مدلسازی رفتار رئولوژیکی از آلیاژهای آلومینیوم در تغییر شکل گرم و چند مرحله ای با استفاده از تجزیه و تحلیل رگرسیون چندگانه و شبکه عصبی مصنوعی

Artificial neural network and multiple regression analysis techniques were applied in modelling the rheological behaviour of AA 6082 aluminium alloy under multistep hot deformation conditions. To this end, multistage torsion tests were carried out in order to obtain the experimental data to be used in the development of the predictive models. The envelope curves predicted by both the ANN- and MRA-based models have shown an excellent fit, in terms of curve shape and stress level, with the experimental ones obtained under the same process conditions, even if the ANN based model has provided the best predictive capability.