Tool Wear Monitoring Using Improved Dragonfly Optimization Algorithm and Deep Belief Network

In recent decades, tool wear monitoring has played a crucial role in the improvement of industrial production quality and efficiency. machining process, it is important to predict both cost life, reduce equipment downtime. The conventional methods need enormous quantities human resources expert skills achieve precise information. To automatically identify types, deep learning models are extensively used existing studies. this manuscript, new model proposed for effective classification serviceable worn cutting edges. Initially, dataset chosen experimental analysis that includes 254 images edge profile heads; then, circular Hough transform, canny detector, standard transform segment 577 images, where 276 disposable 301 functional. Furthermore, feature extraction carried out on segmented utilizing Local Binary Pattern (LBPs) Speeded up Robust Features (SURF), Harris Corner Detection (HCD), Histogram Oriented Gradients (HOG), Grey-Level Co-occurrence Matrix (GLCM) descriptors extracting texture vectors. Next, dimension extracted features reduced by an Improved Dragonfly Optimization Algorithm (IDOA) lowers computational complexity running time Deep Belief Network (DBN), while classifying evaluations showed IDOA-DBN attained 98.83% accuracy patch configuration full division, which superior models.