Physics-guided convolutional neural network (PhyCNN) for data-driven seismic response modeling

• Published in: Engineering structures Vol. 215; p. 110704
• Main Authors: Zhang, Ruiyang, Liu, Yang, Sun, Hao
• Format: Journal Article
• Language: English
• Published: Kidlington Elsevier Ltd 15-07-2020; Elsevier BV
• Subjects: Algorithms; Artificial neural networks; Cluster analysis; Clustering; Computer simulation; Constraint modelling; Datasets; Deep learning; Earthquakes; Fragility; K-means clustering; Machine learning; Mathematical models; Neural networks; Numerical models; Performance evaluation; PhyCNN; Physics; Physics-guided convolutional neural network; Predictions; Robustness (mathematics); Seismic activity; Seismic response; Seismic response prediction; Serviceability assessment; Training; Vector quantization; Deep learning; Serviceability assessment; Seismic response prediction; K-means clustering; PhyCNN; Physics-guided convolutional neural network
• ISSN: 0141-0296; 1873-7323
• DOI: 10.1016/j.engstruct.2020.110704

•A deep physics-guided convolutional neural network (PhyCNN) is developed for structural seismic response estimation.•Available physics 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 proposed approach was successfully demonstrated by both numerical and experimental case studies.
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.