Research On Preheating Temperature Control And Process Parameters Optimization For Selective Laser Sintering

Selective laser sintering(SLS)is an additive manufacturing technology,which uses a highenergy laser beam to sinter and build up parts layer by layer based on a discrete digital model file generated by computer software.During recent years,it has been applied to different degrees in aerospace,automotive,marine,medical,and other fields.Due to the “layer-by-layer sintering and stacking” characteristic of selective laser sintering,stable preheating temperature and reasonable process parameters are the key factors to ensure the forming precision.However,at this stage,the temperature control scheme of the selective laser sintering preheating process generally adopts proportional integral differential(PID)feedback control to maintain the preheating temperature,since there exist hysteresis links in the controlled object,this method often leads to unsatisfactory final molding results because of poor response speed and overshoot.At the same time,the process parameters required for different materials are often varied,but there is no specific scientific guidance to determine these parameters,which are largely dependent on historical experience.This methodology not only costs a lot of time,but also needs to be enhanced in terms of accuracy.In order to further effectively improve the quality of formed parts by selective laser sintering,this paper considers the preheating temperature control and process parameter selection as the research emphasis,and specifically carries out the following work:Considering the tubular preheating device as the research object,the mathematical model of the controlled object in the preheating process is derived.Based on this,an improved Smith prediction fuzzy PID temperature control strategy is proposed,which combines fuzzy control with PID control to achieve adaptive adjustment of control parameters and introduces an improved Smith predictor to compensate for the pure hysteresis link.To verify the effectiveness of the proposed method,a simulation model is constructed in MATLAB/Simulink software to investigate its dynamic performance.The preheating temperature control system is designed based on the previous temperature control scheme,including both hardware design and software design.In terms of the hardware,the selection of the main control chip,solid-state relay,temperature sensor,and human-machine interaction module is completed,and then the power supply circuit,communication circuit,and temperature control circuit are drawn.As for software,five different programs are developed,namely,the main program,communication program,pulse width modulation control program,temperature control program,and data acquisition filter program.Five process parameters,including laser power,preheating temperature,scanning speed,scanning spacing,and delamination thickness are selected for orthogonal tests.The shrinkage rates of the formed parts in X,Y,and Z directions are separately measured to establish a unified objective function.Meanwhile,a novel improved hybrid Aquila optimizer and African vultures optimization algorithm(IHAOAVOA)is applied to optimize the least squares support vector machine(LSSVM),thus completing the initial establishment of the IHAOAVOA-LSSVM prediction model.The above model is trained by using the sample data set,and then the uniform performance of the formed parts under each combination of process parameters is traversed to obtain the optimal process parameters.The above-mentioned preheating temperature control system is installed on the test prototype,and the interface configuration and electrical connection are completed to test the response speed,temperature control accuracy,and partition molding effect of the system.Additionally,the generalization capability of the IHAOAVOA-LSSVM prediction model is evaluated under the customized process parameter combinations.The optimal process parameters predicted by least squares vector machine,BP neural network,extreme learning machine,and IHAOAVOA-LSSVM are further used to guide the processing,and the superiority of the proposed prediction model is highlighted by comparing the dimensional accuracy and mechanical properties of the formed parts.The experimental results demonstrate that the preheating temperature control system of selective laser sintering based on the improved Smith predictive fuzzy PID can significantly reduce the preheating time of the powder material and improve the temperature field uniformity as well as forming precision.And the optimization of the process parameters by the IHAOAVOA-LSSVM prediction model can also be helpful to boost the dimensional accuracy and mechanical properties of the fabricated parts.The two complement each other and are of great significance to promote the widespread application of selective laser sintering technology.