Condition Recognition Method of Rolling Bearing Based on Ensemble Empirical Mode Decomposition Sensitive Intrinsic Mode Function Selection Algorithm
Wang Yu-jing①② Kang Shou-qiang② Zhang Yun① Liu Xue② Jiang Yi-cheng① Mikulovich V I③
①(School of Electronics and Information Engineering, Harbin Institute of Technology, Harbin 150001, China) ②(School of Electrical and Electronic Engineering, Harbin University of Science and Technology, Harbin 150080, China) ③(Belarusian State University, Minsk 220030, Belarus)
Abstract：In order to extract effectively the characteristics of each condition vibration signal for rolling bearing, a sensitive Intrinsic Mode Function (IMF) selection algorithm which based on Ensemble Empirical Mode Decomposition (EEMD) is proposed. First, for obtaining the initial characteristics of the vibration signal, the vibration signal is decomposed by using EEMD, and the sensitive components of obtained IMFs are extracted automatically by using kurtosis combined with correlation coefficient. Then, the feature vectors of each condition vibration signal of rolling bearing are obtained by using Singular Value Decomposition (SVD) and AutoRegressive (AR) model. The obtained feature vectors are regarded as the input of the improved hyper-sphere multi-class Support Vector Machine (SVM) for intelligent recognition. Thereby, the condition recognition of normal state, different fault types and different degrees of performance degradation of rolling bearing can be achieved. The experimental results show that, the proposed method can effectively extract fault characteristics information of rolling bearing more than EMD combined with AR model and EMD combined with SVD method, and the recognition rate is higher.
王玉静, 康守强, 张云, 刘学, 姜义成, Mikulovich V I. 基于集合经验模态分解敏感固有模态函数选择算法的滚动轴承状态识别方法[J]. 电子与信息学报, 2014, 36(3): 595-600.
Wang Yu-Jing, Kang Shou-Qiang, Zhang Yun, Liu Xue, Jiang Yi-Cheng, Mikulovich V I. Condition Recognition Method of Rolling Bearing Based on Ensemble Empirical Mode Decomposition Sensitive Intrinsic Mode Function Selection Algorithm. , 2014, 36(3): 595-600.
2014, Vol. 36 
