Fault Diagnosis of Rolling Bearing Based on Fisher Discrimination Sparse Coding

摘要：

　　In response to mechanical fault in feature extraction problem,this paper presents a Fisher discrimination sparse coding method.This method is achieved by optimizing an objective function that includes two steps.First,this objective function works well in denoising where signals need to be reconstructed.Second,another objective function is added to the sparse coding framework,the discrimination power of the Fisher discriminative methods with the reconstruction property,and the sparsity of the sparse representation that can deal with the fault signal which is corrupted.Finally,the feature is extracted.In rolling bearing fault classification experiments,the new method improves the accuracy of classification.

关键词： Feature extraction Sparse coding Fisher discrimination

作者: Chengliang Li Zhongsheng Wang Chan Ding

作者单位: Northwest Polytechnical University, Xi’an, People’s Republic of China Xi’an Aerospace Precision Electromechanical Institute, Xi’an, People’s Republic of China

会议类型: 国际会议

会议名称: 2013西安国际航空维修与管理学术会议

会议地点: 西安

会议语种:英文

页码: 387-394

在线出版日期: 2013-11-25（万方平台首次上网日期，不代表论文的发表时间）