Tool Condition Monitoring in Metal Cutting

Abstract

Automatic tool condition monitoring is based on the measurements of physical phenomena which are correlated with the tool wear, and thus can be exploited as the tool wear symptoms. However measured quantities depend not only on the tool wear but also on a variety of other process parameters of random nature, making the relationship between tool wear and measured value very complex which has a statistical rather than strict, predictable nature. Therefore, the development of a robust and reliable tool condition monitoring system requires a combination of different, meaningful signal features, which best describe the tool wear. There are numerous signal features (SFs) that can be extracted from time domain, frequency domain or time-frequency domain signal. As it is really not possible to predict which signal features will be useful in a particular case thus these informative, correlated with tool wear, should be automatically selected. The information extracted from one or several sensors’ signals has to be combined into one tool condition estimation. This can be achieved by various artificial intelligence methods.