论文部分内容阅读
在高温温度场重建过程中,仍然需要通过温度场的单点温度值来构建整个温度场的温度分布情况。以焊接温度场为例,通过反距离加权法、克里格法和样条函数法三种插值方法,分别对其空间变异和布局进行了分析和重建。结果表明:不同插值方法对预测精度影响不显著,而采样点数量则显著影响了温度场空间分布的重建精度。在温度场重建过程中,25个采样点进行重建是比较适宜的采样数量。将得出的结论与最佳采样公式进行比较,发现使用公式计算的最佳采样数量相对偏低,说明不考虑采样点实际的空间变异情况,仅使用最佳公式得到的采样数量进行温度场的重建会导致重建结果的不准确。
In the process of reconstruction of high-temperature temperature field, the temperature distribution of the whole temperature field still needs to be constructed based on the single-point temperature value of the temperature field. Taking the welding temperature field as an example, three kinds of interpolation methods: inverse distance weighting, Kriging and spline function, were respectively used to analyze and reconstruct their spatial variability and layout. The results show that different interpolation methods have no significant effect on the prediction accuracy, while the number of sampling points significantly affects the reconstruction accuracy of the spatial distribution of the temperature field. In the temperature field reconstruction process, 25 sampling points for reconstruction is more appropriate sampling number. Comparing the conclusion with the best sampling formula, we find that the optimal sampling quantity calculated by using the formula is relatively low, which means that regardless of the actual spatial variation of the sampling point, only the sampling quantity obtained by the best formula is used for the temperature field Reconstruction will result in inaccurate reconstruction results.