TimesNet-Gen: Deep Learning-based Site Specific Strong Motion Generation

2025-12-04

AI Summary

This study addresses the problem of site-specific strong ground motion generation. We propose TimesNet-Gen, a time-domain conditional generative model that— for the first time—incorporates an implicit site bottleneck into an end-to-end time-series generation framework, enabling high-fidelity seismic acceleration time history synthesis driven by observed accelerograms. Our method integrates conditional generative adversarial learning with latent-space regularization and introduces a site-specificity scoring mechanism based on confusion matrices of fundamental frequency (f₀) distributions to quantitatively evaluate site-identifiability consistency in generated data. Experiments demonstrate that the synthesized time histories closely match recorded data in both HVSR spectral shape and f₀ distribution. Site-level alignment accuracy significantly surpasses baseline methods such as spectrogram-based VAEs. The proposed framework provides an interpretable and verifiable generative paradigm for fine-grained local site seismic risk assessment.

Technology Category

• Natural Language Processing: Generation
• Machine Learning: Deep Generative Models & Autoencoders
• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems

Application Category

• Web Mining and Content Analysis: Web data generation and simulation
• Social Networks and Social Media: Generative AI / large language models and their impact on social systems
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs

Abstract

Effective earthquake risk reduction relies on accurate site-specific evaluations. This requires models that can represent the influence of local site conditions on ground motion characteristics. In this context, data driven approaches that learn site controlled signatures from recorded ground motions offer a promising direction. We address strong ground motion generation from time-domain accelerometer records and introduce the TimesNet-Gen, a time-domain conditional generator. The approach uses a station specific latent bottleneck. We evaluate generation by comparing HVSR curves and fundamental site-frequency $f_0$ distributions between real and generated records per station, and summarize station specificity with a score based on the $f_0$ distribution confusion matrices. TimesNet-Gen achieves strong station-wise alignment and compares favorably with a spectrogram-based conditional VAE baseline for site-specific strong motion synthesis. Our codes are available via https://github.com/brsylmz23/TimesNet-Gen.

Problem

• Generates site-specific strong ground motions from accelerometer records.
• Uses a station-specific latent bottleneck for localized condition modeling.
• Evaluates accuracy via HVSR curves and fundamental site-frequency distributions.

Innovation

• Time-domain conditional generator for earthquake motion
• Station-specific latent bottleneck for site conditions
• Evaluation using HVSR curves and frequency distributions