基于形态优化滤波和最小二乘支持向量机的轴承故障分析-中国测试.pdfVIP

第43卷第4期 中国测试 Vol.43 No.4 2017年4月 CHINA MEASUREMENT TEST April，2017 doi院10.11857/j.issn.1674-5124.2017.04.023 基于形态优化滤波和最小二乘支持向量机的 轴承故障分析 饶 杰袁 张绍旺袁 徐光荣袁 张 勇 （云南省计量测试技术研究院，云南 昆明 650228） 摘 要院为实现滚动轴承故障分析袁提出基于形态优化滤波和最小二乘支持向量机的轴承故障分析方法遥 首先袁通过 判别指标最大化原则确定最佳形态滤波算子曰然后袁用最佳形态滤波算子对滚动轴承实例故障信号进行降噪分析曰最 后袁利用粒子群对模型参数进行寻优袁利用最小二乘支持向量机建立轴承故障分析模型袁并对模型效果进行评价遥 实 验结果表明院该方法从模型稳定性尧预测准确度尧模型复杂度3个方面考虑袁故障分析结果较优袁能够更好地提取轴承 故障特征信息遥 关键词院滚动轴承曰特征提取曰形态滤波曰最小二乘支持向量机曰故障诊断 文献标志码院A 文章编号院1674-5124渊2017冤04-0110-04 Bearing failure analysis based on improved morphological filtering and least square-support vector machine RAO Jie，ZHANG Shaowang，XU Guangrong，ZHANG Yong （Yunnan Institute of Measuring and Testing Techology，Kunming 650228，China） Abstract: In order to realize rolling bearing failure analysis，a method for diagnosis based on improved morphological filtering and least square-support vector machine is proposed. Firstly， confirm the optimal morphological filter by the maximum principle；then，the method is applied to bearing fault diagnosis by using optimal morphological filter；finally，using PSO-LSSVM to establish the model of bearing fault analysis. From the aspects of model stability，prediction accuracy and model complexity，the method has better experiment results. The results show that the method can effectively extract the bearing fault feature information. Keywords: rolling bearing；feature extraction；morphological filtering；least square-support vector machine；failure diagnosis 0 引 言 过程中发生故障时袁各零件发生碰撞表现为非线性尧 滚动轴承被称为野工业的关节冶袁广泛应用于铁 非稳定性尧复杂性的振动信号袁而机械故障特征往往 路尧汽车尧轮船尧航空航天尧机械等领