| The optimal construction direction of 3D printed parts needs to be determined from an infinite number of printable directions,which is a typical optimization problem.A variety of optimization criteria should be considered in the optimal forming direction.Aiming at how to unify various optimization criteria in the optimization of direction to balance the process demand,quality demand and economic demand in the construction process,a multi-objective optimization method can be introduced to effectively solve the problem.At present,great progress has been made in the research of multi-objective optimization methods.However,the application of existing optimization models and solving methods to3 D printing construction direction optimization problems has the following shortcomings:First,the relationship between the optimization criteria is not fully considered,and the existing objective models often assume that the optimization objectives are independent of each other,which is difficult to solve the problem of mutual influence of each objective in the actual direction optimization problem.Second,for the problem of slow calculation of the objective function,the existing research does not take into account that there are a lot of intermediate calculation steps when direction evaluation,and a lot of time needs to be waited for each evaluation according to the existing sequential algorithm.Therefore,in this paper,the cost model of metal 3D printing parts was established to solve the above problems,and a metal 3D printing direction optimization method was proposed based on TLBO algorithm.On this basis,combined with the parallel algorithm model,a fast automatic generation algorithm model for the optimal forming direction of SLM process parts in the field of metal 3D printing was proposed.The research content of this paper is as follows:(1)The cost model of metal 3D printing is established.In order to minimize the printing cost of parts,a comprehensive cost model including raw material cost,energy cost and indirect cost was established by considering the optimization criteria of the minimum overhang area,the shortest construction time and the best surface quality.This model is used as the objective function in the multi-objective optimization,and the weighted sum method is used to transform the multi-objective optimization problem with multiple cost models into a single objective optimization problem with only one comprehensive cost.(2)A SLM component orientation optimization method is designed.The STL model of SLM parts was input,the support structure was generated by Mesh Mixer software,and the volume of STL model,the volume of support structure,and the printing height of the parts were calculated.Then the metal 3D printing cost model was used as the evaluation function,and the TLBO algorithm was used to optimize the rotation Angle.To get the best construction direction with the lowest printing cost.Finally,the effectiveness and efficiency of the proposed method are verified by experimental cases and comparative analysis.(3)A fast direction optimization method for SLM parts is designed.A parallel computing model was used to solve the problem of slow evaluation of TLBO method in direction optimization,and a fast SLM direction optimization algorithm with heterogeneous memory was proposed.Two kinds of parallelization schemes are designed in this algorithm: 1.In the shared memory platform,the molecule group is divided based on Open MP,and each thread in the process corresponds to a subgroup.Among the subgroups,the teaching stage and self-study stage in each iteration are executed in parallel to complete a round of computing tasks.2.In the distributed memory platform,the population iteration times are divided based on MPI,and the work process and scheduling process are selected.According to the load balancing algorithm,the scheduling process assigns a round of calculation tasks to each work process,summarizes the optimization scheme results after calculation,and outputs the optimal construction direction according to the summarized results. |
