Efficient Characterization of Dynamic Response Variation Using Multi-Fidelity Data Fusion through Composite Neural Network
Efficient Characterization of Dynamic Response Variation Using Multi-Fidelity Data Fusion through Composite Neural Network
复制标题
DOI：
10.1016/j.engstruct.2021.111878
复制
发表时间：
2020-05
期刊：
影响因子：
--
通讯作者：
K. Zhou;Jiong Tang
中科院分区：
文献类型：
--
作者：
K. Zhou;Jiong Tang
结构中的不确定性是不可避免的，这通常会导致动力响应预测的变化。对于复杂的结构，响应变化分析的强力蒙特卡罗模拟是不可行的，因为一次运行可能已经计算成本很高。因此，已经探索了数据驱动的元建模方法，以促进有效的仿真和统计推断。Ameta模型的性能取决于训练数据集的质量和数量。然而，在实际应用中，从高维有限元模拟或实验中获得的高保真数据普遍很少，这对建立有限元模型提出了很大的挑战。在本研究中，我们利用了结构动力分析中的多层次响应预测机会，即从降阶建模中快速获取大量低保真数据，从全尺寸有限元分析中准确获取少量高保真数据。具体地说，我们提出了一种能够充分利用所获得的多层次、异质数据集的复合神经网络融合方法。它隐含地识别低保真和高保真数据集的相关性，与最先进的数据集相比，精度有所提高。以频率响应变化表征为例进行了全面的研究，以验证该方法的性能。
Uncertainties in a structure is inevitable, which generally lead to variation in dynamic response predictions. For a complex structure, brute force Monte Carlo simulation for response variation analysis is infeasible since one single run may already be computationally costly. Data drivenmeta-modeling approaches have thus been explored to facilitate efficient emulation and statistical inference. The performance of ameta-model hinges upon both the quality and quantity of training dataset. In actual practice, however, high-fidelity data acquired from high-dimensional finite element simulation or experiment are generally scarce, which poses significant challenge tometa-model establishment. In this research, we take advantage of the multi-level response prediction opportunity in structural dynamic analysis, i.e., acquiring rapidly a large amount of low-fidelity data from reduced-order modeling, and acquiring accurately a small amount of high-fidelity data from full-scale finite element analysis. Specifically, we formulate a composite neural network fusion approach that can fully utilize the multi-level, heterogeneous datasets obtained. It implicitly identifies the correlation of the low- and high-fidelity datasets, which yields improved accuracy when compared with the state-of-the-art. Comprehensive investigations using frequency response variation characterization as case example are carried out to demonstrate the performance.