轴类零件外圆纵向磨削尺寸智能预测和控制系统

针对纵向磨削非线性和非静态的特点,建立了轴类零件纵向磨削的E lman动态神经网络尺寸预测模型。为了提高尺寸预测的准确性,将实际磨削尺寸的一阶导数和二阶导数做为网络的输入。采用论域自调整策略和模糊控制理论建立了纵向磨削的控制模型,选择工件的转速vw作为控制变量。仿真和实验结果表明所建立的神经网络尺寸预测模型和模糊自适应控制模型是正确的。

Intelligent prediction and control system of shaft workpiece size in cylinder traverse grinding

According to the characters of non-liner and non-static state in traverse grinding,a size intelligent prediction and control model based on the dynamic Elman neural network was constructed.The first and the second derivative of the actual amount removed from the workpiece were added into the network input to improve the prediction accuracy.A flexible factor was introduced to the fuzzy control model,which can self-adapt and adjust the universe of discourse in the fuzzy control and the workpiece peripheral speed v_w was selected as control variable.The results of simulation and experiment indicate that the developed neural network size prediction model and adaptive fuzzy control model are feasible and characterized by high prediction and control precisions.