4.7 Article

Development and verification of frequency domain solution methods for rotor-bearing system responses caused by rolling element bearing waviness

期刊

出版社

ACADEMIC PRESS LTD- ELSEVIER SCIENCE LTD
DOI: 10.1016/j.ymssp.2021.108117

关键词

Bearing roundness profile; Experimental verification; Frequency domain solution; Linearized stiffness; Rolling Element Bearing; Waviness

资金

  1. Academy of Finland [313676]
  2. Academy of Finland (AKA) [313676, 313676] Funding Source: Academy of Finland (AKA)

向作者/读者索取更多资源

This study proposed two novel methods, namely Base Excitation Method and Bearing Kinematics Augmented Base Excitation Method, for solving responses caused by bearing waviness excitation. The methods were validated and showed relatively accurate results with reduced computational time compared to previous time domain solutions.
In rotordynamic simulations, rolling element bearing waviness is often accounted using nonlinear models that are solved with a numerical integration scheme in time domain. This approach generates accurate system response, but the method is limited in terms of computational efficiency. This study proposes two novel methods for solution of the responses caused by the bearing waviness excitation in frequency domain, and compares the result with a previously developed, time domain based numerical simulation. The first method known as Base Excitation Method (BEM) considers the waviness as base excitation whereas the second method, known as Bearing Kinematics Augmented Base Excitation Method (BKA-BEM), utilizes a four degree of freedom, quasi-static model to include the bearing kinematics and refine the base excitations due to waviness. The methods are validated with a test case, in which measured low order waviness components of the bearing inner ring roundness profile were used as source for excitation. The accuracy and robustness of the proposed methods in calculating the subcritical harmonic response frequencies and amplitudes are examined for different roundness profiles. The results show that the proposed methods performed relatively well compared to previously developed, time domain solution based numerical model and experimental results. Furthermore, the frequency domain solutions significantly reduce the computational time which makes them easily applicable to simulation-based transfer learning, iterative inverse problems and optimization solutions.

作者

我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。

评论

主要评分

4.7
评分不足

次要评分

新颖性
-
重要性
-
科学严谨性
-
评价这篇论文

推荐

暂无数据
暂无数据