基于损失函数的缩比模型冲击响应预报结果修正方法

1. 1. 南京航空航天大学 机械结构力学及控制国家重点实验室, 南京 210016;
2. 南京航空航天大学 振动工程研究所, 南京 210016;
3. 上海宇航系统工程研究所, 上海 210019;
4. 中国飞机强度研究所 结构冲击动力学航空科技重点实验室, 西安 710065
• 收稿日期:2021-05-11 修回日期:2021-12-17 出版日期:2022-08-15 发布日期:2021-10-27
• 通讯作者: 何欢,E-mail:hehuan@nuaa.edu.cn E-mail:hehuan@nuaa.edu.cn
• 基金资助:
国家自然科学基金(12072153)

Modified method for impact response prediction of scaled model based on loss function

WANG Yuexin1,2, HE Huan1,2, WU Tian3, XI Xulong4

1. 1. State Key Laboratory of Mechanics and Control of Mechanical Structures, Nanjing University of Aeronautics and Astronautics, Nanjing 210016, China;
2. Institute of Vibration Engineering Research, Nanjing University of Aeronautics and Astronautics, Nanjing 210016, China;
3. Shanghai Aerospace System Engineering Research Institute, Shanghai 210019, China;
4. Aviation Key Laboratory of Science and Technology on Structures Impact Dynamics, Aircraft Strength Research Institute of China, Xi'an 710065, China
• Received:2021-05-11 Revised:2021-12-17 Online:2022-08-15 Published:2021-10-27
• Supported by:
National Natural Science Foundation of China (12072153)

Abstract: Prediction of prototype results by the scaled model is one of the commonly used methods to research the impact response of large structures. Due to the strain rate effect, the similarity relationship between the prototype and the scale model is out of balance, which leads to the deviation of prediction of the prototype impact response by the scaled model test results. For this problem, a modified method for impact response prediction of the scaled model is investigated in this paper. A loss function for measuring the overall error of the flow yield stress is proposed to quantify the deviation between the scaled model and the prototype. When the loss function is minimized in the strain rate range, the velocity scaling factor satisfying the condition of the minimum global error can be obtained, by which the initial condition of the scaled model test and the scaling factors of other relevant quantities can be modified and the prediction error can be reduced finally. To verify the effectiveness of the modified method, numerical simulation and impact tests are carried out. The results show that the loss function method can effectively improve the prediction accuracy of the scaled model, and can predict the impact response of the prototype more accurately.