Integrating AI in Structural Health Monitoring (SHM): A Systematic Review on Advances, Challenges, and Future Directions

The integration of Artificial Intelligence (AI) in Structural Health Monitoring (SHM) has garnered significant attention in recent years, driven by the need for enhanced safety, reliability, and efficiency in infrastructure management. This systematic review synthesizes the latest advancements in AI techniques applied to SHM, exploring various methodologies, including machine learning, deep learning, and data-driven approaches. We examine a wide range of applications, from real-time damage detection to predictive maintenance and anomaly detection in diverse structural types, including bridges, buildings, and offshore structures. Despite the promising developments, several challenges hinder the widespread adoption of AI in SHM, including data quality and quantity, interpretability of AI models, and integration with existing monitoring systems. We identify critical gaps in the current literature and propose future research directions that emphasize the need for robust algorithms, interdisciplinary collaboration, and the development of standardized protocols. This review serves as a comprehensive resource for researchers and practitioners aiming to advance the integration of AI in SHM, ultimately contributing to safer and more resilient infrastructure systems.近年来，受基础设施管理领域对提升安全性、可靠性与运行效率的需求驱动，人工智能（Artificial Intelligence, AI）在结构健康监测（Structural Health Monitoring, SHM）中的集成应用已受到广泛关注。本系统性综述梳理了人工智能应用于结构健康监测的最新研究进展，探讨了包括机器学习、深度学习以及数据驱动方法在内的多种技术路径。本综述还考察了该领域的多类应用场景，涵盖从实时损伤检测、预测性维护到各类结构（包括桥梁、建筑与近海结构）的异常检测等方向。尽管相关研究已取得可喜进展，但仍有多项挑战制约着人工智能在结构健康监测中的规模化应用，包括数据质量与数量不足、人工智能模型可解释性欠缺，以及与现有监测系统的集成难题等。本综述梳理了当前研究中的关键空白，并提出了未来研究方向，强调亟需研发鲁棒性更强的算法、推动跨学科协作，以及制定标准化操作规范。本综述可为致力于推动人工智能与结构健康监测集成应用的科研人员与行业从业者提供全面参考，最终助力构建更安全、更具韧性的基础设施系统。