Balanced fidelity digital twin for structural damage monitoring based on Lamb wave multilevel feature enhancement and adaptive space interaction

As structural damage patterns and service environments become increasingly complex, digital twin (DT)-based structural health monitoring, with its unique advantages, can compensate for the limitations of data-driven methods regarding data dependency and model interpretability. However, it still faces challenges in modeling complexity, simulation accuracy, and the discrepancies between real and virtual features. This study proposes a balanced fidelity DT for structural damage monitoring based on Lamb wave multilevel feature enhancement and adaptive space interaction. Firstly, multilevel refined features are extracted from few-shot guided wave signals obtained in physical and digital space, and the adversarial synthetic balancing algorithm is proposed for feature enhancement. Additionally, the learning phase of the damage monitoring model based on the feature-mapping convolutional network is driven by virtual samples of readily accessible balanced fidelity in digital space. To reduce the feature distribution difference between the two spaces, an interactive transfer strategy is introduced to establish a shared DT feature space. The model is further updated by accurate but elusive experimental data from the physical space, making it adaptable for damage diagnosis tasks of real structures in the shared space. This method reduces the dependence on experimental data and the cost of simulations, without compromising prediction accuracy. The case study shows that the damage localization accuracy of this DT system is increased by 22.67–30.55% and the quantification accuracy by 14.23–17.11% compared to a single space. Overall, this study provides a feasible technique to enhance the accessibility and generalizability of DTs for real engineering structures.