当前位置: X-MOL 学术Exp. Mech. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Quasi-Gauss Point Digital Image/Volume Correlation: a Simple Approach for Reducing Systematic Errors Due to Undermatched Shape Functions
Experimental Mechanics ( IF 2.0 ) Pub Date : 2020-02-13 , DOI: 10.1007/s11340-020-00588-3
B. Pan , X. Zou

In using subset/subvolume-based digital image/volume correlation (DIC/DVC), proper shape functions must be defined to depict the underlying displacement of the target subsets/subvolumes to be tracked. Because the local deformation in a subset/subvolume cannot be known a priori, mismatched problems (i.e. undermatched or overmatched) unavoidably occur, which induce either remarkable systematic errors for undermatched cases or doubled random errors for overmatched cases in detected displacements. Also, the use of high-order shape functions (e.g. second-order shape functions) would lead to greatly increased computational complexity since more deformation parameters need to be solved. In this work, a novel and simple quasi-Gauss point DIC/DVC method that only uses first-order shape functions is proposed, which can completely avoid or significantly compromise the systematic errors due to undermatched shape functions. Specifically, based on rigorous theoretical analysis, we find that specific positions (designated as quasi-Gauss points) in a subset/subvolume deliver accurate displacement results even when undermatched issues are present. This new finding inspires us to output the displacements at these specific points rather than subset/subvolume center points. Numerical simulations with different deformation modes validate that the proposed approach can effectively reduce or even eliminate the undermatched error in these deformation modes in both DIC and DVC measurements.

中文翻译:

准高斯点数字图像/体积相关性:一种减少因形状函数不匹配导致的系统误差的简单方法

在使用基于子集/子体积的数字图像/体积相关 (DIC/DVC) 时,必须定义适当的形状函数来描述要跟踪的目标子集/子体积的潜在位移。由于子集/子体积中的局部变形无法先验知道,因此不可避免地会发生不匹配问题(即不匹配或过匹配),这会导致不匹配情况下的显着系统误差或检测到的位移中过匹配情况下的双倍随机误差。此外,使用高阶形状函数(例如二阶形状函数)会导致计算复杂度大大增加,因为需要求解更多的变形参数。在这项工作中,提出了一种新颖且简单的准高斯点 DIC/DVC 方法,该方法仅使用一阶形状函数,这可以完全避免或显着降低由于形状函数不匹配而导致的系统误差。具体来说,基于严格的理论分析,我们发现即使存在不匹配的问题,子集/子体积中的特定位置(指定为准高斯点)也能提供准确的位移结果。这一新发现激励我们输出这些特定点处的位移,而不是子集/子体积中心点。不同变形模式的数值模拟验证了所提出的方法可以有效地减少甚至消除这些变形模式在 DIC 和 DVC 测量中的不匹配误差。我们发现,即使存在不匹配的问题,子集/子体积中的特定位置(指定为准高斯点)也能提供准确的位移结果。这一新发现激励我们输出这些特定点处的位移,而不是子集/子体积中心点。不同变形模式的数值模拟验证了所提出的方法可以有效地减少甚至消除这些变形模式在 DIC 和 DVC 测量中的不匹配误差。我们发现,即使存在不匹配的问题,子集/子体积中的特定位置(指定为准高斯点)也能提供准确的位移结果。这一新发现激励我们输出这些特定点处的位移,而不是子集/子体积中心点。不同变形模式的数值模拟验证了所提出的方法可以有效地减少甚至消除这些变形模式在 DIC 和 DVC 测量中的不匹配误差。
更新日期:2020-02-13
down
wechat
bug