Surface Roughness Prediction And Parameter Optimization Of PCBN Tools For Interrupted Turning Of Hardened Steel

In mechanical production processing,surface roughness is an important index to measure the surface processing quality of the workpiece,and the establishment of accurate surface roughness prediction model is of great practical significance in production to improve the processing quality.Hardened steel,as a cold work tool steel with a wide range of applications,has the advantages of good wear resistance,hardenability and thermal stability,and is generally applicable to a variety of machining situations.In this paper,in order to establish the prediction model of surface roughness of hardened steel for interrupted turning and to carry out multiobjective parameter optimization based on surface roughness,PCBN tools suitable for hard state interrupted cutting are selected,and the test related to interrupted turning is designed by Box-Behnken.The surface roughness RSM prediction model and BAS-BP neural network prediction model based on cutting parameters and workpiece hardness were established respectively,and the prediction accuracy of the two was compared,and the results showed that the established BAS-BP neural network model has good prediction accuracy.At the same time,in order to meet the actual production requirements,the established surface roughness prediction model was combined with transient impact force and cutting ratio energy,and the cutting parameters were optimized by establishing and solving a multi-objective optimization model,and finally the before and after optimization effects were compared to prove the feasibility of the cutting parameter optimization results.The specific research results of this paper are as follows:(1)The design of interrupted turning-related tests and the summary of turning test laws.Through theoretical analysis of the formation process of surface roughness,combined with the special characteristics of interrupted turning hardened steel,the single-factor test was designed to prove the influence of the degree of interruption on the surface roughness,and finally the influencing factors of the multi-factor test were determined as cutting parameters and workpiece hardness;through the Box-Behnken design turning test,the test data were collected and analyzed,and the influence factors and the interaction between the factors on the surface roughness were derived.By using the Box-Behnken design turning test,the test data were collected and analyzed,and the influence law of each influencing factor and the interaction between the factors on the surface roughness was derived,which provided a reference for the actual machining production.(2)The BP neural network was improved and a BAS-BP surface roughness prediction model with high prediction accuracy was established.Based on the experimental data,the Response Surface Method(RSM)was firstly established and the prediction accuracy of RSM model was found to be improved,and then the traditional BP neural network was improved by using the Beetle Antennas Search(BAS),so that the BAS-BP neural network prediction model with higher prediction accuracy was established.By comparing the evaluation parameters of RSM and BAS-BP two prediction models,the BAS-BP neural network was finally determined as the prediction model of interrupted turning surface roughness,which realized the accurate prediction of interrupted turning surface roughness and laid the foundation for the subsequent establishment of a multi-objective optimization model based on surface roughness.(3)The establishment and solution of the multi-objective optimization model,as well as the analysis of the parameter optimization results and optimization effects.Using the established surface roughness prediction model,the multi-objective optimization model was established by combining the minimum cutting specific energy and the minimum transient impact force;the NSGA-II algorithm in genetic algorithm was used to solve the bi-objective and tri-objective optimization models respectively,and the Pareto solution set based on the cutting parameters was obtained;by comparing the empirical values with the optimization target values corresponding to the obtained solutions,the optimal combination of cutting parameters was determined.The optimal combination of cutting parameters was determined by comparing the empirical values with the obtained solutions corresponding to the optimized target values,and comparing the before and after optimization effects to prove the feasibility of the optimization results,which will be helpful for the selection of multi-objective based cutting parameter optimization in actual machining production.