线切割加工中切缝宽度与材料去除率的优化研究

摘要

在精密制造业中,切缝宽度对尺寸精度的影响尤为显著,而材料去除率是影响加工效率的最重要指标,其与切缝宽度之间关系复杂且相互制约,一组加工参数难以同时获得较小的切缝宽度和较高的材料去除率.针对此问题,运用BP神经网络与粒子群算法(PSO)的混合算法建立多目标预测优化模型;以Ti6Al4V合金为实验对象,以水压、脉冲时间、脉冲间隙、伺服电压和电极丝张力为工艺参数,以切缝宽度(Kerf)和材料去除率(MRR)为工艺目标,设计田口实验.结果显示,Kerf和MRR的预测平均相对误差分别为5.32％和6.14％,优化得到单目标和多目标最优工艺参数,Kerf同比降低11.10％,MRR同比提高27.37％,表明对切缝宽度和材料去除率的预测与参数优化效果显著.%Cutting width and the material removal rate in LS-WEDM (low speed wire electrical discharge machining) is important processing indicators.In precision manufacturing,the cutting width particularly influences on the dimensional accuracy,and the material removal rate is the most important indicator influencing the processing efficiency,which has a complex and mutual restraint relationship with cutting width,a set of processing parameters is difficult to obtain smaller cutting width and higher material removal at the same time.To solve this problem,multi-objective optimization model is established by using BP (Back Propagation) neural network and particle swarm optimization (PSO),and in which Ti6Al4V alloy is took as testing object,water pressure,pulse-on time,pulse-off time,servo voltage and wire tension are took as testing parameters,the material removal rate (MRR) and the cutting width (Kerf) are took as processing indicators via Taguchi experiment.The results illustrate the average relative error of Kerf and MRR are of 5.32％ and 6.14％ respectively.Meanwhile the single-objective and multi-objective optimal processing parameters are obtained,and Kerf decreases of 11.10％,MRR increases of 27.37％ comparing with the previous results.The present multi-objective optimization model has a significant effect for predicting and optimizing cutting width and material removal rate.