Optimization Of Selective Laser Melting Process Based On Neural Network And Genetic Algorithm

As one of the most widely used metal additive manufacturing technologies,selective laser melting technology has significant advantages in shortening the manufacturing cycle,reducing costs,and personalized manufacturing.However,there are too many influencing factors in the selective laser melting forming process.If the process parameters are not selected properly,not only the forming process will be unstable,but also the internal defects of the formed part will be generated,which will seriously affect the quality of the formed parts.Therefore,the adjustment and optimization of the process parameters of selective laser melting forming is the key to improving the quality of formed parts.In this paper,the selective laser melting equipment WXL-120 P independently developed by Xiamen Wuxinglong Technology Co.,Ltd.is used as the test platform,and 316 L stainless steel powder is used as the test material to optimize the relative density of the formed parts.The main research contents are as follows:(1)The selective laser melting forming experiment is carried out,and the mapping relationship model between the selected laser melting process parameters and the relative density of the formed part is established by using the BP neural network which has high approximation ability to nonlinear problems.Six sets of test data other than the training sample data are selected to perform a simulation test on the network.The relative error between the test value and the test value of the test sample is less than 8%,which indicates that there is a good mapping relationship between the process parameters and the density value of the formed parts.The model is suitable for the study of the process parameters of selective laser melting.(2)Aiming at the problem that the initial weight of the neural network is too sensitive,the global search function of genetic algorithm is used to optimize the initial weight and threshold of the BP neural network.After the optimization,the maximum relative error of the neural network test sample drops from 7.38% to 4.19%,the average absolute error dropped from 2.8709 to 1.1453,and the average absolute percentage error dropped from 3.67% to 1.46%,which effectively improved the prediction accuracy and stability of the model.(3)The optimized model is transformed into the fitness function of genetic algorithm,and the process parameters are optimized.The optimized process parameters are verified,and it is found that the formed samples are nearly full density and has no obvious metallurgical defects,which proved the feasibility and practicability of the optimization method.At the same time,the influence of process parameters on the density of formed parts is analyzed.(4)In order to facilitate the realization of the selective laser melting forming process parameter optimization process,use the GUIDE of MATLAB software to design the selective laser melting process parameters optimization interface system.Through the system,we can quickly realize the neural network modeling,prediction and process parameters optimization quickly.