Research On Upsetting Technology Of 42CrMo Steel And Development Of Integrated Control Software

With the development of various technologies,the requirements for the quality of forgings are getting higher and higher in the field of precision manufacturing.In order to ensure the quality of the final forging,it is necessary to ensure the good uniformity of the initial billet structure.Upsetting is the most commonly used billet making process.The upsetting hot deformation process of 42 Cr Mo steel is taken as the object in this paper.The microstructure change and high temperature rheological behavior of 42 Cr Mo steel during hot compression were studied.The method of press load and energy prediction and the prediction method of billet microstructure during the upsetting process of 42 Cr Mo steel on electric screw press were explored.An integrated software system for the upsetting process of 42 Cr Mo steel in electric screw press was established.the upsetting process of the42 Cr Mo steel on electric screw press was established.The main research contents and conclusions of this paper are as follows:(1)Through the thermal simulation compression test with a variety of experimental conditions,the microstructural evolution of 42 Cr Mo steel is analyzed by OM,and the processing map of 42 Cr Mo steel was established.The results show that both the strain rate and the deformation temperature have a significant effect on the microstructure after deformation;it shows that when the deformation rate is constant,the increase of the temperature promotes the dynamic recrystallization behavior of the material,the recrystallization nucleation rate increases,then the grain size becomes uniform.However,the recrystallization nucleation rate is saturated when the temperature is too high,which lead to the growing up of the matrix grains and the coarsening of the recrystallized grains;when the temperature is constant,the increase of the strain rate also promotes the occurrence of dynamic recrystallization;with the increase of strain during thermal deformation,the instability interval also gradually increases,and the instability mainly occurs in the region of low temperature and high strain rate.Lower deformation temperatures and faster strain rates should be avoided as far as possible during thermal deformation.(2)The high temperature rheological properties of 42 Cr Mo steel were analyzed based on the experimental data of thermal simulation compression,and a rheological behavior prediction model based on DBN deep confidence network was established.The results show that the flow stress curve of 42 Cr Mo steel is mainly characterized by high temperature deformation softening mechanism and work hardening.With the increase of deformation,the flow stress increases rapidly,and then gradually decreases after reaching the peak value,and finally tends to be stable.The absolute percentage error(MAPE)and root mean square error(RMSE)of the model are 3.29% and 3.3754,respectively,which have high prediction accuracy,indicating that the DBN prediction model can be used to accurately describe the high temperature rheological behavior of 42 Cr Mo steel.(3)The improved neural network screw press load prediction method based on genetic algorithm and the Taylor expansion load prediction method combined with fuzzy reasoning are studied.The absolute value of relative error of most samples of the improved neural network screw press load prediction model based on genetic algorithm is within 0.02,and its performance index value is only 0.71%,which is significantly lower than the performance index of the neural network model before the improvement.The method can more accurately predict the load of the screw press;the performance index of the Taylor expansion load prediction model combined with fuzzy reasoning is only0.15%,which is better than the prediction ability of the traditional Taylor expansion model.(4)The recrystallization fraction and average grain size data of the target points under different process parameters were extracted from the simulation experiments,and a microstructure soft measurement method based on LSTM was proposed.The method has good learning ability and soft measurement accuracy.The energy prediction method in the actual striking process is studied by analyzing the force and energy characteristics of the screw press.An integrated system software for the upsetting process of the 42 Cr Mo steel electric screw press is established based on Lab VIEW,which can realize the final energy prediction output under the input parameters and the output of load data and microstructure data under the energy parameters.