摘要
将人工神经网络引入磨削加工领域。针对BP算法存在收敛速度慢,容易陷入局部极小值以及全局搜索能力弱等缺陷,采用遗传算法训练 BP神经网络,设计了基于进化神经网络的学习算法,建立了外圆纵向磨削表面粗糙度的进化神经网络预测模型。实验和仿真结果表明,基于进化计算的 BP神经网络可以克服单纯使用 BP神经网络易陷入局部极小值等问题,预测精度较高,对提高外圆纵向磨削加工的自动化程度具有重要的意义。通过在线监测磨削参数,所提供的预测方法可以实现对工件表面粗糙度的在线预测。
Artificial neural networks were introduced to the grinding area. Focusing on some disadvantages in BP (Back Propagation) algorithm, such as low convergence rate, easily falling into local minimum point and weak global search capability, a new learning algorithm was presented, that used GA (Genetic Algorithm) to train BP neural networks. The prediction model of surface roughness in cylindrical longitudinal grinding based on evolutionary neural networks was proposed in detail. The experimental and the simulating results show that the combination of BP and GA can effectively overcome the problems of easily falling into local minimum point, and this method can obtain higher accuracy of prediction. By monitoring the grinding parameters, this proposed method can realize on-line prediction for the workpiece roughness.
出处
《中国机械工程》
EI
CAS
CSCD
北大核心
2005年第3期223-226,共4页
China Mechanical Engineering
基金
吉林省科技发展计划资助项目(20020632)
吉林大学青年教师基金资助项目
关键词
外圆纵向磨削
进化神经网络
表面粗糙度
在线预测
cylindrical longitudinal grinding
evolutionary neural network
surface roughness
on-line prediction
