重型车铣复合机床动力学参数（模态参数）具有不确定性，传统模态分析方法辨识的确定性结果无法满足实际分析需要。针对该问题，申请者以燃气轮机机匣车铣复合机床结构主轴部位为具体对象，拟展开如下研究：首先，利用敲击模态实验与空运行自激励模态实验，研究非线性、变工况因素下主轴部位局部模态参数分布规律，建立不确定机制下模态参数的概率表征；其次，基于贝叶斯理论思想，研究不确定机制下模态参数在线辨识模型；针对运行状态下重型机床结构模态参数的时变性与不确定性，引入工况状态向量，建立不确定工况区间分析模型，在多维度工况区间下实现变工况非平稳振动信号的离散与模态参数辨识；最后，利用燃气轮机机匣加工过程中的长期振动大数据，分析重型车铣复合机床结构主轴部位模态参数的长期劣化规律，从而验证本项目不确定理论与方法的有效性。本项目的研究成果可以为重型车铣复合机床结构动力学相关的健康状态监测与主动维护奠定理论与方法基础。

With the improvement of machining precision, the dynamic characteristics of the heavy duty milling compound machining center has become an important index to reflect the machining performance. Heavy duty turning-milling machining center has the characteristics of large stroke and multi process, which leads to structure nonlinear and time-varying conditions. Therefore, the kinetic parameters (modal parameters) have a certain degree of uncertainty. In the past, due to the simplification of mathematical model and the limitation of technical conditions, these uncertainty factors were often neglected, which limits the accuracy and reliability of modal identification results. This project plan to study the uncertainty influence mechanism of the heavy duty machining center, build the on-line identification model of structural parameters of heavy machine tool based on uncertain mechanism. In order to identify the modal parameters of machine tool under large stroke and multiple process conditions, a new method which based on big data analysis such as clustering and Bias estimation is proposed. Finally, by using the long term vibration data in the processing of the heavy gas turbine casing, the modal parameters and the degradation rule of its long-term operation state are identified to verify the validity of the method.