A Review of Artificial Neural Networks (ANNs) as a Potential Predictive Tool for the Performance of Scissor-Type Deployable Bridges

Abstract

After a severe disaster, some places may be unreachable for rescue operations due to bridge destruction. A scissor-type deployable bridge is a novel rescue technology that enables a lifeline to be quickly recovered during a catastrophic event. Several structural analysis approaches were used to predict the structural behavior of deployable bridges. Yet none of the prior studies used Artificial Neural Networks (ANNs) to predict the structural behavior of scissor-type deployable bridges. This research explores the potentiality of Artificial Neural Networks (ANNs) to predict the performance of a scissor-type deployable bridge. The study aims to leverage the capabilities of ANNs in modeling complex relationships to forecast key parameters related to the bridge's functionality. ANNs can assist engineers in optimizing the design parameters of scissor-type deployable bridges by predicting how different configurations affect total deformation and stress levels. The analysis involves training the neural network with relevant data to enable it to learn and generalize patterns, enabling more informed predictions for diverse scenarios. Lastly, the application of ANNs in simulating bridge behavior contributes to advancing research in structural engineering, particularly in the field of deployable structures, by providing insights into complex structural responses that are challenging to model analytically.

Keywords: Deployable Bridge, Mobile Bridge, Scissor-Type Bridge, Aluminum Alloy, Artificial Neural Networks (ANNs), MATLAB