Proposed soft computing models for moment capacity prediction of reinforced concrete columns

Computational intelligence (CI) is a powerful approach to determine the response values of complex systems. Despite their benefits, the way to reach the solution in these approaches is difficult and cannot be expressed in a clear and simple formulation. In recent years, some methods have been proposed to provide simple and efficient mathematical forms in such approaches. In this paper, five of these methods are investigated to estimate the amount of moment capacity in rectangular concrete columns based on the extracted equations of CI. To train, validate and also test the proposed equations, a set of experimental laboratory tests of RC columns were collected from PEER database, and then mathematical frameworks for calculating the target were extracted. The obtained results of the proposed structures are also compared with each other, and it was concluded that all methods with high accuracy were able to estimate the moment capacity, but equation-based neuro-fuzzy system had better results than other presented models. The proposed equations are very powerful tools for determining the final capacity of RC columns as a key element in concrete structures.