乏燃料组件典型尺寸测量对于保障压水反应堆的运行安全具有重要意义, 针对乏燃料组件典型尺寸测量任务, 本文介绍了一种基于退化图像的燃料组件典型尺寸测量方法。该方法中首先对图像进行前处理, 包括利用改进的中值滤波去除图像辐射噪声、基于多帧图像及维纳滤波去除图像热扰动以及基于Radon变换对燃料棒进行倾斜校正; 然后利用标尺对图像的像素尺度进行标定, 利用基于灰度投影的方法对燃料棒边缘及格架边缘进行提取, 并利用模版匹配的方法对燃料组件上下端塞进行匹配定位; 最后根据所标定的像素尺度以及特征检测得到的像素尺寸计算得到燃料组件典型尺寸如燃料棒直径、格架宽度及上、下端塞距离。随后对所提方法进行了理论精度分析, 分析结果表明本方法测量精度与像素尺度及边缘检测误差呈线性相关。最后进行了精度验证试验, 试验结果表明燃料棒直径测量精度小于±0. 05mm, 端塞距离测量精度小于±0. 3mm, 格架宽度测量精度小于±0. 5mm, 具有一定的使用价值。
The measurement of typical dimensions of spent fuel assemblies is of great significance for ensuring the safe operation of pressurized water reactors. In response to the task of measuring typical dimensions of spent fuel assemblies, this article introduces a method for measuring typical dimensions of fuel assemblies based on degraded images. In this method, the image is first preprocessed, including using an improved median filter to remove image radiation noise, using multi frame. images and Wiener filtering to remove image thermal disturbances, and using Radon transform. to correct fuel rod tilt; Then, the pixel scale of the image is calibrated using a ruler, and the edges of the fuel rod and the frame. are extracted using a grayscale projection based method. The upper and lower plugs of the fuel assembly are matched and located using a template matching method; Finally, based on the calibrated pixel scale and the pixel size obtained from feature detection, the typical dimensions of the fuel assembly, such as fuel rod diameter, grid width, and upper and lower plug distances, are calculated. Subsequently, theoretical accuracy analysis was conducted on the proposed method, and the results showed that the measurement accuracy of this method is linearly correlated with pixel scale and edge detection error. Finally, precision verification experiments were conducted, and the results showed that the measurement accuracy of fuel rod diameter was less than ± 0. 05mm, the measurement accuracy of end plug distance was less than ±0. 3mm, and the measurement accuracy of grid width was less than ±0. 5mm, which has certain practical value.