Artificial Intelligence and Data Mining Applications in Fault Diagnosis and Damage Identification of Infrastructure

Submission Deadline: 01 April 2026 View: 273 Submit to Special Issue

Guest Editors

Assoc. Prof. Dr. Kai Tao

Email: kai.tao@njupt.edu.cn

Affiliation: Collage of Automation, Nanjing University of Posts and Telecommunications, Nanjing 210003, China

Homepage:

Research Interests: structural health monitoring, multimodal data mining, machine learning, damage identification

Prof. Nizar Faisal Alkayem

Email: nizar.alkayem@njupt.edu.cn

Affiliation: Collage of Automation, Nanjing University of Posts and Telecommunications, Nanjing 210003, China

Homepage:

Research Interests: structural health monitoring, evolutionary computation, machine vision, deep learning, machine learning, damage detection, applied intelligence, intelligent sensing technology

Summary

Global infrastructure systems, encompassing various types of structures such as civil (bridges, buildings, etc.), mechanical (rotating machinery, wind turbines, pipelines, etc.), power systems, and transportation networks, are susceptible to performance degradation due to aging, environmental extremes, and operational loads. Undetected failure or damage in these systems can cascade into catastrophic failures, endangering safety and functionality. The urgent need for precise fault diagnosis and damage identification drives demand for next-generation structural health monitoring (SHM) solutions.

Advances in artificial intelligence (AI) and data mining enable transformative approaches to interpreting complex SHM data, automating the detection, localization, and severity assessment of damage across diverse infrastructure types. These technologies unlock predictive insights critical for risk mitigation and lifecycle management.

This Special Issue seeks cutting-edge technologies of AI and data mining methodologies for fault and damage diagnosis in infrastructure systems. Submissions should address algorithmic innovation, validation case studies, or scalable data-driven frameworks. All manuscripts will undergo rigorous peer review. Topics include, but are not limited to:
· AI for real-time damage detection in civil/mechanical/power system related structures/tunnel systems.
· Damage identification using machine and deep learning (e.g., crack quantification, corrosion mapping).
· Anomaly diagnosis in sensor networks via machine learning.
· Data mining for large-scale SHM data (e.g., IoT, UAVs, distributed sensors).
· Multimodal data fusion for cross-infrastructure damage assessment.
· Explainable AI (XAI) for interpretable fault diagnosis.
· Transfer learning for cross-domain damage identification (e.g., bridges → tunnels).
· Computer vision for automated defect inspection.
· Natural language processing (NLP) for mining fault records/maintenance logs.
· Resilience-oriented digital twins with embedded AI diagnostics.
· Unsupervised learning for novel fault discovery in unlabeled data.
· AI-optimized sensor placement for critical infrastructure.

Keywords

Published Papers