On-machine measurement and quantitative evaluation of belt wear in robotic grinding using line structured light scanning

Due to the favorable characteristics of flexible contact and low grinding temperature, belt grinding has been widely used for the precision machining of difficult-to-machine materials such as superalloys and titanium alloys. However, belt wear is inevitable during the machining process, posing significant challenges to grinding process planning and quality control. Currently, there is still a lack of an efficient, accurate, and widely applicable method for monitoring belt wear. To address this shortcoming, this study proposes an on-machine measurement and quantitative evaluation method of belt wear in robotic grinding using line structured light scanning. Firstly, a line laser scanning module with integrated dust-removal and dust-proof functions is designed to mitigate the impact of dust contamination on measurement accuracy. Then, this study defines a wear coefficient based on the geometric characteristics and volume variations of agglomerate. Based on this, an evaluation method for belt wear using point cloud processing is proposed, and a wear matrix is employed to better characterize the wear distribution across different regions of the belt. Finally, the full-lifecycle wear experiment of the belt is conducted on a robotic grinding platform. The experimental results demonstrate that, within a 95 % confidence interval, the measurement deviation between the proposed method and the optical profilometer-based measurement method is within 10 %. Additionally, this study investigates the temporal evolution and spatial distribution patterns of belt wear and further explores its effect on grinding performance.