Simulation Of Temperature Field And Prediction Of Microstructural Evolution Of Steels During Heating And Cooling After Rolling

In steel production,heating and the cooling after rooling are two extremely important processes.In the processes,the temperature field and microstructures of the material change greatly,which play crucial roles in improving the final quality of the products.Therefore,it is of great theoretical significance and practical value for the formulation and optimization of heating and the cooling after rolling processes.This thesis simulated the temperature field of the billet heating and cooling process after rolling by using finite element method.A model has been established predicting the microstructural evolution during the cooling process after rolling.In addition,intelligent models were established for the prediction of temperature field and of microstructural evolution during cooling after rolling using artificial neural network method.The heating process of the billet was simulated using FEM with respect to the classical three-step stepwise heating mode.The influence of heating intensity and time on temperature field was analyzed,and nearly 3000 sets of data have been obtained.The cold process of the steel plate was also simulated using FEM with respect to the classical t three-stage cooling mode.The influence of cooling intensity,time,and slab thickness on temperature field was analyzed,and nearly 3,000 sets of data have been obtained.This work lays a good foundation for further research in this field.Based on computational materials science and mathematical statistics,a nucleation model considering the effect of time scale is established.The value of nt determined with unit time used as the reference time step.The average grain sizes predicted by using the proposed model are 23.48 and 22.97 ?m for QP980 and 42CrMo,respectively,which agrees with the actual values of 23.54 and 22.46 ?m obtained experimentally by other researchers.The results of this thesis show that the nucleation model is suitable for the prediction of grain size in the cooling process of steel plate and strip after rolling.An intelligent model for predicting temperature field and microstructures is established using neural network.The date of temperature field and microstructures for X80,42CrMo and QP980 are used to train the network model.By the analysis for Q345E using the established intelligent model,the optimal heating parameters including heating time,heating furnace size and heat transfer coefficient are obtained.The surface temperature of the slab reaches up to 1217.30?.The cross-section temperature difference is 7.20?.An analysis of microalloyed steel to obtain the optimal cooling process including cooling time,cooling device specifications and heat transfer coefficients,its microstructure is 81.61%ferrite and 18.39%pearlite,and the average grain size is 4.61 ?m.The results obtained from the intelligent prediction are in good agreement with the actual results by other researchers.It shows that the prediction model is also suitable for other materials,and exibits good universality and intelligence.