飞轮振源特征参数提取方法及其分布特性

为了从飞轮微振动输出中提取飞轮转子的振源信息，提出利用鲁棒回归辨识飞轮在升速过程中各主要阶次的振幅对转速的二次幂系数，并将其作为飞轮的振源幅值特征参数；为探究参数的分布特性，利用Weibull分布拟合多个飞轮在同一阶次的特征参数，并计算出特定概率点的分位数。振源特征参数的分位数既系统地展示了飞轮振源分布情况，亦可作为振源分级判据。将该方法成功应用于50 Nms飞轮的振源特征参数分布研究中，为该型飞轮的超静性能筛选和设计改进提供了判定准则和参考依据。

Extraction Method and Distribution of Characteristic Parameters of Flywheel Vibration Source

The broadband micro-vibration associating with the high-speed rotation of the rotor in a flywheel is mainly derived from the centrifugal force and couple generated by the unbalance of the rotor and preload force fluctuation of the bearing assembly. A large structural resonance may be excited by the sources when coupled with the whirling motion or the structural modal of the flywheel. To extract the vibration sources information of a rotor from the micro vibration outputted by the flywheel, the two power coefficients of main orders vibration amplitudes with rotational speed are identified by the robust regression while speeding up. They are used as the characteristic parameters of the flywheel vibration sources. To study the parameters distribution, the quantiles at some given probabilities are calculated based on the results of the Weibull distribution that is used to fit the characteristic parameters of the flywheels at the same order. They show the distributions of the flywheel vibration sources integrally. They can also be used as the classification criteria for vibration sources. This method has been successfully applied to study the characteristic parameters distribution of a 50 Nms flywheel. The results provide the grading criteria for its stabilization screening and the references for the improved design of the super static flywheel.