ULTRASONIC GUIDED WAVE DAMAGE MONITORING TECHNOLOGY FOR COMPOSITE STRUCTURES DRIVEN BY ARTIFICIAL INTELLIGENCE

Composite structures have been extensively applied in critical equipment within aerospace and other engineering industries. The timely detection of damage and failure is essential for preventing sudden damage and structural failure. It is urgent to explore the theory and method for quantitative monitoring of damage in composite structures in complex service environment. Aiming at the limitations of the existing structural health monitoring technology, such as low damage identification accuracy, weak cross-domain generalization, and poor performance under small-sample conditions, this paper summarizes the solutions provided by the author's team in the field of ultrasonic guided wave damage monitoring of composite structures driven by artificial intelligence. Firstly, aiming at the challenge of feature extraction from ultrasonic guided wave signals in complex composite structures, a deep learning-based multi-source information feature extraction and fusion method is constructed to achieve high-precision identification of damage location and size in composite structures. Secondly, to address the challenge of poor model generalization performance in different monitoring areas, different sensor layouts and different structures, a cross-domain damage identification method based on transfer learning and domain adaptation is proposed, and the effective transfer and robust characterization of damage features are realized by combining parameter transfer with feature distribution alignment. Then, in view of the scarcity of samples, a data expansion strategy driven by generation model is introduced, and damage monitoring research for limited samples is developed by integrating the digital twin method to improve the prediction performance. The research results of the author's team demonstrate that the proposed method has high accuracy and robustness in damage localization and quantification assessment of composite structures, which provides theoretical basis and technical approach for structural health monitoring of composite structures in complex service environment. Finally, the main challenges of aircraft composite structural health monitoring in actual service environment are analyzed, and the future development trend of aircraft composite structural health monitoring technology is prospected.