采用改进灰色神经网络的铣床热误差补偿研究

数控铣床在铣削零件过程中,主轴会受到温度变化影响而发生热变形,导致铣削零件误差较大,从而降低产品精度.对此,采用一阶线性微分方程推导GM(1,1)模型,创建灰色预测模型.将神经网络模型与灰色预测模型进行组合,建立灰色神经网络预测模型.引用粒子群算法,在粒子群算法中增加变异操作和修改惯性权重系数,给出改进粒子群算法优化灰...

Research on Thermal Error Compensation of Milling Machine Based on Improved Grey Neural Network

In the milling process of NC milling machine,the spindle will be affected by the temperature change and thermal deformation,resulting in greater error of milling parts,thereby reducing product accuracy.Therefore,the GM(1,1)model is derived from the first order Linear differential equation,and the grey prediction model is established.Combining the neural network model with the grey prediction model,the grey prediction model is established.By using particle swarm optimization(PSO)Algorithm,the mutation operation and Inertia Weight Coefficient are added to PSO Algorithm,and the operation steps of improving PSO Algorithm to optimize grey neural network prediction model are given.The thermal errors in the milling process of a milling machine are experimentally measured and compared with the prediction model.The results show that the residual error is larger when the grey neural network model compensates the milling spindle in X,Y and Z axes,and the error is larger when the improved grey neural network model compensates the milling spindle The resulting residuals are relatively small.The prediction model of Grey Neural Network based on improved particle swarm optimization algorithm can improve the milling accuracy of milling machine spindle.