Int J Performability Eng ›› 2019, Vol. 15 ›› Issue (5): 1334-1342.doi: 10.23940/ijpe.19.05.p9.13341342

Previous Articles     Next Articles

Degradation Index Extraction and Degradation Trend Prediction for Rolling Bearing

Xin Zhanga,*, Jianmin Zhaoa, Xianglong Nib, Haiping Lia, and Fucheng Sunb   

  1. a Army Engineering University, Shijiazhuang, 050003, China
    b Electronic Equipment Test Centre of China, Luoyang, 471003, China
  • Submitted on ;
  • Contact: * E-mail address:
  • Supported by:
    This research was partially supported by the National Natural Science Foundation (No. 71871220).

Abstract: In the degradation process of the rolling bearing, the traditional feature trends appear steady in the early stage and then show a sudden change trend in the later stage. For this reason, it is difficult to predict the bearing degradation trend accurately. In order to solve this problem, this paper puts forward a new degradation index extraction method based on dual-tree complex wavelet transform (DTCWT) and isometric feature mapping (ISOMAP). Compared with the traditional characteristic parameters, the new degradation index can better reflect the degradation tendency of the bearing. Considering the data of different time points has different contributions to degradation trend prediction, the improved BP neural network is applied to predict the degradation trend of bearing. The method is verified by using the bearing degradation data.

Key words: rolling bearing, feature extraction, neural network, degradation, prediction