摘要
针对旋转机械故障和故障征兆关系的复杂性及神经网络在故障诊断中存在网络结构复杂和训练时间长等问题,提出了一种基于粗糙集与神经网络结合的故障诊断方法;采用自组织映射方法对属性进行离散化,设计了一种自适应遗传算法对属性进行约简,将获得的最小条件属性集作为神经网络的输入;以轴承的故障诊断为例进行分析,结果表明,该方法在保证诊断正确率的同时,可以有效简化神经网络的结构,降低网络的训练时间;另外,设计的自适应约简算法在保证获得最小约简的基础上,大大加快了收敛速度;该方法可推广应用在其它机械设备的故障中。
To overcome the problem of structure complexity and long training time in neural network method for fault diagnosis of rotating machine with fuzzy fault feature, a new fault diagnosis method based on rough set and neural network is presented. The self-organizing map method is used to get the discrete attributes fist, then an adaptive genetic algorithm is devised for attribute reduction, and finally the results of the attribute reduction is regard as the inputs of the neural network. The experimental results show that the reduction in rough set is improved, the structure of neural network is optimized, and the computational complexity is decreased.
出处
《计算机测量与控制》
CSCD
北大核心
2010年第1期64-66,82,共4页
Computer Measurement &Control
基金
国家自然科学基金(60873196)
西北工业大学科技创新基金(W016144)
关键词
粗糙集
自适应遗传算法
神经网络
旋转机械
故障诊断
rough set
adaptive genetic algorithm
neural network
rotating machine
fault diagnosis
