Machine learning analysis of impact sounding test for monitoring bridge deck interlayer condition

Abstract

The aging infrastructure worldwide requires rapid monitoring and maintenance to ensure the extension of its serviceable life. Bridges are one of the major transportation infrastructures that require intensive monitoring strategies for their proper maintenance. Developments and applications of bridge monitoring systems have become an active research area in recent years addressing the need for rapid assessment and application of mitigation measures in case of disasters such as earthquakes and typhoons. It has been acknowledged that one of the challenges in bridge condition monitoring is its interlayer pavement exposure to direct live loads and harsh environmental conditions. Non-destructive testing strategies are a prevalent monitoring method for bridges, and impact sounding tests are one of them in which integration of machine learning in its analysis improved its speed of providing results. In this study, machine learning methods are implemented to analyze impact-sounding devices for bridge deck pavement condition monitoring.

__________________________________________

*Corresponding author

Email address: cristinaamor.rosales@g.batstate-u.edu.ph