Neural Network Modeling of the Rheology of the Almg6 Alloy under the Dispersoid Barrier Effect and the Inhibition of Dynamic Relaxation Processes

摘要

The paper deals with a neural network to model the flow stress of the AlMg6 alloy at temperatures ranging between 300 and 500 °C and strain rates from 1 to 25 s?1. In this temperature–strain-rate range, the movement of free dislocations is blocked and dynamic relaxation processes are inhibited. The results of training the neural network and its verification at a temperature not used in the training show that neural networks with a single hidden layer can correctly approximate and predict the rheological behavior of the AlMg6 alloy for the studied temperature–strain-rate range of deformation.