结合小波包奇异谱熵和SVDD的滚动轴承性能退化评估

发布时间：2018-04-13 17:38

本文选题：滚动轴承 + 小波包奇异谱熵　；参考：《机械科学与技术》2016年12期

【摘要】：针对设备的视情维修,提出一种将小波包奇异谱熵和支持向量数据描述(SVDD)相结合的滚动轴承性能退化评估方法。先提取轴承全寿命周期内振动信号的小波包奇异谱熵作为轴承状态的特征矢量,然后以轴承正常状态下的特征矢量训练SVDD,得到正常状态下的基准超球体,再计算轴承全寿命周期内的特征矢量与基准超球体之间的相对距离,作为性能退化过程的定量评估指标,并对失效阈值和早期故障阈值进行设定。结果表明,与基于小波包和SVDD的性能退化评估方法相比,该方法的早期故障检测能力更强,对轴承性能退化各个阶段的描述更加准确。最后,利用基于EMD的Hilbert包络解调方法对评估结果的正确性进行了验证。
[Abstract]:Aiming at the maintenance of equipment, a new method of rolling bearing performance degradation assessment is proposed, which combines wavelet packet singular spectrum entropy and support vector data description (SVDDD).Firstly, the wavelet packet singular spectrum entropy of the vibration signal in the whole life cycle of the bearing is extracted as the characteristic vector of the bearing state, and then SVDDs are trained with the characteristic vector of the bearing in the normal state to obtain the reference hypersphere in the normal state.Then the relative distance between the characteristic vector and the reference hypersphere in the whole life cycle of the bearing is calculated as the quantitative evaluation index of the performance degradation process and the failure threshold and the early fault threshold are set.The results show that compared with the performance degradation assessment method based on wavelet packet and SVDD, the early fault detection ability of this method is stronger, and the description of each stage of bearing performance degradation is more accurate.Finally, the correctness of the evaluation results is verified by using the Hilbert envelope demodulation method based on EMD.
【作者单位】：华东交通大学机电工程学院;
【基金】：国家自然科学基金项目(51205130) 江西省科协重点活动项目(赣科协字[2014]88号)资助
【分类号】：TH133.33

【相似文献】