Study On The Prediction Of Superplastic Properties And Microstructure Of WSTI3515S Burn-resistant Titanium Alloy

WSTi3515S alloy is a new type of highly alloyed ?-type burn-resistant titanium alloy successfully developed in C hina in recent years,which has excellent burn-resistance properties and comprehensive properties.However,the grains are coarse and the deformation resistance is large,and the process is easy to crack,which is difficult to form by conventional processing methods.Superplastic forming technology can effectively solve the problem of difficult deformation of titanium alloy processing,but superplastic deformation is a nonlinear process.The superplastic properties are very sensitive to thermal parameters such as temperature and strain rate,and coarse grains will undergo significant changes during superplastic deformation.Therefore,a numerical simulation method is used to establish a prediction model to study the influence of thermal parameters on the superplastic properties and the evolution of microstructures of WSTi3515 S alloy.It is of great significance to predict and control microstructures changes,to formulate a suitable superplastic forming process and shorten the process cycle.BP neural network and cellular automata method are widely used in the field of plastic processing.BP neural network algorithm is very good at dealing with non-linear systems and has high prediction accuracy.The cellular automata method can show the dynamic evolution of microstructure in the process of grain growth and recrystallization.Therefore,based on BP neural network method and cellular automata method,the superplastic properties and microstructure evolution of WSTi3515 S alloy during superplastic deformation can be effectively predicted.In this paper,based on the mechanical properties and microstructure properties obtained from the superplastic tensile test of the WSTi3515 S burn-resistant titanium alloy,a BP neural network was used to establish a superplastic performance prediction model,and a cellular automaton method was used to establish a microstructure prediction model to study the thermodynamic behavior and microstructure evolution of the alloy during superplastic deformation.The main research contents and results are as follows:Based on the mechanical performance data of WSTi3515 S burn-resistant titanium alloy,with deformation temperature,strain and strain rate as input parameters,and flow stress as output parameters,a 3×10×12×1 hidden layer flow stress BP neural network for the prediction model.The correlation coefficient of the experimental value and the predicted value of the rheological stress is 0.99963,and the average relative error is 1.08%.The prediction result and the experimental data are in good agreement,indicating that the BP neural network model has higher accuracy and can better predict the superplastic mechanical properties of WSTi3515 S burn-resistant titanium al oy.Based on the WSTi3515 S burn-resistant titanium alloy superplastic deformation structure parameters data,3×21×14×3 double hidden was established with deformation temperature,strain and strain rate as input parameters,elongation,recrystallization volume fraction and average grain size as output parameters.The correlation coefficient between the experimental value and the predicted value of each structure parameters is 0.99568,and the average relative error is 4.34%.The prediction results are in good agreement with the experimental data,indicating that the BP neural network model has higher accuracy in the prediction of structure parameters data.According to the prediction result of BP neural network model of the superplastic deformation process of WSTi3515 S burn-resistant titanium alloy,three-dimensional map of flow stress,elongation,volume fraction of recrystallization and average grain size were established.It can well describe the change between superplastic properties and thermodynamic parameters of WSTi3515 S burn-resistant titanium alloy during superplastic deformation,and can also predict the true stress-true strain curve,elongation,recrystallized volume fraction and average grain size under any deformation conditions.Based on the principle of cellular automata,and combined with the curvature growth mechanism and the transformation rules of cells,the cellular automaton model of the initial grain structure of WSTi3515 S burn-resistant titanium alloy was established,and compared with the test results,the reliability of the initial structure model is verified.Based on the initial microstructure model and combined with the dynamic recrystallization theory,a dynamic recrystallization model for superplastic deformation of the WSTi3515 S burn-resistant titanium alloy was established.Through the simulation of the recrystallization process of alloy deformation under different deformation conditions,the influence of deformation parameters on the microstructure evolution process and recrystallization volume fraction was analyzed,compared with the test results,the simulation results are consistent with the test results.The cellular automaton model can accurately reveal the dynamic recrystallization microstructure evolution of WSTi3515 S burn-resistant titanium alloy,and can provide theoretical guidance for the structural prediction of superplastic deformation of WSTi3515 S burn-resistant titanium al oy.