论文部分内容阅读
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。
TLS-W50000A computer-controlled spring tester was used to test the uniaxial quasi-static tensile test under constant temperature and constant strain rate for the bronze QSn7-0.2 strong spinning parts with different wall thickness reduction rates. Based on the experimental data obtained, a normal temperature constitutive model based on BP neural network and different wall thickness reduction rates was established. The results show that the constitutive model of BP neural network has a high predictive accuracy and can well describe the stress-strain relationship of the QSn7-0.2 tin bronze under tensile strain at different wall thickness thinning rates. The establishment of the constitutive model provides an accurate and effective method.