基于HPM-JTM混合模型参数估计优化的非高斯过程模拟

PDF(2247 KB)

振动与冲击 ›› 2025, Vol. 44 ›› Issue (14) : 1-10.

振动理论与交叉研究

基于HPM-JTM混合模型参数估计优化的非高斯过程模拟

罗颖*1，程彦凯1，韩艳1，刘雨辰2，胡朋1

作者信息 +

Non-Gaussian process simulation based on the parametric estimation optimization of a HPM-JTM hybrid model

LUO Ying*1，CHENG Yankai1，HAN Yan1，LIU Yuchen2，HU Peng1

Author information +

文章历史 +

摘要

由于实际工程的激励较为复杂，常呈现非高斯特性，导致高斯过程的假设不再适用，因此需要开展非高斯过程模拟。目前而言，常见的方法是通过高斯过程转换实现非高斯过程模拟。相比一般的隐式映射，Hermite多项式模型(Hermite polynomial model，HPM)和Johnson转换模型(Johnson transformation model，JTM)提供了非高斯过程与标准高斯过程之间的显式转换。针对HPM-JTM混合模型，该研究探讨了如何进一步提升模拟效率。首先，为了避免迭代过程，基于支持向量回归优化了HPM和JTM参数估计流程，提高了参数估计效率。随后，通过线性滤波法和谐波合成法的模拟流程对比，在非高斯过程模拟中采用线性滤波法能够提升模拟效率。最后，结合波浪场和脉动风场的实例分析，展示了改进流程的精度和效率。结果表明，改进流程能够在保证精度的同时实现多变量非高斯过程的高效模拟。

Abstract

The excitation source in actual engineering is relatively complicated, which often presents non-Gaussian features.As a result, the assumption of Gaussian process is no longer applicable.Therefore, it is necessary to carry out the simulation of non-Gaussian process.Currently, the usual method is to realize the simulation of non-Gaussian process through the translation of Gaussian process.Compared with the common implicit mapping, Hermite polynomial model (HPM) and Johnson translation model (JTM) provide more efficiently the explicit translation between the non-Gaussian process and standard Gaussian counterpart.How to further improve the simulation efficiency against the HPM-HTM hybrid model was investigated.Firstly, in order to avoid the iterative process, the parameter estimation process of HPM and JTM was optimized based on the support vector regression to improve the efficiency of parameter estimation.Subsequently, through the comparison of the simulation flows based on the linear filtering method and harmony superposition method, the simulation efficiency was improved in the non-Gaussian process simulation.Finally, the examplic analysis results of the wave field and fluctuating wind field were combined presented to demonstrate the accuracy and efficiency of the modified simulation process.The results show that the improved simulation flow can realize the high-efficiency simulation of the multivariate non-Gaussian process with the guarantee of accuracy.

导出引用

罗颖1, 程彦凯1, 韩艳1, 刘雨辰2, 胡朋1. 基于HPM-JTM混合模型参数估计优化的非高斯过程模拟[J]. 振动与冲击, 2025, 44(14): 1-10

LUO Ying1, CHENG Yankai1, HAN Yan1, LIU Yuchen2, HU Peng1. Non-Gaussian process simulation based on the parametric estimation optimization of a HPM-JTM hybrid model[J]. Journal of Vibration and Shock, 2025, 44(14): 1-10

参考文献

[1] Ochi M.K. Non-Gaussian random processes in ocean engineering [J]. Probabilistic Engineering Mechanics, 1986, 1(1): 28-39.
[2] Gusev A.A. Peak factors of Mexican accelerograms: Evidence of a non‐Gaussian amplitude distribution [J]. Journal of Geophysical Research: Solid Earth, 1996, 101(B9): 20083-20090.
[3] Fragoulis A, Fragoulis A. The complex terrain wind environment and its effects on the power output and loading of wind turbines [C]. 35th Aerospace Sciences Meeting and Exhibit.Reno, NV：American Institute of Aeronautics and Astronautics, 1997.
[4] Stathopoulos T. PDF of wind pressures on low-rise buildings [J]. Journal of the Structural Division, 1980, 106(5): 973-990.
[5] Kareem A, Zhao J. Analysis of non-Gaussian surge response of tension leg platforms under wind loads [J]. Journal of Offshore Mechanics and Arctic Engineering，1994, 116: 137-144.
[6] 杨雄伟, 周强, 李明水, 等. 复杂曲面屋盖脉动风压的非高斯特性及峰值因子研究[J]. 振动与冲击, 2021, 40(10): 315-322.
YANG Xiong-wei, ZHOU Qiang, LI Ming-shui, et al. Non-Gaussian features and peak factors of fluctuating wind pressures on a complex curved roof [J]. Journal of Vibration and Shock, 2021, 40(10): 315-322.
[7] Benasciutti D, Tovo R. Fatigue life assessment in non-Gaussian random loadings [J]. International Journal of Fatigue, 2006, 28(7): 733-746.
[8] Grigoriu M. Crossings of non-Gaussian translation processes [J]. Journal of Engineering Mechanics, 1984, 110(4): 610-620.
[9] Gioffrè M, Gusella V, Grigoriu M. Non-Gaussian wind pressure on prismatic buildings. II: numerical simulation [J]. Journal of Structural Engineering, 2001, 127(9): 990-995.
[10] Winterstein S R. Nonlinear vibration models for extremes and fatigue [J]. Journal of Engineering Mechanics, 1988, 114(10): 1772-1790.
[11] Johnson N.L. Systems of Frequency Curves Generated by Methods of Translation [J]. Biometrika, 1949, 36(1-2): 149-176.
[12] Yang L., Gurley K.R, Prevatt D O. Probabilistic modeling of wind pressure on low-rise buildings [J]. Journal of Wind Engineering and Industrial Aerodynamics, 2013, 114: 18-26.
[13] Yang L, Gurley K.R. Efficient stationary multivariate non-Gaussian simulation based on a Hermite PDF model [J]. Probabilistic Engineering Mechanics, 2015, 42: 31-41.
[14] Ding J, Chen X Z. Moment-based translation model for hardening non-Gaussian response processes [J]. Journal of Engineering Mechanics, 2016, 142(2): 06015006.
[15] Wu F, Huang G, Liu M. Simulation and peak value estimation of non-gaussian wind pressures based on Johnson transformation model [J]. Journal of Engineering Mechanics, 2020, 146(1): 04019116.
[16] Wu F, Liu M, Huang G, et al. Simulation of stationary non-Gaussian multivariate wind pressures based on moment-based piecewise Johnson transformation model [J]. Probabilistic Engineering Mechanics, 2022, 68: 103225.
[17] Luo Y, Liu Y, Han Y, et al. Multivariate non-Gaussian process simulation based on HPM-JTM hybrid model [J]. Probabilistic engineering mechanics, 2023, 73:103472.
[18] Fisher R A, Cornish E A. The percentile points of distributions having known cumulants [J]. Technometrics, 1960, 2(2): 209-225.
[19] Winterstein S.R., Kashef T. Moment-based load and response models with wind engineering applications [J]. Journal of Solar Energy Engineering, 2000, 122(3): 122-128.
[20] Vapnik V. The Nature of Statistical Learning Theory [M]. New York: Springer-Verlag, 1995.
[21] Yang H.Y, Liu Y. Methods of support vector machine on classification of expansive soils [J]. Advanced Materials Research, 2012, 531: 562-565.
[22] Su Z., Fu X. Kernel principal component genetic algorithm and improved SVR stock selection model [J]. Statistical Research, 2013, 30(5): 54-62.
[23] 王晓兰, 王明伟. 基于小波分解和最小二乘支持向量机的短期风速预测[J]. 电网技术, 2010, 34(1): 179-184.
WANG Xiao-lan, WANG Ming-wei. Short-Term Wind Speed Forecasting Based on Wavelet Decomposition and Least Square Support Vector Machine [J]. Power System Technology, 2010, 34(1): 179-184.
[24] 张华, 郁永静, 冯志军, 等. 基于小波分解与支持向量机的风速预测模型[J]. 水力发电学报, 2012, 31(1): 208-212.
ZHANG Hua, YU Yong-jing, FENG Zhi-jun, et al. Wind speed forecasting model based on wavelet decomposition and support vector machine [J]. Journal of Hydroelectric Engineering, 2012, 31(1): 208-212.
[25] 雷静. 支持向量机在基于边缘检测的图像分割和函数估计中的应用[D]. 河北工业大学, 2003.
LEI Jing. The application of support vector machine in image segmentation based on edge detection and function regression [D]. Hebei University of Technology, 2003.
[26] 甘旭升, 张洪才, 程咏梅, 等. 基于鲁棒最小二乘支持向量机的气动参数拟合[J]. 计算机工程与应用, 2007, 43(31): 233-235+239.
GAN Xu-sheng, ZHANG Hong-cai, CHENG Yong-mei, et al. Aerodynamic parameter fitting based on robust least squares Support Vector Machines [J]. Computer Engineering and Applications, 2007, 43(31): 233-235+239.
[27] Shinozuka M, Deodatis G. Simulation of stochastic processes by spectral representation [J]. Applied Mechanics Reviews, 1991, 44(4): 191-204.
[28] Iannuzzi A, Spinelli P. Artificial Wind Generation and Structural Response [J]. Journal of Structural Engineering, 1987, 113(12): 2382-2398.
[29] Deodatis G，Micaletti R C． Simulation of highly skewed non-Gaussian stochastic processes [J]. Journal of Engineering Mechanics, 2001, 127(12): 1284-1295．
[30] Owen J S, Eccles B J, Choo B S, et al. The application of auto-regressive time series modeling for the time-frequency analysis of civil engineering structures [J]. Engineering Structures, 2001, 23: 521-536.
[31] Samaras E, Shinozuka M, Tsurui A. ARMA representation of random process [J]. Journal of Engineering Mechanics ASCE, 1985, 111(3): 449-461.
[32] 张文福, 谢丹, 刘迎春, 等. 下击暴流空间相关性风场模拟[J]. 振动与冲击, 2013, 32(10): 12-16.
ZHANG Wen-fu, XIE Dan, LIU Ying-chun, et al. Simulation of downburst wind field with spatial correlation [J]. Journal of Vibration and Shock, 2013, 32(10): 12-16.
[33] 李锦华, 陈水生, 吴春鹏, 等. 基于时变AR模型的非平稳非高斯随机过程的数值模拟[J]. 振动与冲击, 2015, 34(17): 142-146+160.
LI Jin-hua, CHEN Shui-sheng, WU Chun-peng, et al. Simulation of non-stationary non-Gaussian stochastic process based on a time-varying AR model [J]. Journal of Vibration and Shock, 2015, 34(17): 142-146+160.
[34] Choi M, Sweetman B. The Hermite moment model for highly skewed response with application to tension leg platforms [J]. Journal of Offshore Mechanics and Arctic Engineering, 2010, 132(2): 021602.
[35] Tognarelli M A, Zhao J, Kareem A. Equivalent Statistical Cubicization for System and Forcing Nonlinearities [J]. Journal of Engineering Mechanics, 1997, 123(8): 890-893.
[36] Chakrabarti S.K. Technical note: On the formulation of Jonswap spectrum [J]. Applied Ocean Research, 1984, 6(3): 175-176.
[37] Davenport A.G. The dependence of wind loads on meteorological parameters [M]. Wind Effects on Buildings & Structures, 1967.