Determination of hot and cold rolling textures of steels: combined Bayesian neural network model

摘要: The work reported in the present paper outlines use of a combined artificial neural network model capable fast online prediction textures low and extralow carbon steels. We approach problem by Bayesian framework that takes into account as input to influence 23 parameters describing chemical composition thermomechanical processes, such austenite ferrite rolling, coiling, cold working subsequent annealing, involved production route output is form fibre texture data. predictions provide an excellent match experimentally measured results presented demonstrate this can help optimising normal anisotropy rm steel products.