Journal of Japan Society of Civil Engineers, Ser. B3 (Ocean Engineering)
Online ISSN : 2185-4688
ISSN-L : 2185-4688
Annual Journal of Civil Engineering in the Ocean Vol.36 (Special Feature)
EVALUATION METHOD OF RESIDUAL STRUCTURAL PERFORMANCE BY USE OF ARTIFICIAL INTELLIGENCE
Kunihiko UNOKe BAIMitsuyasu IWANAMI
Author information
JOURNAL FREE ACCESS

2020 Volume 76 Issue 2 Pages I_600-I_605

Details
Abstract

 Appropriate maintenance should be conducted for harbor structures due to their vulnerability towards chloride induced deterioration. We have already proposed an evaluation method of residual structural performance based on the judgement result of deterioration by versatile structural analysis software. On the other side, a simple evaluation method without structural analysis is also demanded when deciding the maintenance priority of several piers. In this study, evaluation methods of residual structural performance introducing artificial intelligence to estimate the damaged area ratio towards assumed external force are proposed. The estimation accuracy of artificial intelligence models are verified and damage estimation for piers with random deterioration distribution are conducted. As a result, for both models utilized numerical information and image showed high estimation accuracy, and the image model can also grasp the damage distribution as well as the damaged area ratio.

Content from these authors
© 2020 Japan Society of Civil Engineers
Previous article Next article
feedback
Top