马猛, 王明红.基于进化神经网络的304不锈钢车削加工表面粗糙度预测[J].轻工机械,2019,37(6):44-47 |
基于进化神经网络的304不锈钢车削加工表面粗糙度预测 |
Prediction of Surface Roughness for 304 Stainless Steel Turning Based on Evolutionary Neural Network |
|
DOI:10.3969/j.issn.1005 2895.2019.06.009 |
中文关键词: 金属切削 表面粗糙度预测 正交试验 进化神经网络 |
英文关键词:metal cutting surface roughness prediction orthogonal test evolutionary neural network |
基金项目: |
|
摘要点击次数: 1291 |
全文下载次数: 1233 |
中文摘要: |
表面粗糙度是衡量加工零件质量的重要指标之一,对表面粗糙度进行提前预测有利于提高加工质量。课题组采用正交试验方法进行了YG8硬质合金刀具干式车削304不锈钢棒料的实验,得到不同切削条件下的表面粗糙度。由于BP神经网络的算法预测精度不高而且容易陷入局部极小值,利用遗传算法的全局搜索能力优化BP神经网络的结构和初值,建立基于进化神经网络的表面粗糙度预测模型。结果表明:进化的BP神经网络模型有效地克服了BP神经网络容易陷入局部极小值的缺陷,实现了表面粗糙度的精确预测。 |
英文摘要: |
The surface roughness is one of the important indicators for the quality of machined surface. It is helpful for improving machining quality to predict surface roughness in advance. Dry turning 304 stainless steel with YG8 carbide tool was carried out by orthogonal test method, the surface roughness was obtained under different cutting conditions. Due to the prediction accuracy of BP neural network algorithm not high and easy to get into the local minimum value, the structure and initial value of BP neural network were optimized by the global search ability of genetic algorithm, and the surface roughness prediction model based on evolutionary neural network was established.The results show that evolutionary neural network model overcomes the defect of BP neural network easily falling into local minimum point,and realizes the accurate prediction of surface roughness. |
查看全文 查看/发表评论 下载PDF阅读器 |
关闭 |