The simulated cooling of the hot-rolled structural steel sections

Abstract

Temperature models based on the finite difference, ADI and Runge-Kutta methods have been written in order to establish the most efficient algorithm when simulating the cooling of newly hot-rolled steel sections under a variety of cooling conditions. For air-cooling, the most efficient results were obtained using extended-stability Runge-Kutta methods, together with adaptive step-size control procedures. CPU time-savings of around 85% were achieved when an existing finite difference based section air-cooling model was modified to run using a specially developed, highly stable, second-order Runge-Kutta formula with the method of lines. The ADI approach gave the most efficient results for water spray cooling, producing accurate results in approximately half the CPU time required by the finite difference method.