Physics-guided convolutional neural network (PhyCNN) for data-driven seismic response modeling
Physics-guided convolutional neural network (PhyCNN) for data-driven seismic response modeling
复制标题
DOI：
10.1016/j.engstruct.2020.110704
复制
发表时间：
2020-07-15
影响因子：
5.5
通讯作者：
Sun, Hao
中科院分区：
文献类型：
--
作者：
Zhang, Ruiyang;Liu, Yang;Sun, Hao
准确预测建筑物在地震作用下的反应为评估建筑物的性能提供了可能。为此，我们利用深度学习的最新进展，开发了一种用于数据驱动结构地震响应建模的物理制导卷积神经网络(PhyCNN)。其概念是基于有限的地震输入输出数据集(例如，来自模拟或传感)和物理约束来训练深度PhyCNN模型，从而建立结构响应预测的代理模型。可用的物理(例如，动力学定律)可以对网络输出提供约束，减轻过拟合问题，减少对大训练数据集的需要，从而提高训练模型的稳健性，以实现更可靠的预测。然后，在给定一定的极限状态准则的情况下，利用代理模型进行脆弱性分析。此外，还提出了一种基于K-均值聚类的无监督学习算法，将数据集划分为训练类、验证类和预测类，以最大限度地利用有限的数据集。通过数值算例和实验算例验证了PhyCNN的性能。令人信服的结果表明，PhyCNN能够以数据驱动的方式准确地预测建筑物的地震响应，而不需要基于物理的分析/数值模型。PhyCNN范例的性能也优于非物理引导的神经网络。
Accurate prediction of building's response subjected to earthquakes makes possible to evaluate building performance. To this end, we leverage the recent advances in deep learning and develop a physics-guided convolutional neural network (PhyCNN) for data-driven structural seismic response modeling. The concept is to train a deep PhyCNN model based on limited seismic input-output datasets (e.g., from simulation or sensing) and physics constraints, and thus establish a surrogate model for structural response prediction. Available physics (e.g., the law of dynamics) can provide constraints to the network outputs, alleviate overfitting issues, reduce the need of big training datasets, and thus improve the robustness of the trained model for more reliable prediction. The surrogate model is then utilized for fragility analysis given certain limit state criteria. In addition, an unsupervised learning algorithm based on K-means clustering is also proposed to partition the datasets to training, validation and prediction categories, so as to maximize the use of limited datasets. The performance of PhyCNN is demonstrated through both numerical and experimental examples. Convincing results illustrate that PhyCNN is capable of accurately predicting building's seismic response in a data-driven fashion without the need of a physics-based analytical/numerical model. The PhyCNN paradigm also outperforms non-physics-guided neural networks.