Evaluation of the ultimate eccentric load of rectangular CFSTs using advanced neural network modeling

• An optimization strategy was conducted to obtain a final set of ANN model. The developed model exhibited superior performance than current codes and empirical equations. explicit equation based on Artificial Neural Networks were derived for practical application. In this paper an Network (ANN) is the prediction ultimate compressive load rectangular Concrete Filled Steel Tube (CFST) columns, taking into account eccentricity. To end, experimental database CFST specimens from literature has been compiled, totaling 1224 individual tests, both under concentric eccentric loading. Except eccentricity, other parameters taken consideration include cross section width, height thickness, steel yield limit, concrete strength column length. Both short long evaluated. architecture proposed optimally selected, according predefined metrics. then compared against available design codes. It found that its accuracy significantly improved while maintaining stable numerical behavior. describes mathematically offered in paper, easier implementation evaluation purposes.