抄録
In the advanced manufacturing, it is necessary for the improvement of machining efficiency to select an optimum tool at eachcutting process by tool management. Then, one of the most important issues is to estimate the rest of cutting tool life under a given cutting condition as accurately as possible. In this paper, a structural model based on neural network is proposed to estimate the life and the wear type of cutting tools from their image data and cutting conditions. The input of the model is (1) the states of crater wear and flank wear to cutting tool (e.g. grade of brightness caused by surface roughness, profile), obtained as image data, and (2) the cutting conditions. This system consists of the model based on neural network and image processing device. The output of the system was examined by the experiments under various cutting conditions in turning and the validity of the system was confirmed.
