Prediction of Machining Condition Using Time Series Imaging and Deep Learning in Slot Milling of Titanium Alloy

Low surface quality, undesired geometrical and dimensional tolerances, product damage due to tool wear breakage lead a dramatic increase in production cost. In this regard, monitoring conditions the machining process are crucial prevent unwanted events during guarantee cost-effective high-quality production. This study aims predict critical concerning roughness slot milling of titanium alloy. Using Siemens SINUMERIK Edge Box integrated into CNC machine tool, signals were recorded from main spindle different axes. Instead extraction features signals, Gramian angular field (GAF) was used encode whole signal an image with no loss information. Afterwards, images obtained for training convolutional neural network (CNN) as suitable frequently applied deep learning method images. The combination GAF trained CNN model indicates good performance predicting conditions, particularly case imbalanced dataset. classification resulted recall, precision, accuracy 75%, 88%, 94% values, respectively, prediction workpiece quality breakage.