Search papers, labs, and topics across Lattice.
The paper introduces Ground-Motion Flow (GMFlow), a novel physics-inspired latent operator flow matching framework for generating realistic, large-scale regional ground-motion time-histories conditioned on physical parameters. GMFlow achieves a 10,000x speedup compared to traditional physics-based simulations by operating in a latent space and utilizing flow matching for generative modeling. Validated on earthquake scenarios in the San Francisco Bay Area, GMFlow generates spatially coherent ground motion across 9 million grid points in seconds, enabling rapid uncertainty-aware hazard assessment.
Simulate earthquake ground motion 10,000x faster with a new latent operator flow matching method, opening the door to real-time risk assessment for critical infrastructure.
Earthquake hazard analysis and design of spatially distributed infrastructure, such as power grids and energy pipeline networks, require scenario-specific ground-motion time histories with realistic frequency content and spatiotemporal coherence. However, producing the large ensembles needed for uncertainty quantification with physics-based simulations is computationally intensive and impractical for engineering workflows. To address this challenge, we introduce Ground-Motion Flow (GMFlow), a physics-inspired latent operator flow matching framework that generates realistic, large-scale regional ground-motion time-histories conditioned on physical parameters. Validated on simulated earthquake scenarios in the San Francisco Bay Area, GMFlow generates spatially coherent ground motion across more than 9 million grid points in seconds, achieving a 10,000-fold speedup over the simulation workflow, which opens a path toward rapid and uncertainty-aware hazard assessment for distributed infrastructure. More broadly, GMFlow advances mesh-agnostic functional generative modeling and could potentially be extended to the synthesis of large-scale spatiotemporal physical fields in diverse scientific domains.
