摘要
针对单一分类器进行故障诊断时诊断精度不高、随机性强的问题,提出一种基于改进D-S证据理论的滚动轴承故障诊断方法,构建了信息融合诊断框架。首先,利用BP神经网络、支持向量机、径向基神经网络构建初步诊断层,将提取的特征信息进行初步诊断;然后,利用改进的D-S证据理论构建融合诊断层,将初步诊断层的诊断结果进行融合,并根据诊断规则得到最终的诊断结果;最后,采用不同的信息融合方法对滚动轴承故障数据进行对比研究。试验结果表明:使用改进D-S证据理论的滚动轴承故障诊断方法能够有效提高证据可信度,降低不确定性,提高故障诊断精度和故障诊断模型的鲁棒性。
A fault diagnosis method of rolling element bearings based on improved D-S evidence theory is proposed and a frame of information fusion base diagnosis is constructed. Firstly, the primary diagnosis layer is constructed by back propagation(BP) neural network, support vector machine(SVM) and radial bases function(RBF) neural network for the purpose of diagnosing the fault with the extraction features. Then the information fusion diagnosis layer is constructed by the improved D-S evidence theory, fusing the diagnosis results coming from the primary diagnosis layer and getting the final diagnosis result. Finally, the fault diagnosis of rolling element bearing based on different fusion method is studied. The experiment result indicates that the proposed method can enhance the reliability and decrease the uncertainty effectively, which improves the diagnosis accuracy and robustness of the fault diagnosis model.
作者
张钢
田福庆
梁伟阁
佘博
ZHANG Gang;TIAN Fu-qing;LIANG Wei-ge;SHE Bo(College of Weaponry Engineering, Naval Univ, of Engineering, Wuhan 430033, China)
出处
《海军工程大学学报》
CAS
北大核心
2019年第4期42-47,共6页
Journal of Naval University of Engineering
基金
国家自然科学基金资助项目(61640308)
海军工程大学自主立项资助项目(20161579)
关键词
滚动轴承
故障诊断
改进D-S证据理论
信息融合
多分类器
rolling element bearing
fault diagnosis
improved D-S evidence theory
information fusion
multi-classification