摘要
随着混合模拟试验中子结构自由度和数量及其更新参数的增加,混合模拟试验模型更新中用于寻找更优参数的样本点数量及计算量均会明显增加。从试验设计角度出发,提出基于均匀设计的模型更新混合模拟方法。该方法在试验子结构的基础上,利用均匀设计构造一个验算子结构样本空间,以恢复力为目标,从中找出最符合试验子结构的数值模型作为数值结构本构参数更新的来源,有效提升识别效率。考虑钢材不同本构模型及试验子结构边界处弯矩处理情况,以2层单跨钢框架为例进行了四种类型数值模拟分析。模型更新混合模拟试验采用Matlab与OpenSees混合编程,软件数据交互全采用Socket接口,效率比传统文本读写交互得到有效提升。结果表明,该方法适用于不同边界处理情况,极大地降低了模型更新计算时间,且有效提升了混合模拟试验的模拟精度。
With increase in DOFs, substructure number and updating parameters in hybrid simulation tests, sample points and computational amount needed to search more optimal parameters for model updating hybrid simulation tests can obviously increase. Here, from the perspective of testing design, a model updating hybrid simulation test method based on uniform design was proposed. Based on tested substructures, the uniform design was used to construct a sample space of checking substructures. Taking restoring force as an objective, the most suitable numerical model of checking substructures for tested substructures was found as the source of updating constitutive parameters of the studied numerical structure to effectively improve the identification efficiency. Considering different constitutive models of steel products and bending moment processing situations at boundaries of tested substructures, a single-span two-story steel frame structure was taken as an example to do 4 types of numerical simulations. The model updating hybrid simulation tests adopted Matlab-OpenSees mixed programming, and the software data interaction adopted socket interface. The efficiency was improved effectively compared with the traditional text read-write interaction. Results showed that the proposed method is suitable for different boundary processing situations;the model updating calculation time is reduced greatly, and the accuracy of the proposed hybrid simulation tests is improved effectively.
作者
陈再现
钟炜彭
李秦鸣
CHEN Zaixian;ZHONG Weipeng;LI Qinming(College of Civil Engineering,Harbin Institute of Technology(Weihai),Weihai 264209,China)
出处
《振动与冲击》
EI
CSCD
北大核心
2020年第21期9-16,共8页
Journal of Vibration and Shock
基金
国家自然科学基金项目(51678199)。
