A new constitutive relationship for alloy TC11

Engineers need constitutive relationships for planning engineering of high quality forgings. Accordingly, accurately describing the mechanical performance of material deformation is of decisive importance. In this paper, a new method is presented, in which a four-layer backpropagation neural network is built to acquire the constitutive relationship of the TC11 alloy based on the homogeneous compression test. Temperature, effective strain, and effective strain rate are used as the input vectors of the neural network, and the output of the neural network is the flow stress. After the network is trained with experimental data, it correctly reproduces the flow stress in the sampled data. Furthermore, when the network is presented with nonsampled data, it also correctly predicts the flow stress.