【摘要】 涡轮叶片的内腔多采用陶瓷型芯铸造而成,在其内部的空气通道上如果残留有型芯材料多余物,将会影响叶片的冷却效果。为了对叶片进行检测,提出了一种基于工业CT(computed tomography,以下简称CT)图像局部控制核的多余物检测方法。该方法采用标准叶片和待测叶片的CT图像,首先分别计算两幅图像的局部控制核,然后通过计算对应控制核的相似性来判断待测叶片中是否存在多余物。实验结果表明,该方法能够克服标准CT图像和待测CT图像存在配准误差的影响,能够有效检测出涡轮叶片内腔多余物,并且对噪声不敏感,具有良好的鲁棒性。更多还原

【Abstract】 The internal cavity of turbine blade was usually casted by ceramic core.If there was some residual core material in the cooling channels,it would decrease the blade’s cooling effect.In order to ensure the quality of the blade,a new approach for detecting remainders based on local steering kernel(LSK) of industrial computed tomography(CT) images is proposed.Firstly the LSKs from both the CT images of standard blade and tested blade at all pixel locations were calculated.Then the comparison between the LSKs was carried out to determine whether there were some remainders in the tested blade.The results show that,this method can overcome the influence of the two images’ registration errors.It can effectively detect remainders in the internal cooling channels of the turbine blade,and has a good robustness to noise.更多还原