机构地区: 中南大学信息科学与工程学院
出 处: 《模式识别与人工智能》 2003年第3期323-327,共5页
摘 要: 本文提出了一种新颖的基于Boosting模糊分类的滚动轴承故障诊断方法。利用小波包对滚动轴承的振动加速度信号进行分解,得到滚动轴承动态信号在不同频带的能量,并以此作为滚动轴承的特征向量,采用模糊分类方法进行故障诊断。该方法利用Boosting方法和遗传算法以迭代形式获取一组模糊规则及规则对应的权值,分类器以加权投票方式进行分类决策。对7类滚动轴承进行了实验,结果表明该方法具有很好的故障诊断效果。 In this paper, we present a novel method for fault testing on rolling bearing based on boosting fuzzy classification. First, frequency band is extracted from dynamic vibration signals of rolling bearing gained from acceleration meter using wavelet packets transform. 'Than, boosting fuzzy classification is used to class the faults of rolling bearing. The fuzzy rule base is built using boosting method and Genetic algorithm in an incremental fashion, the Genetic algorithm extracts (me fuzzy classifier rule at a time, and the boosting method reduces the weight of those training instances that are clasified correctly by the new rule, such that the next iteration of the evolutionary algorithm focuses the search on those fuzzy rules that capture the currently misclassified instances. The weight of a fuzzy rule reflects the relative strength the boosting algorithm assigns to the rule class when it aggregates the casted votes. A set of experiments of fault testing on rolling bearing are presented. Experiment results have shown good detective performance of our newly presented method.
关 键 词: 滚动轴承 故障诊断 模糊分类 动态信号 数学模型 模糊分类
领 域: [机械工程]