摘要
利用TLS-W50000A微机控制弹簧试验机,对不同壁厚减薄率下的锡青铜QSn7-0.2强力旋压件进行等温恒应变速率下的单向准静态拉伸试验。基于获得的试验数据,建立基于BP神经网络技术、不同壁厚减薄率下的常温本构模型。结果表明:BP神经网络本构关系模型具有很高的预测精度,可以较好地描述不同壁厚减薄率下锡青铜QSn7-0.2在拉伸变形时的应力-应变关系,为强力旋压工艺本构关系模型的建立提供了一种准确有效的方法。
Using TLS - WS0000A microcomputer control spring testing machine, the uniaxial tensile quasistatic experiment under the iso- thermal constant strain rate was conducted for QSn7 - 0. 2 copper alloy power spinning spieces with different wall thickness reduction rati- os. Based on the obtained experimental data, the BP neural network technology was adopted to establish the normal temperature constitu- tive relationship model under different wall thickness reduction ratios. The results show that the BP neural network constitutive relationship model has high prediction accuracy and can accurately describe the relationship between stress and strain of tin bronze QSn7 - 0. 2 with different wall thickness reduction ratios during tensile deformation, and it provides an accurate and effective method for the constitutive modeling of power spinning.
出处
《锻压技术》
CAS
CSCD
北大核心
2014年第2期150-153,共4页
Forging & Stamping Technology
基金
山西省自然科学基金资助项目(2012011023-2)
山西省高校高新技术产业化项目(20120021)
