Online damage detection of cutting tools using Dirichlet process mixture models

The ability to monitor and predict tool deterioration during machining is an important goal because the state of wear has a significant influence on surface quality machined components. To build up comprehensive condition monitoring system for diagnosis prognosis, however, extensive measurements knowledge required. Collecting labelled datasets that include damage information this purpose can be expensive time consuming. This paper suggests unsupervised clustering approach using Dirichlet process mixture models detect change in characteristics cutting online diagnosis. As well as providing useful tool, potential reduce need exhaustive associated required prognosis. model suited erratic unpredictable nature progression, number clusters determine possible states are not set a-priori. Consequently, method equipped handle variations across homogeneous heterogeneous groups material compositions. proposed demonstrated here trials characterisation new tools. In example shown, results indicate would result around 30% reduction test times (on average) outer diameter turning case hardened steel, 10 Polycrystalline cubic Boron Nitride tools from two different