Optimization And Development Of Selective Laser Melting (SLM) Process For 925Ag Metal Parts With Hollow Structure Features

Selective laser melting(SLM),a kind of 3D printing,is widely used in various fields due to its plentiful printable materials,flexible and complex printing structure.The high value-added of precious metals has become its conjunction with 3D printing.How to quickly determine the appropriate process parameters is the primary condition to ensure the quality of the product in precious metal SLM printing.However,the SLM printing process is complex,the process parameters are numerous and mutual coupling and restriction,coupled with the high cost of precious metal itself,repeated trial printing will cause material loss and cost increasing.Therefore,this paper takes the hollowed out 925 Ag parts as an example,on the basis of process parameters that are initial designed,combining the Simufact Additive finite element simulation and orthogonal experiment analyze the influence of the main printing process parameters on the quality of the parts.Then the prediction model of the deformation of the parts is established based on support vector machine(SVM),and the process parameters are optimized by genetic algorithm(GA),which provides a new method for the selection and regulation of SLM printing process parameters.The main research contents are as follows:(1)Based on the structural characteristics and material characteristics of the925 Ag hollow parts,the selective laser melt printing process of 925 Ag hollow parts was designed by combining the printing equipment with them.The selective laser melt printing process includ supports,placement,scanning strategy,process parameters,etc.and the finite element simulation analysis was based on Simufact Additive to prepare for the next research.(2)Based on the Simufact Additive finite element simulation,the weight order of the influence of different process parameters on the print quality is obtained through the analysis of the results of orthogonal experiment,and the superior combination of process parameters is obtained.A group of optimal process parameters is finally obtained through further simulation and comparative analysis.The method and results can provide theoretical guidance for related process parameters’ design and optimization.(3)Based on the Support Vector Machine(SVM)machine learning method,the prediction model of 925 Ag hollowed out parts deformation is obtained by comparing and analyzing the genetic algorithm(GA)and k-fold cross validation method,it’s combined with the sample set obtained by orthogonal experiment,.The establishment of prediction model replaces the tedious and time-consuming of finite element simulation,and provides reference for relevant personnel to formulate and optimize process parameters.(4)Based on GA＿SVM prediction model,the optimization of the process parameters by using genetic algorithm is established,which will be connect the mapping relationship of process parameters with deformation.and designs GUI system of the mapping relationship between process parameters and deformation.Finally,the simulation verification and enterprise production are done.It avoids the discreteness of orthogonal test and makes the test and selection of reasonable process parameters more convenient and intuitive.