Measurement uncertainty propagation in spindle error separation techniques - Investigation by means of stochastic spectral method
Measurement uncertainty propagation in spindle error separation techniques - Investigation by means of stochastic spectral method
复制标题
主轴误差分离技术中的测量不确定度传播 - 随机谱法研究
DOI：
10.1016/j.ijmachtools.2019.03.006
复制
发表时间：
2019
影响因子：
14
通讯作者：
Czarske Juergen
中科院分区：
文献类型：
--
作者：
Shi Shengyu;Zhang Hao;Qu Jinping;Jin Gang;Kuschmierz Robert;Czarske Juergen
主轴是几乎所有类型机床中最关键的子系统。误差分离技术为主轴误差测量提供了最先进的原理，没有系统偏差。然而，虽然大量的研究工作已经完成，特别是关于他们的谐波抑制问题，他们很少在工业中应用，由于不可预测的和不稳定的测量精度。在此背景下，本文将研究重点从谐波抑制问题转移到测量不确定度的传播问题上。首先，建立了两步误差分离法的系统模型，推导了测头不确定度的传播规律。传播定律允许对所产生的不确定性进行定量预测。在此基础上，提出了三种改进的TSM：角度优化TSM、混合TSM和融合TSM，使测量不确定度显著降低，并彻底解决了谐波抑制问题。蒙特卡罗模拟和实验验证了传播规律和改进方法的可行性。本研究首次实现了测量不确定度的定量评估和降低，并在我们看来，可能会带来误差分离技术的范式转变。
The spindle is the most critical subsystem in almost all types of machine tools. Error separation techniques provide the state-of-the-art principles for the spindle error measurement, suffering no systematic deviation. However, although substantial research effort has been accomplished, particularly regarding their harmonic suppression problem, they have rarely been applied in industry due to the unpredictable and unstable measurement precision. In this context, this paper changes the research focus from the harmonic suppression problem to the propagation of the measurement uncertainty. First, the system model of the two-step error separation method (TSM) is established, from which the propagation law of the probe uncertainty is analytically derived. The propagation law allows for a quantitative prediction of the resulting uncertainty. Furthermore, three improved TSMs are put forward: the angle-optimized TSM, the hybrid TSM, and the fusion TSM, which enable a significant reduction of the measurement uncertainty and solve the harmonic suppression problem completely. Monte Carlo simulations and experiments validate the propagation law and the viability of the improved approaches. This research, for the first time, realizes the quantitative evaluation and reduction of the measurement uncertainty, and in our opinion, may bring about a paradigm shift for the error separation techniques.