改进极限学习机在滚动轴承振动故障诊断中的应用

发布时间：2018-03-29 10:40

本文选题：滚动轴承　切入点：振动故障诊断　出处：《机械设计与制造》2016年01期

【摘要】：滚动轴承广泛应用于机械、铁路运输、航天航空等领域,在旋转机械设备的正常稳定运行中占据着至关重要的位置,其振动故障的准确、快速诊断是保证机械设备正常稳定运行的必要手段,因此,研究一种能够有效诊断滚动轴承振动故障的方法具有重要意义。针对滚动轴承振动信号具有非平稳性、非线性和影响因素相互影响相互作用的特点,结合小波分析理论对滚动轴承振动信号进行故障特征提取,通过借鉴支持向量机的分类思想,从最优化角度出发,引入结构风险最小化原理对极限学习机进行改进及优化,并结合改进后的方法构建了滚动轴承振动故障诊断模型。仿真结果表明,改进的极限学习机进一步提高了滚动轴承振动故障诊断的效率和分类准确率,为滚动轴承的振动故障诊断提供了新思路和新方法。
[Abstract]:Rolling bearings are widely used in machinery, railway transportation, aerospace and other fields. They occupy a crucial position in the normal and stable operation of rotating machinery and equipment, and their vibration faults are accurate. Rapid diagnosis is a necessary means to ensure the normal and stable operation of mechanical equipment. Therefore, it is of great significance to study a method that can effectively diagnose the vibration faults of rolling bearings. Based on the characteristics of the interaction between nonlinear and influencing factors, the fault feature extraction of rolling bearing vibration signal is carried out based on wavelet analysis theory. By using the classification idea of support vector machine for reference, we proceed from the angle of optimization. The structural risk minimization principle is introduced to improve and optimize the ultimate learning machine. In combination with the improved method, a fault diagnosis model for rolling bearing vibration is constructed. The simulation results show that, The improved extreme learning machine further improves the efficiency and classification accuracy of vibration fault diagnosis of rolling bearings, and provides a new idea and method for vibration fault diagnosis of rolling bearings.
【作者单位】：广西职业技术学院机械与汽车技术系;广西水利电力职业技术学院机电工程系;
【基金】：广西高校科研资助项目(YB2014485)
【分类号】：TH165.3;TH133.33

【相似文献】