Research On Cutting Tests And The Optimization Of Cutting Parameter For The Hard-to-cut Material

With the rapid development of science and technology in our country, the level of mechanical manufacturing has increased significantly. Mechanical manufacturing plays an important role in national economy, especially in the field of aviation industry,medical industry and national defense. Therefore, there are higher requirements for the quality of machining. The surface roughness is an important index to measure the quality of machined surface. The tool wear has a direct affect on machining cost,surface quality of parts, machining precision and machining efficiency. Thus, the prediction of machining tool wear and surface roughness of workpiece and the optimization of cutting process parameters were studied in this paper.The strong influence of machining parameters on the tool wear can shorten the tool life. It can not only greatly reduce the machining accuracy and surface quality of workpieces, but also increase the machining cost of workpieces. It is an important task to select a reasonable set of machining parameters during the machining procedure.Therefore, the orthogonal experimental design method was used to carry out the plane-milling experiment with the hard-to-cut material of martensitic stainless steel. The Ultra Depth-Of-field 3D Microscope is adopted to measure the tool flank wear as the training samples. And then, with the nonlinear mapping of neural network, the finite training samples can be employed to formulate the prediction model of tool wear as a function of cutting speed, feed per tooth, depth of cut, and cutting time. Experimental results show that the prediction error of the proposed neural network model is no more than 5.4%. Finally, the optimal model of machining parameters is established with the objective of minimizing the tool wear. According to the tool wear of each generation,the evaluation function is defined for the fitness of the individual. Thus, the genetic algorithm can be skillfully developed to solve the optimal model of tool wear. In comparison with the Taguchi method, the optimal machining parameters which are obtained by the genetic algorithm based optimal model can bring a 6.734 percent decrease of tool wear. The proposed method can not only improve the calculation efficiency and precision, but also provide a basic theory for the selection of machining parameter.The surface roughness is an important parameter to measure the surface quality of a workpiece. It’s critical to properly control the surface roughness so that the machiningquality can be granted. The factors that influence the surface roughness are complicated and uncertain in practical processing.Therefore, the uniform experimental design method was used to carry out the cutting experiment with the hard-to-cut material of TC4 titanium alloy. The MarSurf M 300 C precise surface roughness measuring instrument was employed to measure the surface roughness of workpiece. And then,with the nonlinear regression technology, the prediction model of surface roughness as a function of cutting speed, feed, depth of cut, and length of cut was established. The fitting degree of the prediction model and the significance of all independent input parameters were tested by ANOVA method. The contrast experimental between experimental values and predicted values shows that the prediction model has an error as low as 0.019%, which means the model is highly precise and reliable. Finally, the optimal model of machining parameters was established with the objective of minimizing the surface roughness. The genetic algorithm was used to solve the optimal model of surface roughness. The proposed method can provide a basic theory for surface roughness prediction.