Optimization Study Of Aluminum Alloy Grinding And Polishing Process Parameters

As a representative of metal materials,aluminum alloys are very widely used in optical systems.Grinding and polishing,as the final process of processing,directly affects the quality of the product components in the final finished product.In order to meet the different requirements for quality,the selection of reasonable processing parameters plays an important role in ensuring processing accuracy,improving processing efficiency and saving processing costs,so it is important to study how to improve product quality and production efficiency.The research of this paper is as follows:(1)Four sets of single-factor experiments were designed with surface roughness as the processing performance evaluation index,and the surface roughness values of each group were obtained after measurement to obtain the effects and trends of single-factor changes on surface roughness.According to the conclusions obtained from the results of single-factor experiments,the factor levels of orthogonal experiments were selected and a set of 4-factor,4-level orthogonal experiments were designed.The surface roughness and material removal rate of the workpiece after grinding and polishing with different combinations of parameters were obtained by 16 sets of orthogonal experiments.The order of the parameters affecting the surface roughness and material removal rate of the workpiece was obtained by the extreme difference analysis of the orthogonal experiments.(2)Using the data obtained from the orthogonal experiments,a prediction model of surface roughness was established by MATLAB software using the regression analysis method,and the significance test of the prediction model was conducted,and the established prediction model was significant at the confidence level of 0.05,indicating that the prediction effect of the model was good.The predicted values of surface roughness were obtained by the prediction model,and the measured values of the experimental results were compared with the regression predicted values.It was found that the relative errors of the predicted values of the regression prediction model were lower than 15% in 13 groups,and the errors were within the range of 15%-20% in 3 groups.(3)Using the orthogonal experimental data,a BP neural network with 4 inputs and 1output was written by MATLAB software,and the surface roughness prediction model was established by judging the root mean square error of different hidden layer nodes to select the best number of hidden layer nodes.After the prediction of BP neural network,the relative errors of surface roughness compared with the actual measurement results were within 1%.The relative error is significantly smaller than that of the regression analysis prediction model,so the established BP neural network prediction model is applied to the parameter optimization of aluminum alloy grinding and polishing.(4)Using the established BP neural network prediction model combined with the improved multi-objective particle swarm optimization algorithm to optimize the parameters of aluminum alloy grinding and polishing process,the roulette selection operator method is added on the basis of MOPSO algorithm to prevent the particle swarm from falling into the local optimal solution too early and improve the accuracy of the algorithm.A dual-objective optimization is performed with minimum surface roughness and maximum material removal rate,and the range of machining parameters is the constraint,and the optimized Pareto frontier is obtained by the algorithm,and a set of solutions is preferred from the set of 30 frontier solutions.Finally,the accuracy of the parameter optimization is verified by experiments,and the surface roughness of the workpiece is effectively reduced and the machining efficiency is improved by the parameter optimization.