Integration of Non-Destructive Testing Techniques and Machine Learning Algorithms for Enhanced Structural Health Monitoring of Bridges

Abstract

Aim: To examine the integration of non-destructive testing techniques and machine learning algorithms in order to enhance structural health monitoring of bridges. Problem Statement: Bridges are vital structures in civil engineering which have broad purposes and economic influence. However, they get expired over some period calling for their structural health monitoring from time to time to avoid any catastrophic event that may arise from their collapse. Significance of Study: This technical review critically examines the need to adopt the use of non-destructive testing techniques and machine learning algorithms in order to enhance structural health monitoring of bridges. Methodology: Recent relevant published articles in the area of structural health monitoring of bridges, non-destructive testing techniques and machine learning algorithms were consulted. Discussion: The advancement of technology has greatly influenced the incorporation of non-destructive testing techniques and machine learning algorithms to enhance structural health monitoring of bridges. This technically review has discussed the fundamental principles of structural health monitoring of bridges, non-destructive testing techniques and machine learning algorithms. Various non-destructive testing methods for SHM of bridges were highlighted and major emphasis was laid on visual testing, ultrasonic testing, liquid penetrant testing and radiographic testing. Their respective advantages and shortcomings were discussed. Application of machine learning in bridge SHM to improve the monitoring techniques was discussed. Different supervised and unsupervised learning algorithms that are applicable to the SHM of bridges are explained. Conclusion: The incorporation of non-destructive testing techniques and machine learning algorithms to structural health monitoring of bridges is imperative to integrate and enhance the process.