Abstract：A new method for the identification of dynamic load between satellite and rocket by using BP neural network is proposed. At first, a high-fidelity dynamic model of the satellite structure is built and the data base including both the structural response of the satellite and the base excitation acceleration at the interface between satellite and rocket can be obtained via numerical analysis or ground experiments; Secondly, the relationship between the structural response of the satellite and the excitation acceleration at the interface between satellite and rocket can be trained by using BP neural network, based on which the base excitation acceleration at the interface between satellite and rocket can be identified from measured structural response of the satellite; Finally, by applying the identified base excitation acceleration on the high-fidelity dynamic model of the satellite structure, the dynamic load at the interface between satellite and rocket can be calculated. Numerical simulation and vibration test are conducted to verify the effectiveness of the proposed method, which can provide strong support on the prediction of vibration load environment of the satellite structure under service condition.
收稿日期: 2022-05-16
出版日期: 2023-03-15
引用本文:
陈树海1, 3，郭安丰2，吴邵庆2，费庆国1. 基于BP神经网络的星箭界面动载荷识别[J]. 振动与冲击, 2023, 42(5): 279-286.
CHEN Shuhai1, 3, GUO Anfeng2, WU Shaoqing2, FEI Qingguo1. Dynamic load identification of satellite-rocket interface based on BP neural network. JOURNAL OF VIBRATION AND SHOCK, 2023, 42(5): 279-286.
[1] 马兴瑞, 韩增尧, 邹元杰, 等. 航天器力学环境分析于条件设计研究进展 [J]. 宇航学报, 2012, 33(1):1-12. [Ma Xing-rui, Han Zeng-yao, Zou Yuan-jie, et al. Review and Assessment of Spacecraft Mechanical Environment Analysis and Specification Determination [J]. Journal of Astronautics, 2012, 33(1):1-12.]
[2] Brunner O, Braeken R. Force measurement device for Ariane 5 payloads[C]. Proceedings of the 5th International Symposium on Environmental Testing for Space Programmes, Noordwijk, The Netherlands, June 15-17 2004.
[3] 邱吉宝, 张正平, 李海波. 航天器与运载火箭耦合分析相关技术研究进展[J]. 力学进展, 2012, 42(4): 416-436. [Qiu Ji-bao, Zhang Zheng-ping, Li Hai-bo. Progresses in research on coupled analysis technology for space vehicle and launch vehicles [J]. Advances in Mechanics, 2012, 42(4): 416-436.]
[4] 朱斯岩, 朱礼文. 基于有限元Craig缩聚模型的基础激励和载荷识别[J]. 北京航空航天大学学报, 2011, 37(6):665-668. [Zhu Si-yan, Zhu Li-wen. Base excitation and load identification based on Craig reduction model[J]. Journal of Beijing University of Aeronautics and Astronautics, 2011, 37(6):665-668.]
[5] 张永涛, 周徐斌, 杜冬, 等. 一种星箭动态界面力识别方法[J]. 航天器工程, 2015, 24(1): 62-69. [Zhang Yong-tao, Zhou Xu-bin, Du Dong, et al. An Identification Method for Satellite-launcher Dynamic Interface Force[J]. Spacecraft Engineering, 2015, 24(1): 62-69.]
[6] 尹健, 吴邵庆, 陈树海. 星箭连接界面处环形分布动载荷识别 [J]. 宇航学报, 2019, 40(12):1393-1402. [Yin Jian, Wu Shao-qing, Chen Shu-hai. Identification of Annularly Distributed Dynamic Load at Interface between Satellite and Rocket [J]. Journal of Astronautics, 2019, 40(12):1393-1402.]
[7] Wu S Q, Zheng Y, Sun Y W, et al. Identify the stochastic dynamic load on a complex uncertain structural system [J]. Mechanical Systems and Signal Processing, 2021, 147: 107114.
[8] Liu J, Meng X, Jiang C, et al. Time-domain Galerkin method for dynamic load identification [J]. International Journal for Numerical Methods in Engineering, 2016, 105:620-640.
[9] Liu J, Li K. Sparse identification of time-space coupled distributed dynamic load [J]. Mechanical Systems and Signal Processing, 2021, 148: 107177.
[10] 丁继峰, 韩增尧, 马兴瑞. 航天器动力学模型试验验证技术研究进展. 力学进展, 2012, 42(4):395-405. [Ding Ji-feng, Han Zeng-yao, Ma Xing-rui. Research evolution on the test verification of spacecraft dynamic model[J]. Advances in Mechanics, 2012, 42(4): 395-405.]
[11] Wang L, Liu Y, Gu K, et al. A radial basis function artificial neural network (RBF ANN) based method for uncertain distributed force reconstruction considering signal noises and material dispersion [J]. Computer Methods in Applied Mechanics and Engineering, 2020, 364: 112954.
[12] 夏鹏, 杨特, 徐江, 等. 利用时延神经网络的动载荷倒序识别[J]. 航空学报, 2021, 42(7):224452. [Xia Peng, Yang Te, Xu Jiang, et al. Reversed time sequence dynamic load identification method using time delay neural network [J]. Acta Aeronautica et Astronautica Sinica, 2021, 42(7):224452.]
[13] 张方, 朱德懋. 基于神经网络模型的动载荷识别[J].振动工程学报, 1997, 10(2):156-162. [Zhang Fang, Zhu Demao. The dynamic load identification research based on Neural Network model[J], Journal of Vibration Engineering, 1997, 10(2):156-162.]
[14] 沈花玉,王兆霞,高成耀,秦娟,姚福彬,徐巍.BP神经网络隐含层单元数的确定[J].天津理工大学学报,2008(05):13-15. [Shen Huayu, Wang Zhaoxia, Gao Chengyao, Qin Juan, Yao Fubin, Xu Wei. Determining the number of BP neural network hidden layer units [J], Journal of Tianjin University of technology, 2008 (05): 13-15.]
