Artificial Neural Network–Based Predictive Tool for Modeling of Self-Centering Endplate Connections with SMA Bolts

Endplate moment connections with shape memory alloy (SMA) bolts provide self-centering for the seismic resilience of structures. Predicting response these new beam–column connections, which have not been codified yet, requires conducting experimental tests or detailed continuum finite-element simulations. Computationally efficient predictive tools are needed to facilitate analysis, design, and assessment frames. In this paper, artificial neural networks (ANNs) used develop a MATLAB tool predicting moment-rotation backbone extended endplate SMA bolts. As input networks, model development employs design database parameters from 72 (FE) models seven connection specimens. Neural trained parameters, graphical user interface (GUI). The coefficient determination ANNs is in range 0.92 0.99, indicating acceptable prediction accuracy. Furthermore, optimization studies using multiobjective genetic algorithm performed, seeking minimization material use (steel SMA) improved connection-response characteristics (i.e., stiffness, strength, ductility). A phenomenological also developed OpenSees. ANN-based accurate modeling SMA-based moment-resisting frames illustrated. computation time typical significantly reduced hours ANSYS only three minutes OpenSees while providing same level results confirmed by performing nonlinear pushover history analyses.