ANN/GA (artificial neural network/genetic algorithm) for modeling and optimizing of liquid metal extrusion process

To ensure forming quality of a liquid metal extrusion process, the model of the technological process was built by using artificial neural network and the process parameters were optimized with genetic algorithm (GA). The method was applied to a liquid AlCuSiMg alloy extrusion. It is shown that the predicted values of the hybrid BPF (backpropagation and feedforward) neural network is in agreement with the experimental ones, and the process parameters optimized by GA are 716°C of pouring temperature, 250°C of die temperature, 2.6×10^-3 m/s of pressing velocity, 30 s of delay period before applying pressure and 86.6 MPa of minimum deforming force. These predicted optimal values agree well with test results.