An Assessment of Quantitative Predictions of Deterministic Mixed Lubrication Solvers
An Assessment of Quantitative Predictions of Deterministic Mixed Lubrication Solvers
复制标题
DOI：
10.1115/1.4047586
复制
发表时间：
2020-07
期刊：
影响因子：
--
通讯作者：
Yuechan Wang;Abdel Dorgham;Y. Liu;Chun Wang;M. Wilson;A. Neville;Abdullah Azam
中科院分区：
文献类型：
--
作者：
Yuechan Wang;Abdel Dorgham;Y. Liu;Chun Wang;M. Wilson;A. Neville;Abdullah Azam
模拟混合润滑问题的能力有了很大提高，特别是在集中润滑的接触中。提出了一种基于半系统方法的混合润滑仿真方法，该方法对提高混合润滑仿真的稳定性和稳健性具有重要意义。然后，通过改变雷诺方程中Couette流项的离散格式、改变密度导数的计算方法和改变系数矩阵中项的贡献，得到了模型的不同变体。结果表明，压力分布、油膜厚度分布、波长比、重合度和计算时间与求解方案的选择有很大关系。这表明，混合润滑解算器的输出可以很容易地用于定性和参数研究，但在进行定量预测时应小心。
The ability to simulate mixed lubrication problems has greatly improved, especially in concentrated lubricated contacts. A mixed lubrication simulation method was developed by utilizing the semi-system approach which has been proven to be highly useful for improving stability and robustness of mixed lubrication simulations. Then different variants of the model were developed by varying the discretization schemes used to treat the Couette flow terms in the Reynolds equation, varying the evaluation of density derivatives and varying the contribution of terms in the coefficient matrix. The resulting pressure distribution, film thickness distribution, lambda ratio, contact ratio, and the computation time were compared and found to be strongly influenced by the choice of solution scheme. This indicates that the output from mixed lubrication solvers can be readily used for qualitative and parametric studies, but care should be taken when making quantitative predictions.