摘要
Abstract
In order to effectively extract useful information from chaotic time series,this article proposes a novel fault diagnosis method based on Hierarchical Differential Symbolic Entropy(HDSE)and Hierarchical Prototype classifier(HP).Firstly,in response to the shortcomings of Differential Symbolic Entropy(DSE),the article proposes HDSE and uses HDSE to comprehensively extract features from the original vibration signal.Secondly,in order to avoid excessive redundant information in the sample that affects the accuracy of fault recognition and to extract feature information at a deeper level,Linear Discriminant Analysis(LDA)is used to perform dimensionality reduction on the feature vectors.Finally,a Hierarchical Prototype classifier was trained using feature vectors and its performance was tested.In order to test the effectiveness of the method proposed in this article,bearing test-bench data was used for experimental analysis.Through experimental analysis,it was found that the fault recognition accuracy reached95.423%.关键词
轴承/层次差分符号熵/线性判别分析法/HP分类器/故障识别Key words
bearings/hierarchical differential symbol entropy/linear discriminant analysis/HP classifier/fault identification分类
机械工程
