Study for Artificial Neural Network of Aluminum Benchmark Bridge
- September 16, 2020
- Posted by: RSIS Team
- Categories: IJRIAS, Mathematics
International Journal of Research and Innovation in Applied Science (IJRIAS) | Volume V, Issue III, February 2020 | ISSN 2454–6186
Study for Artificial Neural Network of Aluminum Benchmark Bridge
Sertaç TUHTA1, Furkan GÜNDAY2
1,2Department of Civil Engineering, Ondokuz Mayis University, Turkey
Abstract—Today, civil engineering structures suffer from dynamic effects. Earth on structures have been severely damaged by the earthquake. Thus, there has been loss of life and property. This has particularly affected countries located on active fault lines. Pre- and post-earthquake measures have been developed in world. For these reasons, it is necessary to determine the dynamic performance of structures around the world. There are various methods for determine the dynamic performance. System identification is one of these methods. Mathematical model of the structural system is obtained by system identification method. Artificial Neural Networks (ANN) is a system identification method. Artificial Neural Networks (ANN) can adapt to their environment, adapt, work with incomplete information, make decisions under uncertainties and tolerate errors. Aluminum benchmark bridge sample was used in this study. The system identification of the aluminum benchmark bridge with the ANN method of 0.87 was made successfully. As a result of this study, The ANN approach can provide a very useful and accurate tool to solve problem in modal identification studies.
Keywords—System Identification, ANN, Levenberg-Marquardt algorithm, Aluminum Bridge
