二维缺陷的定量识别技术研究

Study on quantitative identification technique of two-dimensional subsurface defect

内部缺陷边界的识别方法是传热反问题的重要研究内容,也是红外无损检测技术从定性到定量发展所迫切需要的理论基础。结合对缺陷试件内部传热规律的分析,将缺陷边界的识别算法研究和缺陷的红外检测方法优化相结合提出了多次测量综合方法,从而解决了缺陷边界识别结果中部分边界对初始边界假设比较敏感的问题。数值算例证明了算法的有效性。表面温度测量误差、初始边界假设对识别精度的影响不大。

The research on the of inverse heat transfer rmographic nondestructive test piece with subsurface optimiza method inverse identification of the subsurface defect boundary is an important problem, and is also needed by the quantitative development of the technique. Based on the analysis of the heat transfer mechanism inside defect, this paper combined the identification algorithm study with the tion of thermographic detection technique to develop a multiple measurements combination （MMCM） initial guess of the to de presented method. T shape have negligible solve the problem that some parts of the defect boundary are very sensitive to the feet boundary shape. The numerical experiments certified the effectiveness of the he temperature measurement error and the initial guess of the defect boundary effect on the identification result.