Reliability Prediction of Near-Isothermal Rolling of TiAl Alloy Based on Five Neural Network Models.

The near-isothermal rolling process has the characteristics of multi-variable and strong coupling, and the industrial conditions change constantly during the actual rolling process. It is difficult to consider the influence of various factors in industrial sites using theoretical derivation, and the compensation coefficient is difficult to accurately determine. The neural network model compensates for the difficulty in determining the compensation coefficient of the theoretical model. The neural network can be trained in advance through historical data, the trained network can be applied to industrial sites for prediction, and previous training errors can be compensated for through online learning using real-time data collected on site. But it requires a large amount of effective historical data, so this research uses a combination of production data from a controllable two-roll rolling mill and finite element simulation to provide training data support for the neural network. Five trained neural networks are used for prediction, and the results are compared with industrial site data, verifying the reliability and accuracy of genetic algorithm optimized neural network prediction. We successfully solved the problem of low control accuracy of TiAl alloy outlet thickness during near-isothermal rolling process.

Application of Mean Modulus in Three-Point Bending and Roll Forming.

Nonlinear unloading plays an important role in predicting springback during plastic forming process. To improve the accuracy of springback prediction which could provide a guide for precision forming, uniaxial tensile tests and uniaxial loading-unloading-loading tensile tests on SUS304 stainless steel were carried out. The flow stress mathematical model and chord modulus mathematical model were calibrated according to the test results. A constant elastic modulus three-point bending finite element model (E0FEMB) and a constant elastic modulus roll forming finite element model (E0FEMR) were established in MSC.MARC. The chord modulus was output by the PLOTV subroutine to determine the mean modulus of different regions, and the mean modulus three-point bending finite element model (E¯cFEMB) and the mean modulus roll forming finite element model (E¯cFEMR) were defined. The constant modulus finite element model (E0FEM) simulation results and the mean modulus finite element model (E¯cFEM) simulation results were compared with the three-point bending tests and roll forming tests test results. The difference between the simulation results and the test results was small, indicating that the mean modulus was feasible to predict the springback, which verified the suitability of the E¯cFEM.