A study on tool wear monitoring using time-frequency transformation techniques

It is in a high demand to automatically monitor and diagnose tool wear, tool fault, or tool damage during machining process to increase efficiency and product quality and reduce production cost. This paper investigates an online tool condition monitoring method using acoustic emission signal in milling operation. The flank wear (VB) is investigated as the system fault. The nature of faulty signals in tool condition monitoring (TCM) is time varying. Therefore time-frequency transformation is an ideal analysis tool for signal interpretation. Short-time Fourier transform (STFT), Wavelet transform, S-transform, the smoothed pseudo-Wigner-Ville distribution and the Choi-Williams distribution are used for signal decomposition and two-dimensional (2D) principal component analysis (PCA) is implemented for dimensionality reduction. A 2D correlation analysis represents the deviation of the faulty signals from the healthy signal and a curve fitting approach is used to find the tool fault. Experimental tests are used for validation and the efficiency of each time-frequency transformation method in the designed TCM system is evaluated and compared.