Grinding chatter is the main manifestation of grinder machine malfunction.The signals of grinding chatter are usually non-stationary and non-linear.It will be more difficult to detect and predict in the strong background signals and noise.In order to know the occurrence of chatter beforehand, grinding chatter prediction method based on AR (p) type high-order fuzzy time series is proposed.Firstly, by analyzing the vibration signal in the grinding process, the feature quantity real-time variance that can intuitively reflect the change of grinding state is extracted.Then, An AR (p) type high-order fuzzy time series prediction model is established using a set of real-time variance of grinding chatter signals as a training set.Finally, the method proposed in this paper is verified by the actual measured chatter signal of the grinder.Experimental results show that the predicted values obtained by using the real-time variance time series to establish AR (p) type high-order time series prediction model are very close to the true values, which verifies the effectiveness and feasibility of the proposed method.%针对磨床磨削颤振信号的非平稳和非线性特点, 提出一种基于AR (p) 型高阶模糊时间序列的磨削颤振预测方法.首先对磨削过程振动信号进行分析, 提取能够直观反映磨削颤振状态变化的特征量———实时方差;然后以一组磨削颤振信号实时方差时间序列作为训练集, 建立AR (p) 型高阶模糊时间序列预测模型;最后通过实测磨床颤振信号对所提出的方法进行了验证.结果表明:利用实时方差时间序列建立AR (p) 型高阶模糊时间序列预测模型来预测磨削颤振状态, 得到的预测值与真实值十分接近.
展开▼
