为了提高涡流轴绕式探头(Bobbin)定量检测的准确性, 需要明确缺陷定量检测影响因素的主次关系, 在标定管制作等方面开展针对性的工作, 采用权重分析对此类影响因素主次关系进行分析。以裂纹深度为因变量, 以缺陷特性参数为自变量, 采用最小二乘回归和岭回归分析对其定量关系进行了分析, 并在此基础上获取了权重因子。

This work aims at the improvement of quantitative Bobbin eddy current inspection for the heat exchange tubes of steam generator (SG) in nuclear power plant. The accuracy of quantitative Bobbin inspection of heat transfer tubes in SG is affected by various of factors, such as the consistence of the material properties of the tubes; the characteristics of the defects; the types of defects in between the calibration tubes and actual tubes one; the property of testing equipment, the environmental features and so on. Previous researches were mostly concerned about the factors like characteristics of defects (such as length of defect, volume of defect and angle between defect and tube axial), and they were thought as the main factors, which may influence the quantitative accurancy. In order to control these factors for an improved tube Bobbin inspection, primary and secondary of these influencing factors should be clear. The weight factor analysis as a useful method is mainly applied. With crack depth as the dependent variable, least squares and ridge regression methods are quantitatively analyzed for the relationship between the influence factors and the variable. On this basis, the weights of these factors are obtained.