NEURAL NETWORK MODELING FOR ROTATIONAL CAPACITY OF COLD-FORMED PURLIN STEEL SECTIONS

The possibility of consuming artificial neural networks (ANN) using Matlab software to calculate the rotational capacity of steel cold-formed C- and Z-section purlins. Rotational capacity is a significant phenomenon as in the situation of steel purlins which are extensively used in roofing wide-ranging buildings. The complex conducts of such members make the conventional design approaches not satisfactory from a reliability standpoint. The main aim of this paper was to give a quick and precise technique for estimating local buckling capacity of C-and Z-section purlin. Good agreement was attained concerning (ANN) technics outcomes and data from literature. Trained neural network develops easy to-utilize method for calculating yielding and ultimate moment's capacity of C-and Z-section. Broad parametric investigations were additionally performed and introduced graphically to analyse the impact of geometric and mechanical properties on rotational capacity. It was found that the proposed (ANN) based technics is practical in predicting both the yield and buckling rotational strength of cold-formed purlin steel sections.