Investigation Of Tool Wear In Machining H13 Steel And Prediction Of Machined Surface

H13 steel has the characteristics of high strength,high hardness,good corrosion resistance,wear resistance and heat resistance and usually works under the conditions of high temperature,high pressure and periodic load,so it is prone to wear,breakage,plastic deformation and other failures.In addition to improving material performance,optimizing mold structure and changing the heat treatment process,improving the finished surface quality had been treated as an important method to improve the performance and service life of H13 by researchers.H13 has a high hardness(usually greater than 50HRC)and therefore tool wears more quickly during traditional turning and milling H13 processes,When the tool wear occurred,its shape changed gradually,which affected the process of machining and the quality of machined surface.It is necessary to study the influence of tool wear pattern on the surface appearance of H13 systematically and control the surface roughness by selecting different cutting process parameters.We can not only improve the machining efficiency of H13 steel,but also improve the service life and reliability of the parts through controlling surface quality,which is of great significance for the improvement of machining H13 steel technology.In this dissertation,the machining H13 steel process was investigated.First of all,a three-dimensional finite element model of predicting wear morphology in turning H13 steel was established,in which the performance of tool wear was shown by the node displacement and the amount of wear was determined by the finite element result and wear rate equation.The wear morphology prediction model had high accuracy by comparing with the experimental results.The length of the wear area at the rake surface was consistent with the experimental results in the simulation results,while the simulation value had some deviation with the experimental value in the prediction of the depth of the wear area.This is because part of the material was bonded to the rake face during the machining process,resulting in an increase in the height of the rake face,which was inconsistent with the simulation's assumption that tool wear was a loss of tool material process.The main reason for wear at the flank face was abrasive wear and the simulation results had a good agreement with experiment results.Secondly,the influence of wear morphology on cutting performance and roughness of machined surface was studied by using the established FE model.The influence of tool wear evaluation and morphology on cutting force,cutting temperature,friction coefficient and roughness of the machined surface were investigated by comparing results with different cutting parameters.Once wear occurred at the rake face,the distribution and direction of stress changed,resulting in an increasing of real cutting front angle and the maximum equivalent stress.Besides,serious front face wear would cause cutting edge deformation or cutting edge and tool damage.The wear at the flank face led to the increase of contact area as well as cutting temperature and the serious flank wear was caused by the bonding and friction,which lowed the quality of the machining surface.Finally,an BP artificial neural network model was established to predict the roughness of the machined surface by considering the simulation results and experimental data.The Levenberg-Marquardt algorithm was selected in the neural network model and the input data contained the cutting parameters,tool geometry parameters and wear values.Compared with the experimental results,the improved neural network had better performance in the convergence speed and prediction accuracy in the case of less input training data.