【摘要】 轴承作为机械传动系统上不可缺少单元，在长期交变载荷作用下容易产生内部损伤，判断其故障状态尤为重要。综合复合多元多尺度加权排列熵（CMMWPE）与天牛须搜索支持向量机（BSASVM）算法优点，设计了一种基于CMMWPE-BSASVM复合算法的轴承故障智能诊断方法。以调心球轴承运行情况为例，开展故障诊断分析。研究结果表明：利用本文CMMWPE算法进行处理时形成了比MWPE、CMWPE算法更加平滑的熵值变化曲线，能够实现以上样本的精确区分，有助于更准确提取出滚动轴承的特征信号。采用CMMWPE可以实现更高效的故障模式识别性能；相比较其他分类器，BSASVM分类器处理效率更高，因此建立的CMMWPE和Isomap特征集识别运行故障获得了100%准确率，确保滚动轴承故障达到精确、高效识别的效果。该研究可以拓宽到相关机械传动领域，具有很好的应用价值。更多还原

【Abstract】 As an indispensable unit of mechanical transmission system,bearing is easy to produce internal damage under long-term alternating load,so it is particularly important to judge its fault state. The composite multivariate multi-scale weighted permutation entropy,CMMWPE and BSASVM algorithm advantages. An intelligent bearing fault diagnosis method based on CMMWPE-BSASVM composite algorithm is designed. Taking the operation of aligning ball bearing as an example,the fault diagnosis analysis is carried out. The research results show that the CMMWPE algorithm in this paper forms a smoother entropy change curve than the MWPE and CMWPE algorithms,which can realize the accurate discrimination of the above samples,and is helpful to extract the characteristic signals of rolling bearings more accurately. CMMWPE can achieve more efficient fault pattern recognition performance. Compared with other classifiers,BSASVM classifier has higher processing efficiency,so the established CMMWPE and Isomap feature sets achieve 100% accuracy in identifying running faults,ensuring that rolling bearing faults can be accurately and efficiently identified. The research can be extended to the field of mechanical transmission and has good application value.更多还原