A Multi-Fidelity Convolutional Autoencoder-Transfer Learning Framework for Guided-Wave-Based Damage Diagnosis Using Large Simulated and Limited Experimental Datasets
arXiv:2606.27304v1 Announce Type: new Abstract: Guided wave-based structural health monitoring (GWSHM) with onboard transducers offers significant potential for the early diagnosis of damage in engineering structures. However, the practical deployment of deep learning models is often hindered by the limited availability of labelled experimental data and the high computational cost of generating large-scale high-fidelity simulation datasets. This study presents a multifidelity transfer learning framework that integrates lightweight physics-based simulations, convolutional autoencoder (CAE)-base
The increasing complexity and aging of critical infrastructure, coupled with advancements in AI and simulation capabilities, are driving the need for more efficient and robust structural health monitoring solutions.
This development offers a pathway to significantly improve the reliability and reduce the maintenance costs of essential engineering structures like aircraft, bridges, and power plants through more effective damage diagnosis.
The proposed framework enables the practical deployment of deep learning models for structural health monitoring by effectively bridging the gap between abundant simulated data and scarce experimental data, making AI solutions more accessible.
- · Aerospace industry
- · Civil engineering firms
- · AI/ML solution providers
- · Infrastructure operators
- · Traditional inspection methods
- · Companies reliant on solely experimental data
Reduced catastrophic failures and extended lifespan of critical infrastructure components.
Lower insurance premiums and operational costs for industries managing large-scale physical assets.
Acceleration of autonomous inspection systems and 'self-healing' infrastructure concepts facilitated by continuous, precise damage detection.
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