PCA-RPF构成的数控转台精度衰退模型

Positional Accuracy Decline Model of NC Rotary Table Based on PCA-RPF

针对数控转台精度衰退特性缺乏有效的评估方法的问题,提出一种数控转台重复定位精度衰退趋势预测模型,该模型将主成分分析法(principal component analysis,PCA)的多维融合能力与正则化粒子滤波算法(Regularized Particle Filter,RPF)的状态跟踪与预测的能力相结合。将数据驱动的方法引入到数控转台精度模型中,提取采集到的振动数据的时域和时频域特征;通过PCA融合多维特征向量,建立转台精度衰退趋势;在此基础上,利用RPF建立精度失效预测模型,预测精度的剩余寿命(RUL)。模型计算与试验测量的结果表明,该模型能够比较准确地得到数控转台精度失效的预测结果。

For NC rotary table,due to the lack of the effective evaluation methods for the deterioration characteristics of the repeatability positioning accuracy,a novel prediction approach based on the principal component analysis(PCA)and regularized particle filter(RPF)was proposed here.The model combines the multidimensional fusion of PCA with the state tracking and state prediction of RPF.The data driven method is introduced into the accuracy model,and the time domain and time-frequency domain features of the vibration data are extracted.Through PCA fusion of multidimensional feature vectors,the decline trend of rotary table accuracy is established.On this basis,the failure prediction model of precision is established by using RPF,and then the remaining life(RUL)of prediction accuracy is predicted.The results of model calculation and experimental measurement show that the model can accurately predict the accuracy failure of NC rotary table.