An Innovative Forecasting Formula for Axial Compression Capacity of Circular Steel Tubes Filled with Concrete through Neural Networks

Document Type : Original Article

Authors

1 Structural and Earthquake Research Center, Taft Branch, Islamic Azad University, Taft, Iran

2 Department of Civil Engineering, Vali-e-Asr University of Rafsanjan, Rafsanjan, Iran

10.22115/cepm.2023.394004.1235

Abstract

The manufacturing industry widely employs concrete and steel as building materials. These materials can be cleverly combined to create an efficient and innovative system, commonly referred to as a composite system. Despite the advantages and high performance of circular concrete filled steel tube (CCFST), there is a lack of reliable and accurate relationships for estimating their ultimate capacity. To address this issue, a wide range of valid experimental tests have been collected as a reference for actual data. By utilizing intelligent systems, such as artificial neural networks (ANN), the data can be effectively used to estimate the ultimate capacity of CCFST columns. Selecting the appropriate algorithm is critical for ANNs to eliminate unnecessary errors and produce optimal outputs. This study proposes a relation created by ANN to determine the ultimate capacity of CCFST columns and assesses its accuracy. Finally, a comparison with existing formulas has been conducted. The proposed network introduced enough accuracy compare to other existing methods.

Keywords

Main Subjects


  • Receive Date: 01 April 2023
  • Revise Date: 18 April 2023
  • Accept Date: 22 April 2023
  • First Publish Date: 22 April 2023