arXiv:2606.20015v1 Announce Type: new Abstract: Long-span roadway bridges exhibit highly localized structural responses under vehicular loading, making repeated FE analysis computationally expensive for applications such as influence surface generation and structural digital twins. Existing SciML approaches struggle to accurately capture these localized responses. To address this challenge, this study proposes an adaptive-trunk DeepONet for localized structural response prediction in large-scale bridge systems. The framework dynamically constructs a load-dependent learning domain using a KNN s
The continuous development in AI, particularly deep learning and scientific machine learning (SciML), is enabling more sophisticated applications in engineering previously limited by computational costs.
This development represents a significant step towards more efficient and accurate infrastructure monitoring and digital twinning, crucial for maintaining complex systems like long-span bridges.
The ability to accurately predict highly localized structural responses in large-scale infrastructure using AI reduces computational expense and enhances the feasibility of real-time monitoring and predictive maintenance.
- · Civil Engineering firms
- · Infrastructure owners and operators
- · AI/ML model developers
- · Smart City initiatives
- · Traditional structural analysis software providers reluctant to integrate AI
- · Infrastructure maintenance companies relying solely on periodic manual inspectio
Improved safety and extended lifespan of critical infrastructure through better monitoring and predictive maintenance.
Reduced operational costs for maintaining large-scale structures and enhanced resilience against unexpected events.
Acceleration of digital twinning for physical assets across various industries, creating a more interconnected and intelligently managed built environment.
This signal links to a primary source. Continuum Brief monitors and indexes it as part of the live intelligence stream — we do not republish source content.
Read at arXiv cs.LG