Influence of Parameter Uncertainties on the Computation of Stability Lobe Diagrams

Abstract

To predict stability boundaries of cutting operations models of process forces and dynamic machine behavior are required. Uncertainties of model parameters may lead to significant differences between predicted and measured stability boundaries. To run a lot of simulations with random parameters is a possible approach to take these uncertainties into account when analyzing the stability behavior. However, this is a time consuming task. This paper presents a time efficient method to provide confidence levels of computed stability lobe diagrams based on stability analysis in frequency domain. The required computational time is compared to the effort for an explicit calculation with different stability algorithms. The confidence levels are compared to experimentally determined stability lobe diagrams.