建立准确的结构动力学模型是结构响应分析的基础，由于模型简化的不确切等因素，必然会带来一定的误差，为了获得高精度的动力学分析模型，需要结合试验数据对模型进行修正。模态试验结果中包含了试件不同状态不同阶次的频率和振型信息，模型修正时需要建立多个目标函数，提出了一种基于动态加权系数的多目标模型修正方法。通过对解的群体实施进化，在每一代非劣解中，挑选各个子目标函数的局部最优解，计算各个局部最优解与子目标期望值的差距，并根据差距对加权系数动态调整，从而在进化过程中对加权系数进行优化，避免维数灾难问题，实现各个子目标函数的快速收敛。采用该方法对导弹全弹动力学模型进行了修正，子目标函数个数达到16个，与基于Pareto最优的模型修正方法相比，用较少的代数实现了各个子目标函数的收敛，提高了群体搜索的效率，取得了较好的修正效果。
Abstract
The establishment of an accurate structural dynamic model is the basis for structural response analysis.Due to inaccurate model simplification and other factors, it will inevitably bring certain errors.In order to obtain a high-accuracy model, it needs to be modified in conjunction with experimental data.Modal test results usually contain frequency and mode shape information of different orders of different states of the test piece; thus, multiple objective functions need to be established.A multi-objective model updating method based on dynamic weighting coefficients was proposed, through the evolution of the solution group, in each generation of non-inferior solutions, the local optimal solution of each sub-objective function was selected, the gap between each local optimal solution and the expected value was calculated, and the weighting coefficient was dynamically adjusted according to the difference.In the course of evolution, the weighting coefficients were optimized to avoid the dimension disaster problem and realize rapid convergence of each sub-objective function.This method was used to update a missile’s dynamic model.The number of sub-objective functions reaches 16.Compared with the Pareto-optimal model updating method, the convergence of each sub-objective function was realized with few algebras, and the efficiency of group search is improved and a better updating effect is achieved.
关键词
模型修正 /
多目标进化 /
进化算法 /
加权系数
{{custom_keyword}} /
Key words
model updating /
multi-objective evolution /
evolutionary algorithm /
weighting coefficient
{{custom_keyword}} /
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] 严海, 马锐磊等. 基于模态参数的水下航行器楔环结构有限元模型修正[J]. 鱼雷技术, 2016, 24(2):87-93.
YAN Hai, MA Rui-lei, et al. Finite Element Model Correction for Wedged-Ring Connection Structure of Underwater Vehicle Based on Modal Parameters[J]. TORPEDO TECHNOLOGY, 2016, 24(2):87-93.,
[2] 宋正华, 姜东等. 含铰可展桁架结构非线性模型修正方法研究[J]. 振动与冲击,2018, 37(1):21-26.
SONG Zhenghua, JIANG Dong, et al. Nonlinear model updating method for deployable trusses with joints[J]. JOURNAL OF VIBRATION AND SHOCK, 2018, 37(1):21-26.
[3] 秦仙蓉, 潘杰等. 塔式起重机结构有限元模型修正的响应面方法[J]. 振动与冲击,2018, 37(6): 244-250.
QIN Xianrong, PAN Jie, et al. Response surface method for the structural finite element model updating of tower cranes[J]. JOURNAL OF VIBRATION AND SHOCK, 2018, 37(6): 244-250.
[4] 袁昭旭, 于开平. 高温环境下结构动力学模型修正方法研究[J]. 振动与冲击,2017, 36(15) :171-180.
YUAN Zhaoxu, YU Kaiping. Structural dynamic model updating under high temperature environment[J]. JOURNAL OF VIBRATION AND SHOCK, 2017, 36(15) :171-180.
[5] 王毅，朱礼文等. 大型运载火箭动力学关键技术及其进展综述[J].导弹与航天运载技术，2000(1):29-37.
Wang Yi, Zhu Liwen, et al. Summary of Some Key Technologies on Dynamics of Large Launch Vehicle[J]. MISSILES AND SPACE VEHICLES, 2000(1):29-37.
[6] 崔逊学.多目标进化算法及其应用[M].北京：国防工业出版社，2006:9-10.
Cui Xun-Xue. Multiobjective Evolutionary Algorithms and their Application[M]. Beijing: National Defense Industry Press，2006:9-10.
[7] 杨 帆，邓斌等. 基于Pareto排序遗传算法的改进型扩张室压力脉动衰减器多目标优化[J].振动与冲击，2018，37(12):1-8.
YANG Fan，DENG Bin，et al. Mult-optimization on improved expansion chamber hydraulic pulsation attenuators using a NSGA[J]. JOURNAL OF VIBRATION AND SHOCK, 2018,37(12):1-8.
[8] 贾洁，万一品等. 基于多目标遗传算法的振动熨平板压实系统参数优化 [J].振动与冲击，2017，36(12):230-235.
JIA Jie ,WAN Yi-pin, et al. Parameter optimization for compaction system of vibration screed of asphalt payer based on multi-objective genetic algorithm[J]. JOURNAL OF VIBRATION AND SHOCK, 2017, 36(12):230-235.
[9] Zhenan He, Gary G. Yen. Many-Objective Evolutionary Algorithms Based on Coordinated Selection Strategy[J]. IEEE Transactions on Evolutionary Computation, 2017, 21(2):220-233.
[10] Zhenhua Li, Qingfu Zhang. A Simple Yet Efficient Evolution Strategy for Large-Scale Black-Box Optimization[J]. IEEE Transactions on Evolutionary Computation, 2018, 22(5):637-646.
[11] Shouyong Jiang, Shengxiang Yang. A Strength Pareto Evolutionary Algorithm Based on Reference Direction for Multi-objective and Many-objective Optimization. IEEE Transactions on Evolutionary Computation[J], 2017, 21(3):329-346.
[12] 王乐，冷德新. 基于Pareto最优的Timoshenko梁模型修正方法[J].导弹与航天运载技术，2015(5):73-76.
Wang Le, Leng De-xin. A Method of Model Correction of Timoshenko Beam Based on Pareto Front[J]. MISSILES AND SPACE VEHICLES, 2015(5):73-76.
[13] 郝武伟，李俊吉. 粒子群算法对高维问题的优化研究[J]. 控制工程, 2018(05): 870-877.
HAO Wu-wei, LI Jun-ji. Study on the High Dimensional Problem Optimization by Particle Swarm Algorithm[J]. Control Engineering of China, 2018(05): 870-877.
[14] A. Cano, C. García-Martínez, S. Ventura. Extremely high-dimensional optimization with MapReduce: Scaling functions and algorithm[J]. Information Sciences, 2017, 415–416: 110-127.
[15] Hong Qian, Yang Yu. Solving High-Dimensional Multi-Objective Optimization Problems with Low Effective Dimensions[C]. The Thirty-First AAAI Conference on Artificial Intelligence (AAAI-17):875:881.
[16] H. Xu, C. Caramanis, S. Mannor. Statistical Optimization in High Dimensions[J]. Operations Research, 2014, 64:958-979.
[17] Y. Ren, Y. Wu. An efficient algorithm for high-dimensional function optimization[J]. Soft Computing, 2013, 17(6):995-1004.
[18] 汤波，范瑞祥等. 悬挂工装对运载火箭模态试验影响分析[J]. 宇航学报,2017,38(12):1354-1360.
TANG Bo, FAN Rui-xiang, et al. Effects of Suspension Rope on Modal Experiment of Rockets[J]. Journal of Astronautics, 2017,38(12):1354-1360.
{{custom_fnGroup.title_cn}}
脚注
{{custom_fn.content}}
PDF(1335 KB)
