摘要
为了优化油膜附水滴切削工艺参数,降低切削力,减少刀具磨损和提高切削加工质量,结合神经网络和遗传算法,基于MATLAB软件对切削试验数据进行训练。通过构建径向基函数RBF和反向传播BP神经网络模型,并对切削力和表面粗糙度综合加权后综合目标进行预测。对比两种模型的精度后,选择更精确的RBF神经网络模型。利用遗传算法GA优化RBF神经网络模型的切削工艺参数,研究工艺参数间的交互作用。结果表明:最优参数组合为切削速度130m/min,进给量0.13mm/r,背吃刀量0.17mm。可见对基于油膜附水滴技术的切削工艺参数优化具有参考价值。
To optimize cutting process parameters of the oil on water(OoW),for reducing the cutting force,reducing the tool wear and improving the quality of cutting,through combining neural network and genetic algorithm,the cutting test data is trained based on MATLAB platform.The radical basis function(RBF)neural network model and the back propagation(BP)neural network model are built on to predict the comprehensive of cutting force and surface roughness after comprehensive weighted.After comparing two models,the higher precision RBF model is selected.The cutting process parameters of RBF neural network model is optimized by the genetic algorithm,and the interaction in the process parameters is studied.Experimental results show that when the cutting speed is 130m/min,the feed is 0.13mm/r,and the cutting amount is 0.17mm,the parameter combination is optimal.So the optimization of cutting process parameters based on oil on water technology has a certain reference value.
作者
李俊杰
刘永姜
郭文轩
于婷
Li Junjie;Liu Yongjiang;Guo Wenxuan;Yu Ting
出处
《工具技术》
北大核心
2022年第1期56-59,共4页
Tool Engineering
基金
先进制造技术山西省重点实验室开放基金资助项目(XJZZ201806)。
关键词
油膜附水滴
神经网络
遗传算法
参数优化
oils on water(OoW)
neural network
genetic algorithm(GA)
parameter optimization
