Study On Warm-hot Deforming Behavior Of Mg-5.15Y-3.75Gd-3.05Zn-0.75Zr Rare Earth Magnesium Alloy

Magnesium alloys are widely used as light metal materials because of light weight,low density,high specific strength and high specific rigidity.It has been widely used in aerospace,3C,automotive et.al.The rare earth elements on the Earth exist in the form of oxides,in addition,the purification process of rare earth elements is very complicated and costly.Therefore,the research on the properties of rare earth magnesium alloys is of vital importance to its wider application.The microstructure evolution and rheological behavior of metals in the process of high temperature deformation are very complicated.The microstructure of the material has a direct effect on the mechanical properties of the material,but microstructure and the flow stress during material deforming affected by the deformation temperature,strain rate,and strain.And the mechanism of influence is very complicated due to the influence of various factors.Therefore,building the thermal deformation constitutive equation,flow stress prediction model and the processing map of the hot deformation material provide a scientific basis and theoretical guidance for the formulation of the material deformation parameters and the control of its deformation quality.Because the flow stress of the material is complicated and non-linear affected by various factors,so this paper builds a self-organized and autonomous Artificial Neural Network model(ANN)to predict the thermal rheological behavior of Mg-5.15Y-3.75Gd-3.05Zn-0.75Zr rare earth magnesium alloy.This paper carries out a hot compression experiment with Gleeble-1500D thermal system,and the experiment is performed in the temperature range between 350? and 400?,and strain rates between o.1s^-1 and 5s^-1.Using the obtained experimental data,the constitutive equation of the material at the peak stress is constructed based on the Arrhenius hyperbolic sine function model.The correlation between the experimental values and the predicted values by the constitutive model is as high as 0.95696.However,due to the complex calculation of the constitutive equation and the influence of the strain variable is not considered,Then an artificial neural network prediction model of the alloy flow stress is constructed,based on the BP artificial neural network and experimental data.The correlation between the experimental values and the predicted values of the model was as high as 0.99982,and the largest relative error was 3.478%,the minimum is-2.660%,the average value is 0.062%,the prediction accuracy has been improved more than the traditional constitutive equation.Moreover the calculation process of this model is simpler and more convenient.Finally,this paper draw the hot processing map Mg-5.15Y-3.75Gd-3.05Zn-0.75Zr rare earth magnesium alloy in the strain of 0.1,0.3,0.5,based on the materials'processing map theory and the experimental data.Analyzes the relationship between the energy dissipation rate ?,the instability parameter?(?)and the deformation temperature and strain rate.The result shows that the preferred deformation condition range of the alloy is:the deformation temperature is 450-500?,and the strain rate is 0.1-1 s^-1.