• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文期刊>Measurement Science & Technology >Radial basis function interpolation of unstructured, three-dimensional, volumetric particle tracking velocimetry data
【24h】

Radial basis function interpolation of unstructured, three-dimensional, volumetric particle tracking velocimetry data

机译：径向基函数插值的非结构化三维体积粒子跟踪测速数据

获取原文
获取原文并翻译 | 示例
       
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

Unstructured three-dimensional fluid velocity data were interpolated using Gaussian radial basis function (RBF) interpolation. Data were generated to imitate the spatial resolution and experimental uncertainty of a typical implementation of defocusing digital particle image velocimetry. The velocity field associated with a steadily rotating infinite plate was simulated to provide a bounded, fully three-dimensional analytical solution of the Navier-Stokes equations, allowing for robust analysis of the interpolation accuracy. The spatial resolution of the data (i.e. particle density) and the number of RBFs were varied in order to assess the requirements for accurate interpolation. Interpolation constraints, including boundary conditions and continuity, were included in the error metric used for the least-squares minimization that determines the interpolation parameters to explore methods for improving RBF interpolation results. Even spacing and logarithmic spacing of RBF locations were also investigated. Interpolation accuracy was assessed using the velocity field, divergence of the velocity field, and viscous torque on the rotating boundary. The results suggest that for the present implementation, RBF spacing of 0.28 times the boundary layer thickness is sufficient for accurate interpolation, though theoretical error analysis suggests that improved RBF positioning may yield more accurate results. All RBF interpolation results were compared to standard Gaussian weighting and Taylor expansion interpolation methods. Results showed that RBF interpolation improves interpolation results compared to the Taylor expansion method by 60% to 90% based on the average squared velocity error and provides comparable velocity results to Gaussian weighted interpolation in terms of velocity error. RMS accuracy of the flow field divergence was one to two orders of magnitude better for the RBF interpolation compared to the other two methods. RBF interpolation that was applied to vortex identification in experimental data showed reduced noise and reliable calculation of vortex ring geometry.
机译：使用高斯径向基函数（RBF）插值对非结构化三维流体速度数据进行插值。生成数据来模拟散焦数字粒子图像测速的典型实现的空间分辨率和实验不确定性。模拟了与稳定旋转的无限板有关的速度场，以提供Navier-Stokes方程的有界，全三维解析解，从而可以对插值精度进行可靠的分析。数据的空间分辨率（即粒子密度）和RBF的数量是变化的，以便评估对精确插值的要求。插值约束（包括边界条件和连续性）已包含在用于最小二乘最小化的误差度量中，该误差度量确定插值参数，以探索改善RBF插值结果的方法。还研究了RBF位置的均匀间距和对数间距。使用速度场，速度场的散度和旋转边界上的粘性扭矩评估插值精度。结果表明，对于本实施方式，RBF间距为边界层厚度的0.28倍足以进行精确插值，尽管理论误差分析表明，改进的RBF定位可能会产生更准确的结果。将所有RBF插值结果与标准高斯加权和泰勒展开插值方法进行比较。结果表明，基于平均速度误差平方根，RBF插值法与泰勒展开法相比，将插值结果提高了60％至90％，并且就速度误差而言，该结果可与高斯加权插值法相媲美。与其他两种方法相比，RBF插值法的流场发散的RMS精度要好一到两个数量级。在实验数据中应用于涡流识别的RBF插值显示出降低的噪声和可靠的涡流环几何形状计算。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱：kefu@zhangqiaokeyan.com

京公网安备：11010802029741号 ICP备案号：京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有

  • 客服微信

  • 服务号