Article navigation
Purpose

The paper proposes a promising approach by integrating 1DCNN and BiGRU to enhance the detection and assessment of structural damage.

Design/methodology/approach

This paper is organized into five main sections. Section 1 outlines the problem and highlights the innovative contributions of this paper. Section 2 introduces the theoretical background of the DL models, including 1DCNN, GRU, BiGRU, 1DCNN-GRU and the proposed 1DCNN-BiGRU hybrid. Section 3 details the dataset used in the model. Section 4 presents the results obtained from the proposed method. Lastly, the conclusion summarizes the key findings and outcomes of the study.

Findings

The proposed approach combines 1DCNN and BiGRU strengths. 1DCNN extracts key features from input data, while BiGRU learns and classifies sequential data bidirectionally, capturing vital temporal details effectively. Data augmentation uses random shifts, noise and reversal to enhance dataset diversity. A softmax layer calculates class probabilities, improving model confidence assessment per prediction. Validated on an experimental bridge, the method excels in detecting, analyzing and evaluating structural damage.

Originality/value

This paper is an original work authored by the authors and all rights are reserved. The content, including text, figures, tables and data, is the intellectual property of the authors unless otherwise cited. No part of this manuscript has been previously published or submitted elsewhere. The authors confirm that this work does not infringe upon any existing copyrights, and all sources have been appropriately acknowledged.

Licensed re-use rights only
You do not currently have access to this content.
Don't already have an account? Register

Purchased this content as a guest? Enter your email address to restore access.

Pay-Per-View Access
$41.00
Rental

or Create an Account

Close Modal
Close Modal