Monitoring Machining Processes Based on Discrete Wavelet Transform and Statistical Process Control

This paper presents a new method to monitor machining processes based on a combination of discrete wavelet transform (DWT) and statistical process control (SPC), called a multi-scale statistical approach. First, DWT is applied to decompose the sensor signal onto different scales. Next, the detection limits are formed for each decomposed signal components, called the sub-signals, using Shewhart control charts. Finally, by inverse wavelet transform of the threshold crossing points of the sub-signals, malfunctions can be detected. Based on a test on the tool condition monitoring in turning using acoustic emission (AE) signal, it is shown that the new method is effective and robust.