To solve the ill-posed problem and noise sensitivity in the recognition technology of shock load, an improved Bayesian regularization method based on Tikhonov regularization technology was proposed. By introducing the wavelet thresholding method to the Bayesian regularization method, it further solved the poor recognition accuracy of multi-source/continuous shock loads under high noise level, allowing for more accurate selection of regularization parameters and more reasonable elimination of noise influence. By carrying out the finite element simulation analysis of different impact loads and signal-to-noise ratio noise of aviation aluminum wall plate structures, and taking the correlation coefficient and relative error as the evaluation indexes, the recognition effects of the Tikhonov regularization method based on the L-curve method, the Tikhonov regularization method based on the Generalized Cross Validation method and the improved Bayesian regularization method are compared and discussed. The results show that the improved Bayesian method takes into account the smoothness of the curve and the accuracy of peak recognition. At the high noise level of 20dB, the average error of the peak value is controlled within 14% when the continuous shock load is recognized. In addition, the impact test based on the actual reinforced aviation aluminum wall plate structure is carried out, which verifies the ability of the method of this article to accurately identify the peak value of typical impact load in practical application, and the average error of the peak is controlled within 18%, which provides an effective way to solve the load identification problem in engineering.
Key words
Load identification /
Bayesian regularization /
Wavelet thresholding method /
Ill-posed problem
{{custom_keyword}} /
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
References
[1] 周盼,张权,率志君等.动载荷识别时域方法的研究现状与发展趋势[J].噪声与振动控制,2014,34(01):6-11.
ZHOU Pan, Zhang Quan, Sui Zhijun, et al. Research Status and Development Trend of Time Domain Method for Dynamic Load Identification [J]. Noise and Vibration Control, 2014,34(01):6-11.
[2] Golub G H, Wahba H G. Generalized cross-validation as a method for choosing a good ridge parameter [J]. Technometrics, 1979,21(2): 215-223.
[3] B.Radhika,C.S.Manohar.Reliability models for existing structures based on dynamic stateestimation and data based asymptotic extreme value analysis. Probabilistic Engineering Mechanics, 2010,25: 393-405.
[4] Sanchez J, Benaroya H. Review of force reconstruction techniques [J]. Journal of Sound and Vibration, 2014 ,333(14):2999-3018.
[5] 张玉良,杨飞,岳洪浩等.基于频域法的星箭连接分离装置的冲击载荷识别[J].振动与冲击, 2018,37(17):79-85.
ZHANG Yuliang, Yang Fei, Yue Honghao et al. Shock load identification of satellite-arrow connection separation device based on frequency domain method [J]. Vibration and Shock, 2018,37(17):79-85.
[6] 王能建,任春平,刘春生.一种新型分数阶Tikhonov正则化载荷重构技术及应用[J].振动与冲击, 2019,38(06):121-126+158.
WANG Nengjian, Ren Chunping, Liu Chunsheng. A new fractional-order Tikhonov regularized load reconstruction technology and its application [J].Journal of Vibration and Shock, 2019,38(06):121-126+158.
[7] 夏鹏,杨特,王乐等.平稳随机载荷的灵敏度分析识别方法[J].振动与冲击, 2022,41(23):300-306.
XIA Peng, Yang T, Wang Le, et al. Sensitivity analysis and identification method for stationary random load [J].Vibration and Shock, 2022,41(23):300-306.
[8] 姜金辉,陈国平,张方.多点平稳随机载荷识别方法研究[J].振动工程学报, 2009,22(02):162-167.
JIANG Jinhui, Chen Guoping, Zhang Fang.Identification method of multi-point stationary random load[J].Chinese Journal of Vibration Engineering, 2009,22(02):162-167.
[9] Tikhonov A N，Arsenin V Y. Solutions of lll -posed problems
[M]. New York: Wiley,1977.
[10] 常晓通，阎云聚，刘鎏等．基于Green函数和正则化的动态载荷识别方法[J] ．振动﹑测试与诊断，2014，34(1): 124-129.
CHANG Xiaotong，YAN Yunju，LIU Jian，et al. Dynamic load identification method based on Green function and regularization method [J]. Journal of Vibration, Measurement and Diagnosis, 2014,34(1):124-129.
[11] Choi H G, Thite A N, Thompson D J. Comparison of methods for parameter selection in Tikhonov regularization with application to inverse force determination [J]. Journal of Sound & Vibration, 2007,304(3):894-917.
[12] 陈建鼎. 动载荷识别的贝叶斯正则化方法及实验研究[D].南京航空航天大学, 2021.
CHEN Jianding. Bayesian regularization method and experimental study of dynamic load recognition [D]. Nanjing University of Aeronautics and Astronautics, 2021.
[13] 石张昊. 基于正则化的飞机动载荷识别技术研究[D].中国民航大学, 2022.
SHI Zhanghao. Research on aircraft dynamic load recognition technology based on regularization [D].Civil Aviation University of China, 2022
[14] 矫媛,黄斌文.基于小波变换的图像去噪方法综述[J].电子制作, 2015(07):55-56.
JIAO Yuan, Huang Binwen. Review of image denoising method based on wavelet transform [J].Electronic Production, 2015(07):55-56.
{{custom_fnGroup.title_en}}
Footnotes
{{custom_fn.content}}
PDF(2388 KB)
