摘要
以工程陶瓷切削过程为研究对象,通过DEFORM有限元分析、灰色系统和BP神经网络算法,研究了工程陶瓷切削温度特征。根据切削温度有限元仿真数据,构建了不等时距灰色预测模型,预测了工程陶瓷切削温度,在此基础上建立了两层BP神经网络,结合两种算法的优点,提高了预测精度。研究结果表明,切削开始阶段切削温度迅速达到峰值,随后趋于稳定。随着切削速度、进给速度、切削深度和主偏角的增加,切削温度均呈上升趋势,其中切削温度对切削速度最为敏感。
The object of study is engineering ceramics cutting process,the temperature characteristic of engineering ceramics was studied respectively by finite element analysis DEFORM software,gray system and BP neural network algorithm.The unequal interval gray prediction model,was established based on the twotier BP neural network,which simulated the cutting temperature in the paper.Combining the advantages of two algorithms,the prediction accuracy was improved.The results indicated that the temperature quickly reached a peak in beginning of cutting,and then stabilized,the cutting temperature was increased with increasing of cutting speed,feed speed,cutting depth and main angle.The influence of cutting speed on cutting temperature was most significant.
出处
《组合机床与自动化加工技术》
北大核心
2014年第7期1-4,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
切削温度
灰色系统
BP神经网络
有限元
工程陶瓷
cutting temperature
gray system
BP neural network
finite element simulation
engineering ceramics
