Optimisation of variable helix end millings tools by minimising self excited vibration

During machining, self-excited vibrations known as regenerative chatter can occur. This instability can be avoided by modifying the tool geometry, thereby influencing the time delay terms that arise in the governing equations. The present study uses Differential Evolution (DE) to optimise the tool helix geometry, so as to avoid chatter. The results are compared to those from Sequential Quadratic Programming (SQP). It is shown that the DE approach can significantly increase the chatter stability, and substantially out-performs the SQP algorithm. The performance of the DE approach is due to its ability to perform global optimisation in the presence of significant nonlinearities.