关键词: 排序模式分析, 信息散度, 状态监测, 滚动轴承

Abstract: Ordinal pattern analysis might map one-dimensional vibration time series into probability distribution of ordinal patterns in the high-dimensional phase space, and then reveal the internal tiny variations in ordinal pattern structures. It provided a new research view for the weak vibration signal feature extraction. A new monitoring indicator named ordinal information divergence was proposed herein based on ordinal pattern analysis and information divergence, to quantitatively describe the ordinal pattern distribution difference of vibration signals in the high-dimensional phase space between the current status and the reference health status. Two experiments, including the damage degree assessment of bearing inner race and the run-to-failure bearing degradation trend analysis, were used to validate the effectiveness of the proposed new indicator. The comparative studies were performed with the traditional statistics and several existing nonlinear indicators including wavelet entropy, approximate entropy and permutation entropy. Experimental results demonstrate that the proposed new indicator presents better quantitative ability for different damage degrees and is more sensitive to the incipient fault and more stable and efficient in computation, thus easy to implement in engineering applications.

Key words:  ordinal pattern analysis, information divergence, condition monitoring, rolling bearing