Application of singular spectrum analysis to tool wear detection using sound signals

Abstract

The aim of the present work is to study the applicability of singular spectrum analysis (SSA) to the processing of the sound signal from the cutting zone during a turning process, in order to extract information correlated with the state of the tool. SSA is a novel non-parametric technique of time series analysis that decomposes a given time series into an additive set of independent time series. The correspondence between the singular spectrum obtained using SSA and the frequency spectrum of the signal is the basis of this processing technique. Finally, some of the features extracted from the SSA-processed sound signal were presented to a feedforward back-propagation (FFBP) neural network to determine the tool flank wear. The results showed that the proposed processing technique is well suited to the task of signal processing in the area of tool condition monitoring (TCM).

Using spindle noise to monitor tool wear in a turning process

“…4. Many of the previous studies in this area use the sound signals present in the frequency range above machine background noise to develop the tool wear monitoring system [6][7][8][9][10][11][12][13][14][15]. However, as there is little reporting of using spindle noise in tool wear monitoring, this present work concentrates on the spindle noise signal as a signal of interest to extract a promising feature for tool condition monitoring.…”

“…This processing technique was used to extract the sound features from the cutting zone during the turning process. The results showed that the extracted features from sound and feed motor current signals correlate with tool wear state [14]. Raja used sound signal analysis, using the Hilbert-Huang Transform (HHT), to monitor flank wear [8,15].…”

“…Alonso et al used sound and feed motor current to develop the monitoring system to detect the tool wear. The features extracted from sound and feed motor current signals by SSA correlate with tool wear state and were trained by neural network to estimate the tool flank wear [14]. A radial basis function was used by Manan to train the network which consists of features extracted from sound signal and surface roughness [10].This classification model is able to recognise tool wear states which classify tool states (defined as 'sharp', 'semi-dull' and 'dull'), and the proposed system can be utilised to effectively monitor the condition of the cutting tool.…”

Tool Flank Wear Estimation using Emitted Sound Signal Analysis by PCA – SER Based Peak to Peak Measurements

“…The acquisition of process variable values like power used for cutting, amount of heat vibration, current for spindle motor current, surface roughness and their correlation to the tool wear were considered to monitor the tool wear indirectly considered. Researches like Alonso et.al [1] suggested a mechanism to predict the degrees of tool flank wear by artificial neural network that works on the sound signals emitted at the time of conversion procedure by considering feed cutting force. Sadettin et.al [2] establish the vibration of amplitude is straightforwardly to the intensification tool wear.…”

The milling tool wear monitoring using the acoustic spectrum

“…asperity height, size and distributions. Salgado and Alonso [11,12] presented a tool condition monitoring system (TCMS) for on-line tool wear monitoring using the feed motor current and the sound signal. The results showed that the proposed TCMS is fast and reliable for tool condition monitoring.…”