GH4169高温合金电火花线切割加工质量预测研究  

Research on Wire Electrical-Discharge Machining Quality Prediction of GH4169 Superalloy

作　　者：马军 冯士浩 曹阳 李晓科 MA Jun;FENG Shi-hao;CAO Yang;LI Xiao-ke(He’nan Key laboratory of Intelligent Manufacturing of Mechanical Equipment,Zhengzhou University of Light Industry,He’nan Zhengzhou450002,China)

机构地区：[1]郑州轻工业大学河南省机械装备智能制造重点实验室,河南郑州450002

出　　处：《机械设计与制造》2020年第8期151-154,158,共5页Machinery Design & Manufacture

基　　金：国家自然科学基金(11602230);河南省高校科技创新团队支持计划(18IRTSTHN015);河南省高等学校重点科研项目(19A460031)。

摘　　要：针对镍-铬-铁基高温合金GH4169电火花线切割加工质量预测问题,通过正交实验设计并结合方差分析,分析放电电流及脉冲宽度、放电间隙、管数、加工限速对GH4169高温合金电火花线切割加工表面粗糙度和线切割速度的影响规律。在实验数据基础上训练BP神经网络,建立脉冲宽度、放电间隙、管数、加工限速对表面粗糙度和线切割速度影响的预测模型,预测结果表明,将正交实验与BP神经网络融合应用在GH4169高温合金电火花线切割加工中,不仅减少了工艺参数优化选择的盲目性,也提高了加工质量的预测精度和效率。Aiming at the wire electrical-discharge machining(WEDM)quality prediction of nickel-chromium-iron-based superalloy GH4169,the orthogonal experiment design and analysis of variance were used to study the discharge current,pulse width,discharge gap,tube number and processing speed limit for GH4169 superalloy.The influence of spark wire cutting on surface roughness and wire cutting speed. Based on the experimental data,BP neural network was trained to establish a prediction model of the influence of pulse width,discharge gap,tube number and processing speed limit on surface roughness and wire cutting speed.The analysis results show that the orthogonal experiment and BP neural network are applied.In the GH4169 superalloy WEDM,it not only reduces the blindness of process parameter optimization,but also improves the prediction accuracy and efficiency of processing quality.

关 键 词：GH4169 电火花线切割 加工质量 正交实验 BP神经网络 

分 类 号：TH16[机械工程—机械制造及自动化]