An overview of damage and crack detection in structures using metaheuristic algorithms and artificial neural networks

Monitoring structural health is very important to maintain the useful life of civil structures. Many monitoring methods have been developed to provide practical tools for early warning against structural damage or any type of anomaly. Therefore, the health monitoring method of the structure is considered the main approach for the management of identification and diagnosis of damage in different areas. The need to monitor the behavior of the structure is increasing every day, but due to the development of new materials and more complex structures, this leads to the development of strong and sensitive methods for the health monitoring of the structure. Artificial intelligence is an efficient alternative approach to classical modeling methods. Solutions based on artificial intelligence are good alternatives for determining engineering design parameters when testing is not possible; Therefore, it leads to a significant saving of human time and effort in experiments. Today, machine learning has become the most successful sub-branch of artificial intelligence. Identifying damage using intelligent signal processing and optimization algorithms based on vibration criteria is one of the important things. Some recent studies on the applications of artificial neural networks for damage and crack detection have been reviewed in this paper. An attempt has been made to provide a comprehensive review of the published articles in the field of application of optimization methods and inverse methods, artificial intelligence and machine learning, and assessment of damage and cracks in various structures using artificial neural networks, with a view, especially on the studies conducted in the past decades.