摘要
通过神经网络技术建立了快走丝电火花线切割加工工艺模型 ,利用穷举法建立了具有一定人工智能的工艺参数全局优化系统 ,开发了模具电火花加工过程仿真系统。该系统不仅可以精确预测加工效果 ,而且克服了工艺参数表的局限性 ,弥补了建立在工艺参数表基础上的参数自动选取系统的缺陷 ,实现了工艺参数全局最优化。测试结果及实际使用结果表明本文所建立的仿真系统反映了机床的加工工艺特性 ,预测误差基本控制在 8%内 ,系统的参数优化选取功能使机床的加工性能得以充分发挥。仿真系统具有广泛的通用性 ,可适用于不同类型的线切割加工机床。
An attempt at modeling the fast cutting wire electrical discharge machining(WEDM) process through artificial neural networks was presented. Computer simulation system of fast cutting WEDM process was developed. The system can not only forecast the cutting speed and surface roughness accurately, but also fully optimize all the machining parameters. Verification experiments indicated that the estimated error of the system was less than 8%. The optimization result was also proved to be reasonable.
出处
《机械科学与技术》
CSCD
北大核心
2000年第5期777-779,共3页
Mechanical Science and Technology for Aerospace Engineering
基金
上海市青年科技启明星计划资助项目!( 96QF14 0 0 6)
关键词
神经网络
仿真系统
电火花线切割加工
快走丝
Wire electrical discharge machine(WEDM)
Artificial neural networks
Process modeling
Parametric optimization
Computer simulation
