基于GA随机共振的模糊SVDD轴承性能退化评估  

Performance Degradation Quantitative Assessment Method for Bearings Based on GA Stochastic Resonance and Fuzzy SVDD

作　　者：高淑娟[1] GAO Shu-juan(School of Mechatronics Engineering, Zibo Vocational Institute, Zibo Shandong 255013, China)

机构地区：[1]淄博职业学院机电工程学院

出　　处：《组合机床与自动化加工技术》2019年第8期53-58,共6页Modular Machine Tool & Automatic Manufacturing Technique

基　　金：国家自然科学基金(51506221)

摘　　要：提出了一种基于遗传算法随机共振与模糊支持向量数据描述相结合的轴承性能退化评估方法。首先结合遗传算法,应用稳定约束条件下的自适应随机共振方法对轴承故障微弱信号进行提取,在保证收敛性的条件下得到所需的特征集;然后从得到的特征集中选取多个特征作为模糊支持向量数据描述的输入,利用训练样本得到超球体模型,进一步提出一个轴承退化综合性能评估指标。最后利用不同方法对实验数据进行评估对比验证了该方法能够更早的检测出轴承的性能退化时刻,并且能够更加形象的描述轴承性能退化过程。A performance degradation degree quantitative assessment method for rolling bearings was proposed, which integrates the methods of genetic algorithm stochastic resonance and fuzzy Support Vector Data Description(SVDD). Firstly, the stability constrain adaptive stochastic resonance was applied to extract weak bearing fault signal combined with genetic algorithm, by which the required character set was obtained in the convergence conditions. Then the multiple characteristic signal regarded as the input of fuzzy SVDD was selected from obtained character set, and a comprehensive performance evaluation index of bearing degradation was proposed. Finally, the conclusion that the proposed method could detect initial degradation point earlier and more vividly describe the performance degradation process was verified by comparison of different methods to evaluate the experimental data.

关 键 词：轴承 遗传算法 随机共振 支持向量数据描述 性能退化评估 

分 类 号：TH162[机械工程—机械制造及自动化] TG506[金属学及工艺—金属切削加工及机床]