基于增广SVM的结构动力学模型修正方法研究

PDF(1776 KB)

振动与冲击 ›› 2017, Vol. 36 ›› Issue (15) : 194-202.

论文

陈喆，何欢，陈国平

作者信息 +

The research of model updating of using support vector machine based on hybrid basis functions

CHEN Zhe，HE Huan，CHEN Guo-ping

Author information +

文章历史 +

摘要

研究了基于代理模型的有限元模型修正方法，针对支持向量机(SVM)在处理非线性程度不高函数时容易出现过拟合，提出了一种采用混合基函数形式的增广SVM方法。该方法首先是在结构动力学试验结果和结构有限元模型计算分析结果的基础上，根据设计要求、灵敏度计算或工程经验选择适合的待修正参数、修正范围来确定修正样本空间，并给出样本点，其次采用增广SVM方法构造每组样本点和与之对应的目标函数之间的代理模型，采用基于Pareto最优解的多目标优化方法，以代理模型输出为目标，样本空间为变量，寻找待修正参数在修正区间内的全局最优解。用代理模型代替原有的有限元模型进行相关的计算分析，避免在模型修正过程中反复调用原有限元模型进行计算带来的高昂计算成本。通过算例一表明，增广SVM的预测结果较传统SVM方法精度更高，而算例二、三则说明本文所提出的基于增广SVM方法的结构动力学模型修正方法具有实际应用价值，同时计算结果具有很高的精度。

Abstract

In this paper, the model updating method based on surrogate model is presented. A support vector machine (SVM) based on hybrid basis function is proposed for solving the over fitting results while SVM is used to deal with weak nonlinear function. According to the dynamic experimental measured results and finite element calculated results, suitable modified parameters and modified ranges were chosen based on design requirements, sensitivity analysis results or engineering experiences, and the suitable design of experiment method (DOE) was used to choose sample points based on modified parameters and modified ranges. Then, the surrogate model for each group of sample points and corresponding objective function using SVM based on hybrid basis functions was constructed, and the multi-objective optimization algorithm is introduced to obtain the global optimal solution. It appears that surrogate model using SVM based on hybrid basis functions does a higher precision than normal SVM, and it can be used in practical application.

导出引用

陈喆，何欢，陈国平. 基于增广SVM的结构动力学模型修正方法研究[J]. 振动与冲击, 2017, 36(15): 194-202

CHEN Zhe，HE Huan，CHEN Guo-ping. The research of model updating of using support vector machine based on hybrid basis functions[J]. Journal of Vibration and Shock, 2017, 36(15): 194-202

参考文献

[1] Mottershead J E, Friswell M I. Model updating in structural dynamics: a survey[J]. Journal of sound and vibration, 1993, 167(2): 347-375.
[2] Brownjohn J M W, Dumanoglu A A, Severn R T. Ambient vibration survey of the Fatih Sultan Mehmet (Second Bosporus) suspension bridge[J]. Earthquake engineering & structural dynamics, 1992, 21(10): 907-924.
[3] RODDEN W P. A method for deriving structural influence coefficients from ground vibration tests[J]. AIAA journal, 1967, 5(5): 991-1000.
[4] Berman A, Flannelly W G. Theory of incomplete models of dynamic structures[J]. AIAA journal, 1971, 9(8): 1481-1487.
[5] Balci O．Verification validation and accreditation for simulation models[C]．Proceedings of the 29th Conference on Winter Simulation，1997.
[6] Aeschliman D P，Oberkampf W L．Experimental methodology for computational fluid dynamics code validation[R]．Sandia National Laboratories，Albuquerque，NM，1997．
[7] Khuri AI, Cornell JA. Response surface designs and analysis[M]. Marcel Dekker, Inc,1987.
[8] Myers R H, Montgomery D C, Anderson-Cook C M. Response surface methodology: process and product optimization using designed experiments[M]. John Wiley & Sons, 2016.
[9] Marwala T. Finite-element-model Updating Using the Response-surface Method[J]. Finite-element-model Updating Using Computional Intelligence Techniques: Applications to Structural Dynamics, 2010: 103-125.
[10] Krishnamurthy T. Comparison of response surface construction methods for derivative estimation using moving least squares, kriging and radial basis functions[C]//Proceedings of the 46th AIAA/ASME/ASCE/AHS/ASC structures, structural dynamics and materials conference, AIAA-2005-1821. 2005: 18-21.
[11] Kumer, S. Neural Networks[M]. Beijing: Tsinghua University Press. 2006.
[12] Buhmann, M.D. Radial Basis Functions:Theory and Implementations[M], Cambridge University Press, Cambridge, UK. 2003.
[13] WIM C M Van Beers, Jack P C Kleijnen. Kriging interpolation in simulation: a survey[C]. Simulation Conference 2004. Proceedings of the 2004 Winter Simulation Conference：107-115.
[14] Cressie, N.A.C. Statistics for Spatial Data[M], revised edition. New York:Wiley. 1993.
[15] Gunn S R. Support vector machines for classification and regression[J]. ISIS technical report, 1998, 14.
[16] Vapink, V. Statistical learning theory[M], New York: Wiley.1998.
[17] Clarke S M, Griebsch J H, Simpson T W. Analysis of support vector regression for approximation of complex engineering analyses[J]. Journal of mechanical design, 2005, 127(6): 1077-1087.
[18] Ayestaran, R.G., Heras, F.L. Support vector regression for the design of array antennas. IEEE Antennas and Wireless Propagation Letters. 2005, 4:414-416
[19] Yun Y, Yoon M, Nakayama H. Multi-objective optimization based on meta-modeling by using support vector regression, Optimization and Engineering. 2009, 10(2):167-181.
[20] Saqlain, A., He, L.S. Support vector regression-driven multidisciplinary design optimization for multi-stage space launch vehicle considering throttling effect. 44th AIAA. Aerospace Sciences Meeting. 2006, 6:4089-4102.
[21] Wang, B.P., Divija, O., Lee, Y.J. Structural optimization using FEMLAB and smooth support vector regression. 48th AIAA/ASME/ASCE/AHS/ASC Structures, Structural Dynamics, and Materials Conference. 2007, 3:2568-2577.
[22] Qazi M U D, He L, Mateen P. Hammersley sampling and support-vector-regression-driven launch vehicle design[J]. Journal of Spacecraft and Rockets, 2007, 44(5): 1094-1106.
[23] Wang H, Li E, Li G Y. The least square support vector regression coupled with parallel sampling scheme metamodeling technique and application in sheet forming optimization[J]. Materials & Design, 2009, 30(5): 1468-1479.
[24] 朱跃，张令弥，郭勤涛。基于SVR 方法的有限元模型确认中不确定性建模研究[J]。振动与冲击，2010,18(3):190-196.
ZHU Yue, ZHANG Ling-mi, GUO Qin-tao. SVR based uncertainty modeling m ethod for structures finite element model validation[J]. Journal of Vibration and Shock, 2010,18(3):190-196.
[25] 朱跃，张令弥。基于确定性计算响应面的复杂工程结构不确定性建模研究[J]。计算力学学报，2011,28(3):412-416.
ZHU Yue, ZHANG Ling-mi. Deterministic computer simulation response surface research on complex engineer structure uncertainty[J]. Chinese Journal of Computational Mechanics, 2011,28(3):412-416.
[26] 费庆国, 张令弥, 李爱群, 郭勤涛. 基于统计分析技术的有限元模型修正研究[J]. 振动与冲击, 2005, 24(3): 23-26.
Fei Qingguo, Zhang Lingmi, Li Aiqun, Guo Qintao. Finite element model updating using statistics analysis[J]. Journal of vibration and shock, 2005,24(3),23-26.
[27] 张冬冬, 郭勤涛. Kriging响应面代理模型在有限元模型确认中的应用[J]. 振动与冲击, 2013, 32(9): 187-191.
Zhang Dongdong, Guo Qintao. Application of Kriging response surface in finite element model validation[J]. Journal of vibration and shock, 2013, 32(9): 187-191.
[28] Ren W. X., Chen H. B. Finite element model updating in structural dynamics by using response surface method[J]. Engineering Structures, 2010, 32(8): 2455-2465.
[29] V.N.Vapnik. An overview of statistical learning theory. IEEE Transactions on Neural Networks,1999,10(5):988-999