Integrating Fourier Neural Operators with Diffusion Models to improve Spectral Representation of Synthetic Earthquake Ground Motion Response

2025-04-01

AI Summary

In nuclear facility seismic design, high-fidelity synthetic ground motion generation remains challenging in the absence of recorded strong-motion data. Method: This paper proposes a synergistic modeling framework integrating Fourier Neural Operators (FNOs) and Denoising Diffusion Probabilistic Models (DDPMs). The FNO captures source–site elastic dynamic response, ensuring physical interpretability, while the DDPM specifically corrects its mid-frequency spectral attenuation bias. The method unifies data-driven refinement with physics-based constraints by incorporating an approximate seismic Green’s operator and spectral goodness-of-fit (GOF) evaluation. Contribution/Results: Experiments demonstrate significant reduction in mid-frequency spectral distortion, improved GOF scores, rapid inference capability, and strong generalization across diverse sites and source mechanisms. This work presents the first synergistic neural-operator–diffusion architecture for ground motion synthesis, establishing a novel paradigm for seismic safety assessment of nuclear power plants in record-scarce scenarios.

Technology Category

• Search and Optimization: Mixed Discrete/Continuous Search
• Natural Language Processing: Code Generation / Program Synthesis from Natural Language
• Reasoning under Uncertainty: Stochastic Optimization

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Web Mining and Content Analysis: Web data generation and simulation

Abstract

Nuclear reactor buildings must be designed to withstand the dynamic load induced by strong ground motion earthquakes. For this reason, their structural behavior must be assessed in multiple realistic ground shaking scenarios (e.g., the Maximum Credible Earthquake). However, earthquake catalogs and recorded seismograms may not always be available in the region of interest. Therefore, synthetic earthquake ground motion is progressively being employed, although with some due precautions: earthquake physics is sometimes not well enough understood to be accurately reproduced with numerical tools, and the underlying epistemic uncertainties lead to prohibitive computational costs related to model calibration. In this study, we propose an AI physics-based approach to generate synthetic ground motion, based on the combination of a neural operator that approximates the elastodynamics Green's operator in arbitrary source-geology setups, enhanced by a denoising diffusion probabilistic model. The diffusion model is trained to correct the ground motion time series generated by the neural operator. Our results show that such an approach promisingly enhances the realism of the generated synthetic seismograms, with frequency biases and Goodness-Of-Fit (GOF) scores being improved by the diffusion model. This indicates that the latter is capable to mitigate the mid-frequency spectral falloff observed in the time series generated by the neural operator. Our method showcases fast and cheap inference in different site and source conditions.

Problem

• Improving synthetic earthquake ground motion spectral representation
• Reducing computational costs in earthquake model calibration
• Enhancing realism of synthetic seismograms with AI

Innovation

• Combines Fourier Neural Operators with Diffusion Models
• Corrects seismograms using denoising diffusion probabilistic model
• Enhances realism and reduces frequency biases efficiently