Monitoring Method of High-speed Tool Wear Level based on Machine Vision

Abstract

In order to improve accuracy and efficiency of high-speed cutting tool wear monitoring and classification, the novel method was proposed to identify the tool wear level and estimate wear capacity based on machine vision and fuzzy statistical learning. Wear feature parameters were extracted by analyzing the high-speed cutting tool wear morphology and wear mechanism. The feature vector of each worn tool was established, and the membership functions were designed, which correspond with low, medium and high wear level. The method represents the probability of belonging to the low, medium and high wear level. Finally, we had experimented in high speed machining shaft on the HTC2550hs high-speed Computer Numerical Control (CNC) turning center in Shenyang Machine Tool Co.,LTD.. Experimental results show that the method can effectively identify the tool wear level, and identification accuracy can get 98.7%. The estimate of wear capacity provides an important basis for tool radius compensation when the wear level is located in the L class (low) level, and allows to replace the tool when the wear level is located at the end of the M class (medium), preventing that the tool enters into the H class (high).