基于模糊系统的车削工件直径误差预测方法

Diameter error prediction of workpiece in turning process based on fuzzy system

为满足特定轴类工件的加工精度要求,根据车削过程中工件直径误差的特点,提出了用模糊系统预测车削工件直径误差的方法。通过设计车削实验,得到训练数据和测试数据。在分析梯度下降算法和传统遗传算法优缺点的基础上,将梯度下降算法嵌入传统遗传算法形成混合遗传算法。通过训练数据分别用梯度下降算法、传统遗传算法和混合遗传算法训练Mamdani型模糊系统,混合遗传算法的收敛效果优于梯度下降算法和传统遗传算法,用测试数据对三种算法训练的模糊系统进行测试,混合遗传算法的预测效果也是三种算法中最好的。预测结果表明,在一定的工件结构和工况条件下,用混合遗传算法训练的Mamdani型模糊系统进行车削工件直径误差的预测是可行的。

To satisfy machining accuracy requirements for specific workpieces,fuzzy system was adopted to predicate the diameter error of workpiece in turning process based on the characteristics of diameter error.Turning experiment was designed to obtain the original training data and testing data.After analyzing the advantages and disadvantages of gradient descent algorithm and traditional genetic algorithm,gradient descent algorithm was imbedded into traditional genetic algorithm to form the hybrid genetic algorithm.Using training data,Mamdani fuzzy system was trained by gradient descent algorithm,traditional genetic algorithm and hybrid genetic algorithm respectively.Results showed that hybrid genetic algorithm had better convergence than gradient descent algorithm and traditional genetic algorithm.The fuzzy system which was trained by three algorithms was tested by testing data respectively,the prediction effect of hybrid genetic algorithm was the best of three algorithms.The prediction results showed that under certain structure of workpiece and working conditions,Mamdani fuzzy system which was trained by hybrid genetic algorithm was feasible in predicating diameter error of workpiece in turning process.