Abstract
This paper proposed an evolutionary fuzzy neural network (EFNN) for tool wear prediction. The material chip is affected by cutting conditions during the cutting process. The different tool wear status causes different chip color which means the color of material chip can be an important factor for tool wear prediction. In this study, an industrial camera is used to capture chip image and convert it into CIE xy chromaticity features through a color conversion model. In addition, to improve the prediction accuracy, a dynamic group cooperative particle swarm optimization (DGCPSO) is proposed to optimize the EFNN parameters. The cutting time and CIE xy value are used as the input of the EFNN, and the output is predicted tool wear value. The experimental results show that the mean absolute percentage error (MAPE) of the proposed EFNN is 2.83% better than other methods.
Tool Wear Prediction Using a Hybrid of Tool Chip Image and Evolutionary Fuzzy Neural Network
