中文摘要：

简述了恒转速条件下滚动轴承故障信号共振解调的优点和基本原理，通过实验和理论分析研究了变转速轴承故障振动信号的特点．指出了轴承损伤点冲击信号的非周期性、轴承共振频率随转速变化的变频特性，以及故障信号的双变频调制特性．建立了由变频转速信号为调制信号、变频共振衰减信号为载波的滚动轴承故障模型．给出了阶比循环平稳自相关函数的计算方法．利用循环平稳分析对旋转机械振动信号的解调功能，结合连续隐马尔可夫模型（CHMM）对动态信号的识别能力，提出了一种适用于变转速运转条件下的滚动轴承故障诊断方法，通过实验验证了方法的可行性．

英文摘要：

Based on the brief introduction to the advantage and the basic principle of the resonance demodulation method for rolling bearing fault diagnosis at constant rotational speed, the characteristics of the bearing fault vibration signals at varying rotational speed are studied by experiments and theoretical analyses. A scheme is designed to reveal the non-periodic feature of the bearing damage point impact signal, the frequency changeability of the bearing resonance frequency along with the bearing rotational speed changes, and the dual frequency modulation characteristics of bearing fault signal A rolling bearing fault model is established, in which the changing frequency rotational speed signal is used as modulating signal and the attenuation resonance frequency signal as carrier wave, and then an algorithm formula of order cyclostationary autocorrelation is given. Using the excellent demodulation feature of cyclostationary analysis to rotating machinery vibration signal, and with the recognition ability of continuous hidden Markov model （CHMM） for dynamic signal, a fault diagnosis method for rolling bearing under the condition of variable-speed operation is proposed. Finally, the feasibility of the method is verified by tests.