Turbulent flow-induced vibration in nuclear fuel may cause fretting wear of fuel rod at grid support locations. An advanced nuclear fuel for Korean PWR standard nuclear power plants (KSNPs), has been developed to get higher performance comparing to the current fuel considering the safety and economy. One of the significant features of the advanced fuel is the conformal shape in mid grid springs and dimples, which are developed to diminish the fretting wear failure. Long-term hydraulic tests have been performed to evaluate the fretting wear of the fuel rod with the conformal springs and dimples. Wear volume is a measure to predict the fretting wear performance. The shapes of a lot of scars are non-uniform such as wedge shapes, and axially non-symmetric shapes, etc., depending on the contact angle between fuel rod and springs/dimples. In addition, conformal springs and dimples make wear scars wide and thin comparing to conventional ones with convex shape. It is found that wear volumes of these kinds of non-uniform wear scars are over-predicted when the traditionally used wear depth-dependent volume calculation method is employed. In order to predict wear volume more accurately, therefore, the measuring system with high accuracy has been used and verified by the known wear volumes of standard specimens. The wear volumes of the various wear scars have been generated by the measuring system and used for predicting the fretting wear-induced failure time. Based on new evaluation method, it is considered that the fretting wear-induced fuel failure duration with this conformal grid has increased up to 8 times compared to the traditionally used wear depth-dependent volume calculation method.
An Object Recognition and Volume Calculation Method Based on Yolov3 and Depth Vision
