Computer Vision Techniques to Develop an Online Tool Life Prediction System on CNC Lathe Machine

Abstract

Tool wear play an important role in cutting industry for higher productivity and a good product quality. Tool wear are selected based on their life criterion which determine their cutting accuracy of the machining, its stability and reliability. This works focuses on developing an algorithm using fuzzy logic and exponential degradation model to predict the life of a tool wear in CNC lathe machine. The image data set was developed by capturing images of tool wear in experimental studies. The algorithm is divided into two parts firstly it will segment the tool part of the tool wear from the image by using fuzzy logic of artificial intelligence and secondly the exponential model is used for predicting the tool life. Currently 100 image data set have been developed to test the system accuracy 90% of image are used to train the system and 10% image are used for testing. The effective development of the tool condition monitoring systems can provide a practical tool to reduce downtime related with tool changes and limits the amount of scrap in metal cutting industry. Implications of the experimental study and recommendations for further research were provided.