The prediction of chatter-free machining parameters is not reliable for industrial applications without guaranteed robustness against modeling uncertainties. Measurement inaccuracies, fitting uncertainties, simplifications and modeling errors typically lead to a mismatch between the mathematical model and the real physical system. This paper presents the robust stability analysis of milling operations based on a pseudospectral approach to take into account the effect of bounded parametric uncertainties both in the cutting coefficients and in the modal parameters. In order to make the predictions more accurate, the operational modal analysis of the spindle during rotation was conducted. The natural frequencies and damping ratios of the dominant vibration modes were identified from the impact tests of the rotating tool at different spindle speeds. The uncertainties of the fitted parameters were included in the computation of the pseudospectral radius of the monodromy operator of the time-periodic system. The solver is tested in a case study and experimental chatter tests are also included for demonstration purposes.

没有颤动加工参数的预测对于工业应用来说是不可靠的，如果没有针对模型不确定性的鲁棒性保证。测量误差，拟合不确定性，简化和建模误差通常会导致数学模型与实际物理系统之间不匹配。本文介绍了一种基于伪光谱方法的铣削操作的鲁棒稳定性分析，以考虑有限的参数不确定性对切削系数和模态参数的影响。为了使预测更准确，进行了旋转过程中主轴的运行模态分析。从不同主轴转速下旋转工具的冲击测试中确定了主要振动模式的固有频率和阻尼比。拟合参数的不确定性包括在时间周期系统的单峰算子的伪谱半径的计算中。案例研究中对求解器进行了测试，还包括实验性颤振测试，以进行演示。