The study presents a procedure for detecting tool failure in drilling of CFRP/Ti6Al4V aeronautical structural components. To accurately detect tool breakage, a method based on the Daubechies 2 Discrete Wavelet Transform (DWT) is proposed to analyze the spindle power consumption signals in the time-frequency domain. The signals are collected from the automatic drilling machine of the industrial system. At each level of decomposition, statistical measures are calculated to extract signal features that are then used as inputs to a classifier algorithm to identify signal patterns or anomalies. The Recall and Precision scores of 1 indicate the effectiveness of the proposed method.
Domínguez-Monferrer, C., Guerra-Sancho, A., Caggiano, A., Nele, L., Miguélez, M.H., Cantero, J.L. (2024). Tool Failure Detection in CFRP/Ti6Al4V hybrid stacks drilling using Discrete Wavelet Transform. In 17th CIRP Conference on Intelligent Computation in Manufacturing Engineering (CIRP ICME ‘23) (pp.406-411) [10.1016/j.procir.2024.08.387].
Tool Failure Detection in CFRP/Ti6Al4V hybrid stacks drilling using Discrete Wavelet Transform
Caggiano, A.;
2024-01-01
Abstract
The study presents a procedure for detecting tool failure in drilling of CFRP/Ti6Al4V aeronautical structural components. To accurately detect tool breakage, a method based on the Daubechies 2 Discrete Wavelet Transform (DWT) is proposed to analyze the spindle power consumption signals in the time-frequency domain. The signals are collected from the automatic drilling machine of the industrial system. At each level of decomposition, statistical measures are calculated to extract signal features that are then used as inputs to a classifier algorithm to identify signal patterns or anomalies. The Recall and Precision scores of 1 indicate the effectiveness of the proposed method.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
