A probabilistic model for visual inspection of concrete shear walls

Arvin Ebrahimkhanlou; Salvatore Salamone

doi:10.1117/12.2258614

12 April 2017 A probabilistic model for visual inspection of concrete shear walls

Arvin Ebrahimkhanlou, Salvatore Salamone

Proceedings Volume 10168, Sensors and Smart Structures Technologies for Civil, Mechanical, and Aerospace Systems 2017; 101680Y (2017) https://doi.org/10.1117/12.2258614
Event: SPIE Smart Structures and Materials + Nondestructive Evaluation and Health Monitoring, 2017, Portland, Oregon, United States

Abstract

This paper presents a probabilistic model, called Bayesian networks, to visually assess the state of damage in reinforced concrete shear walls. The goal of this research is to reduce the inspection time and decrease the chance of missing or underestimating the state of damage in such structures. To develop this model, we define six types of visible damage on concrete shear walls. The model describes the causal relationship of such damage signs with the design parameters and damage states of the walls. To train and test the model, a database of all visually documented experimental works on concrete shear walls was collected from the literature. The model is trained on ninety percent of the database, and its performance is successfully validated on the ten percent remaining unseen portion of the database. The results show that the model can classify the images of yielded and failed walls. Additionally, it can prognosticate the most probable failure scenario for a yielded wall.

Citation Download Citation

Arvin Ebrahimkhanlou and Salvatore Salamone "A probabilistic model for visual inspection of concrete shear walls", Proc. SPIE 10168, Sensors and Smart Structures Technologies for Civil, Mechanical, and Aerospace Systems 2017, 101680Y (12 April 2017); https://doi.org/10.1117/12.2258614

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available