基于软竞争Yu范数自适应共振理论的轴承故障诊断方法
发布时间：2018-11-10 15:20
【摘要】：传统自适应共振理论网络模型利用硬竞争机制对故障类边界处的样本进行分类时易造成误分类,为此,提出了基于软竞争Yu范数自适应共振理论的轴承故障诊断方法。将基于模糊竞争学习的软竞争方法引入Yu范数自适应共振理论模型中,根据模式节点与输入样本间隶属度的大小,对竞争层多个节点进行训练和学习。通过对轴承故障试验数据的诊断分析可知,该方法不但能有效识别不同类型的故障,而且能识别不同严重程度故障,且诊断精度优于自适应共振理论模型和模糊C均值聚类模型。
[Abstract]:The traditional network model of adaptive resonance theory makes use of hard competition mechanism to classify samples at fault class boundary which is prone to misclassification. Therefore a bearing fault diagnosis method based on soft competition Yu norm adaptive resonance theory is proposed. The soft competition method based on fuzzy competition learning is introduced into the Yu norm adaptive resonance theory model. According to the size of membership degree between the mode node and the input sample, multiple nodes in the competition layer are trained and learned. Through the diagnosis and analysis of bearing fault test data, it can be seen that this method not only can effectively identify different types of faults, but also can identify different severity of faults. The diagnostic accuracy is better than the adaptive resonance theory model and fuzzy C-means clustering model.
【作者单位】： 武汉科技大学机械自动化学院;
【基金】：国家自然科学基金资助项目(51405353)
【分类号】：TH133.3
[Abstract]:The traditional network model of adaptive resonance theory makes use of hard competition mechanism to classify samples at fault class boundary which is prone to misclassification. Therefore a bearing fault diagnosis method based on soft competition Yu norm adaptive resonance theory is proposed. The soft competition method based on fuzzy competition learning is introduced into the Yu norm adaptive resonance theory model. According to the size of membership degree between the mode node and the input sample, multiple nodes in the competition layer are trained and learned. Through the diagnosis and analysis of bearing fault test data, it can be seen that this method not only can effectively identify different types of faults, but also can identify different severity of faults. The diagnostic accuracy is better than the adaptive resonance theory model and fuzzy C-means clustering model.
【作者单位】： 武汉科技大学机械自动化学院;
【基金】：国家自然科学基金资助项目(51405353)
【分类号】：TH133.3
【参考文献】
相关期刊论文 前8条
1 徐增丙;李友荣;王志刚;轩建平;;基于ART和Yu范数的聚类方法在齿轮故障诊断中的应用[J];武汉科技大学学报;2016年02期
2 王洪明;郝旺身;韩捷;董辛e，
本文编号：2322812
本文链接：https://www.wllwen.com/jixiegongchenglunwen/2322812.html
