Intelligent design in continuous galvanizing process for advanced ultra-high-strength dual-phase steels using back-propagation artificial neural networks and MOAMP-Squirrels search algorithm

摘要

In this research work, the optimization of a back-propagation artificial neural network (BPNN) using a new multi-objective bio-inspired algorithm based on squirrels is proposed in order to optimize the main continuous galvanizing process parameters such as the initial cooling rate (CR1), the isothermal holding time at 460(o)C (t(g)), and the final cooling rate (CR2). The computational approach predicts in a satisfactory way the most important mechanical properties including yield strength (YS), ultimate tensile strength (UTS), and elongation at fracture (EL) of cold rolled low carbon DP steels treated under continuous galvanizing thermal cycle conditions. The experimental production of galvanized ultra-high-strength DP steels from cold rolled low carbon sheets with a minimum UTS of 1100 MPa, YS between 550 and 750 MPa, and a minimum elongation of 10% is possible using the proposed methodology.