Analysis of different machine learning algorithms to learn stability lobe diagrams

Chatter is a limiting factor for productivity in milling. Choosing cutting parameters that ensure a stable and productive process is not a trivial task. Stability lobe diagrams (SLD) help to find suitable parameters for machining. This paper examines the suitability of support vector machines (SVM) and artificial neuronal networks (ANN) for this application. In addition, kernel interpolation as a new algorithm for this approach is introduced. The algorithms are tested on simulated as well as on measurement data from a real process. It is shown that ML algorithms are able to learn SLDs during process.