Regional-scale nonlinear structural seismic response prediction by neural network.

• An end-to-end neural network-based framework for predicting seismic responses of building clusters. • The framework adopts the idea of embedding to fully characterize the earthquake and structure features and their interaction processes. • The framework applies to different prediction tasks such as pre-earthquake or post-earthquake assessments, and engineering scenarios with highly limited training data. • The framework achieves a balance of accuracy and efficiency with the representation of embeddings. Urban seismic damage assessment has recently become an emerging research topic due to the accelerating global urbanization trend, to which the seismic responses of building clusters to various earthquakes are a prerequisite. Traditional methods for this task, including vulnerability analysis and time-consuming time history analysis, may suffer from accuracy or efficiency problems especially for nonlinear response calculation. Machine learning methods allow for rapid and accurate response prediction, but current applications still lack scalability (on the size of structures or earthquakes) and the corresponding real datasets. To tackle this issue, this paper proposes an artificial neural network framework for simultaneously predicting nonlinear seismic responses of all buildings in a cluster subjected to multi-earthquake inputs. Inspired by the advanced collaborative filtering techniques, the framework converts the regional response prediction into a matrix completion problem, thereby aggregating information extracted from historical response records and physical characteristics to improve performance. The framework is used to assess nonlinear responses of a real urban region consisting of 2788 buildings subjected to 3798 measured ground motions. The results clearly demonstrated that the proposed framework achieves orders of magnitude faster than time history analysis and average errors below 3 % on several response metrics, showing high computational efficiency and accuracy. [ABSTRACT FROM AUTHOR]