PulseExploreJournal ClubDebatesTrendingResearchersJournals
Instagram
HomeExploreJournal ClubTrending
Synapse
⌘+K
Synapse
May 20, 20260 citationsOpen Access

Soft Computing-Based Model for the Ultimate Strength Prediction of Concrete-Filled Circular-Steel Columns

View Full Paper
SASuha Rasheed AbbasAHAmmar N. HanoonAAA Abdulhameed

Key Points

  • To predict the axial compression capacity of concrete-filled steel tube (CFST) columns accurately using a soft computing model.
  • Employ a grey wolf optimizer (GWO) technique to optimize model membership functions.
  • Utilize a comprehensive database of 561 experimental tests for model design and validation.
  • Assess model accuracy and robustness using various statistical criteria.
  • The model achieved a coefficient of variation (CoV%) of 16.89%.
  • The estimate was found to be 1.08, indicating predictive accuracy.
  • A correlation coefficient (R) of 0.9663 confirmed high accuracy and repeatability.

Abstract

Concrete-filled steel tube (CFST) columns are popular in construction because of the composite effect between the inside concrete and the outside steel tube, improving structural resistance. Accurately predicting the axial compression capacity of CFST columns is critical to maintaining structural stability and avoiding collapse. This paper uses a grey wolf optimizer (GWO) technique to determine the compressive strength of circular CFST columns. The technique is used to optimize the membership functions of the introduced model. A comprehensive database of 561 experimental tests from the published literature was used to design and validate the model. Several statistical criteria were utilized to evaluate the model's accuracy and robustness. The presented empirical model provides a concise, intuitive, and powerful description of the final load capacity of CFST columns, applicable to both ordinary and high-strength concrete and steel. Comparisons between measured and predicted values of these parameters show the model's accuracy. The designed model is expected to help engineers effectively assess the axial capacity of CFST columns in accordance with design requirements. The proposed maximum-force model yielded a coefficient of variation (CoV%) of 16.89%, an estimate of 1.08, and a correlation coefficient (R) of 0.9663, indicating high accuracy and repeatability of the output.

Ask AI
Helpful
Bookmark
Share
View Full Paper

Cite This Study

Abbas et al. (2026) studied this question.

synapsesocial.com/papers/6a0d5064f03e14405aa9c27fhttps://doi.org/10.25130/tjes.33.1.25
Ask AI
Helpful
Bookmark
Share
View Full Paper