An Improved Constitutive Model Based on BP Artificial Neural Network and 3D Processing Maps of a Spray-Formed Al–Cu–Li Alloy

The flow stress behavior of an innovative spray-formed aluminum–copper–lithium (Al–Cu–Li) alloy was successfully investigated via isothermal compression tests under a deformation temperature range of 350–450 °C (25 °C interval) and a strain rate range of 0.01, 0.1, 1, 5, 10 s−1. The constitutive relationship was established based on backpropagation artificial neural network (BP-ANN) algorithm. And the 3D processing maps were constructed as well. The results show that the constitutive model is in great agreement with the experiment data where the correlation coefficient goes up to 0.99963 and the average residual error lies only 1.06%. Moreover, from the 3D processing maps, the area of the instable regions tends to enlarge by virtue of the increasing strain, and the optimum processing domain is advised to be 440–450 °C. The microstructure evolution is found consistent with the prediction of the processing map.