摘要
基于RBF神经网络理论,对工程陶瓷点磨削表面硬度HV与切削速度vc、进给速度f、切削深度ap、倾斜角α、和偏转角β五个工艺参数的关系进行了单因素数值拟合,并以拟合优度对拟合结果进行了检验,检验结果表明模型具有较高可信度。基于遗传优化算法,对点磨削表面硬度关于五个工艺参数的多元模型进行了优化建模,设计了正交实验对模型进行检验,最大误差在10%以内,表明模型具有较高的可靠性。
Inpoint grinding engineering ceramic, the relationship between surface hardness HV and cutting speed vc, feed speedf, cutting depth ap,inclining angleαanddeflecting angleβwere univariate numerical fitting based on RBF neural network. Andfitting resultsare tested by the coefficient of determination, test results show that the model has high credibility. Amultivariate modelissolvedbetween surface hardness with fiveprocess pa-rametersbased on genetic optimization algorithmin point grinding. The orthogonal experiment was designed to test the model,the maximum error is less than 10%. The results indicated that the model has high reliability.
出处
《组合机床与自动化加工技术》
北大核心
2015年第1期30-33,共4页
Modular Machine Tool & Automatic Manufacturing Technique
基金
国家自然科学基金项目资助(51275083)
关键词
表面硬度
数值模拟
遗传算法
RBF神经网络
点磨削
工程陶瓷
surface hardness
numerical simulation
genetic algorithms
RBF neural network
point grinding
engineering ceramics
