Abstract: The non-linear correlation and probability density function of each intrinsic mode function after the set ensemble empirical mode decomposition (EEMD) are not fully considered in the traditional correlation coefficient method, an improved EEMD method is proposed, which is combined with fuzzy theory to evaluate the performance evolution degree of rolling bearings during service. The effectiveness of the proposed method is verified by experiments.The research results of the case show that the performance evolution coefficients of rolling bearings have obvious monotonic decreasing trends in the service process, and the minimum evolution coefficient is less than 0.5, which indicates that the vibration performance of the bearings have undergone serious evolution.

Key words: Rolling bearing; Performance evolution; EEMD method; Fuzzy theory

摘要： 针对传统相关系数法没有充分考虑集合经验模态分解（EEMD）后的各内禀模式函数 与原数据的非线性相关性及概率密度函数问题，提出一种改进的 EEMD 法，结合模糊理论对 滚动轴承服役过程中的性能演变程度进行评估，通过实验验证所提方法的有效性。案例研究 结果表明，滚动轴承服役过程中性能演变系数有明显的单调减小趋势，且最小演变系数值小 于 0.5，表明轴承振动性能已经发生严重演变。

关键词: 滚动轴承；性能演变；EEMD 法；模糊理论