基于光顺的网格优化算法及其关键因素分析

Mesh optimization-based smoothing algorithm and its key influence factors analysis

为了探讨单元数、迭代次数以及需求精度等关键因素对于网格优化效果的影响,选用1种较优的四面体单元质量衡量准则建立了网格优化算法的目标函数,在此基础上提出了1种四面体网格质量提高算法,并通过某离心泵蜗壳数值算例对比分析了这些关键因素在网格质量提高和优化耗时方面的异同.结果表明:优化后网格质量在0～0.1区间内的劣质单元数明显减少,同时基于光顺的网格优化算法能够较好地提高网格整体质量,此外,随着单元数的增加,优化后的网格平均质量和优化时间会有所增加,但优化后的最差单元质量出现了波动,存在着1个极值;迭代次数的变化对于优化后的网格质量和优化时间影响较小;需求精度的提高会使得网格质量和优化时间同时增加.

To investigate the influences of element number, iteration number and required accuracy on optimization, a tetrahedron quality metric was chosen to establish objective function to propose a tetrahe-dral mesh optimization algorithm. Based on volute calculating example of a centrifugal pump, the influ-ences of the key factors on the optimized mesh quality and time consuming were compared. The results show that the poor-quality elements in the range from 0 to O. 1 are significantly reduced. The whole mesh quality can be improved obviously by the mesh optimization-based smoothing algorithm. The average mesh quality and the time cost are increased with the increase of element number, but the worst-quality element is fluctuant with one extremum. Iteration number has slight effect on the mesh quality and time cost. The mesh quality is improved with increased time cost by the increasing of required accuracy.