面向飞机结构实时服役状态监测需求，基于柔性传感、5G无线传输与数据融合技术，提出了一种实时采集数据驱动的数字孪生模型构建方法。首先，研制了以聚酰亚胺为柔性基底的阵列化柔性传感器，采用单片机作为轻量化数据采集设备，结合5G无线传输技术实现结构离散测点应变的高效高精度无线采集。然后，通过参数化模型生成仿真数据，训练了基于本征正交分解的降阶模型，实现离散测点应变到全场应力等力学响应的快速重构。最后，采用桥函数方法融合离散测量数据与全场仿真数据，构建了实时高精度数字孪生模型，提高了全场力学响应预测精度。开展飞机典型壁板结构试验验证，结果表明：提出方法从数据采集到数字孪生模型可视化显示的平均时延为0.5093s，对von Mises等效应力的预测精度为94.26%，建立的数字孪生模型能实现对结构强度实时、高精度监测。

Facing the demand for real-time service condition monitoring of aircraft structures, a real-time acquisition data-driven construction method for digital twin models is proposed by integrating flexible sensing, 5G wireless transmission and data fusion technologies. Firstly, an arrayed flexible sensor with polyimide as the flexible substrate is developed. Taking a single-chip microcontroller as the lightweight data acquisition terminal, combined with 5G wireless transmission technology, efficient and high-precision wireless collection of strain data at discrete structural measuring points is realized. Secondly, parametric finite element modeling is adopted to generate simulated datasets, and a proper orthogonal decomposition (POD)-based reduced-order model is trained to rapidly reconstruct full-field mechanical responses such as structural stress from discrete measured strains. Finally, the bridge function method is employed to fuse discrete measured data and full-field simulated data, constructing a real-time and high-precision digital twin model and effectively improving the prediction accuracy of full-field mechanical responses. Experimental verification is carried out on a typical aircraft stiffened panel structure. The results show that the average latency from data acquisition to visual display of the digital twin model is 0.5093 s, and the prediction accuracy of von Mises equivalent stress reaches 94.26%. The established digital twin model can realize real-time and high-precision monitoring of aircraft structural strength.