| 摘要: |
| 风洞模型支撑机构有限元建模对预测风洞试验中的结构动力学响应至关重要。然而采用模态试验修正有限元模型需要多轮迭代优化，随着修正参数的增多，数值优化耗时将呈指数化增长。为解决上述问题，联合ABAQUS、MATLAB搭建iSIGHT结构有限元模型修正平台，基于此平台开展某模型支撑机构有限元动力学模型修正，根据试验设计变量对响应的贡献量，筛选灵敏度较高的修正变量；构建前４阶频率差和振型相关性为多目标函数，使用近似建模得到修正变量和目标函数的响应面模型，采用多目标优化方法NSGA-II开展模型修正；修正后的模型前４阶频率差均在10%以内，振型相关性均大于0.8。对修正模型开展动力学响应确认，使用结构模态动力学响应与锤击试验响应作对比，结果显示，修正模型满足工程需求，可用作下一步复杂载荷作用下的结构动力学响应预测。 |
| 关键词: 有限元模型修正 近似建模 多目标优化 模型支撑机构 模型确认 |
| DOI： |
|
| 基金项目:中国空气动力研究与发展中心基础与前沿技术研究基金(PJD20200224) |
|
| Finite Ｅlement Ｍodel Ｕpdating and Ｖalidating for Ｍodel Ｓupport Ｍechanism of Ｗind Ｔunnel |
| MA Yueyin,HAO Peng,MA Bin,CHEN Wanhua |
| (China Aerodynamics Research and Development Center;State Key Laboratory of Structural Analysis for Industrial Equipment，Department of Engineering Mechanics，Dalian University of Technology) |
| Abstract: |
| Accurate finite element model of wind tunnel model support mechanism is important to predict the dynamic response of the model in wind tunnel test. With the increase of the modified parameters, the time consumption of numerical optimization will increase exponentially. In order to solve this problem, the computer-aided optimization software iSIGHT, ABAQUS and MATLAB are used to establish a platform of the structural finite element model updating, which can automatically integrate the flow of components. Based on this platform, the finite element dynamic model updating of a model support system is carried out, and the updating variables are selected according to the contribution of the variables to the response; the first four mode frequency differences and mode shape correlations are constructed as multi-objective functions, and the response surface models of the updating variables and objective functions are obtained by using approximate modeling method, and the model updating is conducted by using a variety of multi-objective optimization methods; the first four updated mode frequency differences are less than 10%, and the correlations of the mode shapes are greater than 0.8. The results show that the updated model meets the engineering requirements and can be used to predict the dynamic response of the structure under the complex load condition. |
| Key words: Ｆinite element model updating Ａpproximate modeling Ｍulti-objective optimization Ｍodel support system Ｍodel validating |