基于步长控制的铣削稳定性预测全离散法

发布时间：2018-06-29 03:52

本文选题：铣削稳定性 + Floquet理论　；参考：《组合机床与自动化加工技术》2017年12期

【摘要】：基于控制离散步长来保证预测精度的思想,提出了一种改进的用于预测铣削稳定性叶瓣图的时域全离散法。一直以来,铣削工艺系统的动力学过程被看作带有单个延时反馈量的周期性线性系统,其数学模型是无限维时滞微分方程组。全离散法在一个周期内将时滞微分方程离散为有限个常微分方程,求解前后两步的状态转移矩阵,继而得到一个完整周期内的状态转移矩阵。在选定的转速范围内循环求解过程中,通过控制离散步长小于设定值来达到提高低转速条件下全离散法的收敛率的目的。通过仿真和切削实验,证明方法具有较高的预测精度和计算效率。
[Abstract]:Based on the idea of controlling discrete step size to ensure prediction accuracy, an improved time domain full discrete method is proposed to predict the stability of milling vanes. All along, the dynamic process of milling process system is regarded as a periodic linear system with a single time-delay feedback, and its mathematical model is an infinite dimensional delay differential equation system. The full discrete method discretizes the delay differential equation into finite ordinary differential equations in a period, solves the state transfer matrix of two steps before and after, and then obtains the state transfer matrix in a complete period. In order to improve the convergence rate of the full discrete method under the condition of low rotational speed, the discrete step size is smaller than the set value in the process of cycle solution in the selected rotational speed range. Through simulation and cutting experiments, it is proved that the method has high prediction accuracy and calculation efficiency.
【作者单位】：清华大学机械工程系;精密超精密制造装备及控制北京市重点实验室;海军航空工程学院飞行器工程系;
【基金】：国家04科技重大专项课题(2013ZX04001-021)
【分类号】：TG54

【相似文献】