Recently, customers are demanding for hot rolled strip products to have tight oxide scales on the surfaces. Therefore, high finishing rolling temperature, low coiling temperature and fast finishing rolling speed have to be used to obtain tight oxide scale, which is different from conventional controlled rolling. In order to ensure the mechanical properties at the same time, a framework consisting of the Bayesian neural network and multi-objective particle swarm optimization has been established to determine the optimal hot strip rolling parameters. Due to excellent generalization ability, the Bayesian neural network was employed to develop the model for the prediction of mechanical properties of hot rolled automotive beam steels. The accuracy between the measured and predicted values was within ±30 MPa and ±4% for strength and elongation, respectively, providing a reliable model for the optimal process design. By applying multi-objective particle swarm optimization, the optimized hot rolling process was obtained for the production of hot rolled automotive beam steel with "Tight Oxide Scale". Industrial trials have been carried out, which showed good agreement with the optimized hot strip rolling processes. It has been theoretically and practically proven that the optimal process design framework can effectively locate the optimal processing window for hot strip rolling.

No abstract

No abstract

No abstract

No abstract