Optimization Of Cutting Parameters And Study Of Surface Topography In Turning Of Ti6Al4V

Compressor disk is one of the crucial components in the aero-engine,and the machining accuracy of compressor disk is a key point to ensure the stable operation of the aero-engine.Ti6Al4 V is usually selected as a component material.Due to the poor machinability of titanium alloy,it is hard to balance the efficiency and quality in the manufacturing process.In this paper,based on the nested artificial neural network(ANN)and multi-objective teaching-learning-based optimization(TLBO),a cutting parameters optimization model,which aims to minimize the objective function of surface roughness and maximize the material removal rate,is proposed and solved.Additionally,the generation process of surface textures is investigated considering the influence of axial relative vibration between tool and work-piece in face turning.The main work and achievements are as follows:(1)Experimental study on turning of Ti6Al4 V.Turning experiments are conducted based on the principle of Box-Behken Design(BBD).Response surface methodology(RSM)is applied to build a linear RSM model and a quadratic RSM model using cutting parameters as input factors.The validity of the models and the effect of each cutting parameters are investigated through the analysis of variance(ANOVA).A linear regression model using cutting parameters and cutting states as input factors is developed to map the interactions.The degree of association between surface roughness and input factors including cutting parameters and cutting states is conducted by a correlation analysis.(2)Surface roughness prediction model based on nested neural network.This paper demonstrates a new method to build a nested artificial neural network(ANN).Using this method,a nested-ANN model predicting surface roughness is established,considering the effects of cutting forces and tool vibrations and using only cutting parameters as inputs.To verify the effectiveness of the nested-ANN model,it is compared with other mathematical and statistical models based on conventional BP and RSM(Response Surface Methodology)using the same experimental data.The results show that the nested-ANN uses less input variables and obtains superior prediction accuracy than other models.(3)Optimization of cutting parameters based on multi-objective teaching learning based optimization.An improved multi-objective TLBO algorithm is proposed based on the fast non-dominated sorting and crowding distance.Combined with the established nested ANN model,a cutting parameters optimization model,which aims to minimize the objective function of surface roughness and maximize the material removal rate,is established.The optimization model is solved by the multi-objective TLBO algorithm.The results show that the proposed method can find the evenly distributed Pareto optimal solutions effectively.(4)Simulation of 3D surface topography and prediction of surface texture feature in face turning.The surface of work-piece is generated based on the tool profile and real tool path as a footprint process.Through the analysis of surface generation under the influence of the axial relative vibration between tool and work-piece,the surface topography simulation model is build.Base on this model,different surface topographies under different frequency ratios,amplitudes and feed rates are simulated.The prediction method of the surface texture is also summarized.The experimental results show that the experimental results are in good agreement with the simulation results not only in the micro-surface structure but also in the macroscopic surface texture.The optimization of the cutting parameters and the simulation of the surface topography in turning of Ti6Al4 V can provide a theoretical guidance for the reasonable selection of the cutting parameters in manufacturing process.