摘要
纳米压痕实验由于试样准备简便、使用范围广等优势,在材料力学测试领域得到了广泛关注。本研究建立了一种考虑纳米压痕实验不确定性的Al 2024-T3铝合金材料塑性参数识别方法。首先,针对Al 2024-T3铝合金开展了纳米压痕实验,获取了载荷-位移曲线,由于材料存在不均匀性,所以实验曲线存在不确定性。基于超参数优化的人工神经网络,建立了材料性能参数与压痕响应加载曲线的关联。基于区间优化理论,引入压痕实验曲线的不确定性,以压痕实验曲线加载曲率为不确定性量,提出了基于双层嵌套遗传算法的材料参数反分析识别区间优化模型,并进行了参数识别反问题的求解。该方法的优势在于能够考虑到实验测量的不确定性,识别结果更可信。所建立方法的有效性在Al 2024-T3铝合金塑性参数识别中得到了验证,识别误差分别为:屈服应力-0.87%,硬化指数2.76%。该识别方法可用于小尺寸试样局部力学性能的检测领域。
Nanoindentation experiment has been widely concerned in the field of material mechanics testing due to its advantages of simple sample preparation and wide range of use.A plastic parameter identification method of the Al 2024-T3 aluminum alloy material considering the uncertainty of nanoindentation experiment is established.First of all,the nanoindentation experiment is carried out on the Al 2024-T3 alloy,and the load-displacement curve is obtained.Due to the inhomogeneity of the material,the experimental curve is uncertain.Based on the artificial neural network of superparametric optimization,the relationship between the material performance parameters and indentation response loading curve is established.Based on the interval optimization theory,the uncertainty of the indentation test curve is introduced.Taking the loading curvature of the indentation test curve as the uncertainty quantity,the interval optimization model of the material parameter identification based on the double-layer nested genetic algorithm is proposed,and the inverse problem of the parameter identification is solved.The advantage of this method is that it can take into account the uncertainty of experimental measurement,and the recognition result is more reliable.The validity of the established method has been verified in the identification of the Al 2024-T3 alloy plastic parameters.The identification errors of the yield stress and hardening index are-0.87% and 2.76%,respectively.This recognition method is expected to be used in the detection of mechanical properties of small size specimens.
作者
张桂涛
黄想
侯冰玉
王明智
ZHANG Guitao;HUANG Xiang;HOU Bingyu;WANG Mingzhi(School of Mechanical and Electrical Engineering,Xidian University,Xi'an 710071,China)
出处
《材料开发与应用》
CAS
2023年第6期41-51,共11页
Development and Application of Materials
关键词
参数识别
不确定性优化
神经网络
区间优化
压痕
parameter identification
uncertain optimization
neural network
interval optimization
indentation
