Abstract: The digital-twin system for the internal mechanical state of Carbon Fiber Reinforced Polymer (CFRP) wing structure, which serves as an effective means for achieving structural health monitoring of CFRP components. By using the real-time mapping between the physical space and the digital space, the dynamic human-machine interaction during the service life of critical structures can be achieved. Addressing the difficult issues in clear the internal mechanical state of CFRP structure during actual service based on four key parts: CFRP wing preparation, mechanical simulation, construction of radial basis function (RBF) surrogate model, and unidirectional data reduction processing. Then, a novel method for constructing digital-twin framework of the internal mechanical state of CFRP wings based on high fidelity proxy model is proposed. Taking the CFRP wing structure as a case study, high-fidelity data obtained from sensors and low-fidelity data derived from simulation are employed in reduced-order processing with one-way reduced-order model, and the RBF surrogate model is utilized to interpolate the stress values at each node. Ultimately, these data are used to train the surrogate model, which then repairs and reconstructs sample points for the wing's digital twin. It is helpful to facilitate the construction of a digital-twin system for the internal mechanical state of CFRP wing, enabling real-time mapping between the physical CFRP wing and its twin model.

Key words: digital twin, internal mechanical state, high-fidelity surrogate model, CFRP wing structure