中文摘要：

带式输送机传动滚筒轴承发生故障时,特别是早期故障,其振动信号中隐含的脉冲故障信息很微弱,且常被淹没在强烈的噪音中,直接做频谱分析或包络分析,很难提取其故障特征。最小熵解卷积（Minimum Entropy Deconvolution,MED）通过最优滤波器对轴承微弱故障信号进行最优滤波,提高了信号的信噪比,然后对滤波后的信号进行包络解调分析,能够提取出信号中隐含的故障特征。将该方法应用于带式输送机传动滚筒中的滚动轴承故障诊断,成功提取出了轴承内圈的早期微弱点蚀故障特征。对FIR滤波器阶数L的选择进行了分析,以确保最优的MED解卷积效果。仿真与应用验证了最小熵解卷积方法在滚动轴承故障诊断的有效性和优点。

英文摘要：

When the fault occurred from the rolling element bearing in the belt conveyor driving drum,especially an incipient fault,the impulse-like fault information hidden in vibration signals with strong background noise is usually very weak. It is very difficult to extract the fault features with a frequency spectrum or an envelope analysis. The method for minimum entropy deconvolution（ MED） achieves the optimal filtering for the bearing＇s weak fault signals through a designing optimal filter,and improves the signals to noise ratio（ SNR）,and then the concealed fault feature was extracted from the MED filtered signals with an envelope demodulation analysis. The proposed method was applied to the bearing fault diagnosis in the belt conveyor driving drum,finally the incipient weak pitting fault feature was successfully extracted from the bearing inner race. At the same time,the parameter of FIR filter length L was discussed so as to ensure the optimal deconvolution result with MED. Simulation and application results verify the effectiveness and advantage of the MED method in the fault diagnosis for rolling element bearings.