dazhou 635000, Chin)
出处
《计算机测量与控制》
2018年第9期5-8,共4页
Computer Measurement &Control
基金
自然科学基金(2014JY0111)
四川省教育厅科技计划项目(18ZA0415)
关键词
智能诊断
不完备信息系统
粗糙集理论
蚁群算法
神经网络
intelligent diagnosis
incomplete information system
rough set theory
ant colony optimization
RBF neural network
