基于QGA优化广义S变换的滚动轴承故障特征提取

Fault feature extraction for rolling bearings based on generalized S transformation optimized with Quantum genetic algorithm

考虑到实际工程环境中噪声对故障特征提取的影响,提出了基于量子遗传算法(QGA)优化广义S变换的滚动轴承故障特征提取方法。该方法以时频分布集中程度为评价标准,首先采用量子遗传算法自适应地选取广义S变换中最优窗口控制参数,然后提取信号变换后复时频矩阵的模向量作为滚动轴承故障特征向量。利用该方法提取的滚动轴承故障特征与其它故障特征进行故障识别对比研究,实验结果表明该方法能够更准确地提取出故障特征,验证了方法的优越性。此外,对不同噪声强度背景下的滚动轴承振动信号进行故障特征提取,诊断结果进一步显示所提方法具有良好的抗噪性和健壮性。

Considering noise,s effects on fault feature extraction of rolling bearings in practical engineering environment, a novel method for rolling bearing fault feature extraction based on the generalized S transformation optimized with the quantum genetic algorithm （ QGA） was proposed. Firstly, the optimal window control parameters of the generalized S transformation were selected adaptively with QGA taking the concentration the level of time-frequency distribution as the evaluation standard. Then the mode vectors of the complex time-frequency matrix formed after fault vibration signals of rolling bearings were transformed with the generalized S transformation were extracted as rolling bearing fault feature vectors. The method was applied to extract rolling bearing fault feature and compared with other methods using fault diagnosis tests. The results showed that the proposed method can extract fault features more accurately than other methods can. Moreover, the fault feature extraction tests of rolling bearing vibration signals under different levels of background noise indicated that the proposed method has a good anti-noise capability and a strong robustness.