Dimensionality Reduction of Sensorial Features by Principal Component Analysis for ANN Machine Learning in Tool Condition Monitoring of CFRP Drilling

Abstract

With the aim to perform sensor monitoring of tool conditions in drilling of stacks made of two carbon fiber reinforced plastic (CFRP) laminates, a machine learning procedure based on the acquisition and processing of thrust force, torque, acoustic emission and vibration sensor signals during drilling is developed. From the acquired sensor signals, multiple sensorial features are extracted to feed artificial neural network-based machine learning paradigms, and an advanced feature extraction methodology based on Principal Component Analysis (PCA) is implemented to decrease the dimensionality of sensorial features via linear projection of the original features into a new space. By feeding artificial neural networks with the PCA features, the diagnosis of tool flank wear is accurately carried out.