【摘要】 对于滚动轴承而言,工程实际中存在诊断样本与训练样本故障类型相同(如均为滚动体故障)但故障程度却不同的现象,同时滚动轴承发生故障时其振动信号表现出明显的非平稳性,因此文中提出一种基于S变换和改进奇异值分解的滚动轴承故障程度鲁棒的智能诊断方法。首先利用S变换得到滚动轴承振动信号时频分布矩阵,再利用改进奇异值分解方法对时频矩阵进行降维进而得到约简的特征向量,最后将提取到的故障特征向量作为支持向量机的输入,利用支持向量机识别轴承所属的故障类型。实验结果表明,该方法能有效地解决滚动轴承训练样本与测试样本故障程度不一致时的诊断问题,效果优于传统滚动轴承诊断方法。更多还原

【Abstract】 Considering rolling bearing vibration signals exhibiting non-stationary characteristics,an intelligent diagnosis method of rolling bearing to fault degree robust was proposed based on S transformation and improved singular value decomposition( ISVD). In practical engineering,testing samples have the same fault type with training samples( such as ball fault),however the degree of fault may vary,and the proposed approach solved the problems of fault diagnosis in practical engineering. The vibration signals of rolling bearings were firstly conducted S transformation by time-frequency method,the improved singular value decomposition was utilized to extract feature vectors from transformed matrices,and then input to support vector machines( SVMs) as feature vectors. The support vector machine was used to simultaneously judge bearing fault type. The results verify the effectiveness of the proposed approach and this method is better than the results of traditional methods for the practical problems to fault degree robust.更多还原