Warpage Optimization of Polystyrene in Selective Laser Sintering Using Neural Network and Genetic Algorithm

摘要

Warpage is a crucial factor to accuracy of sintering part in selective laser sintering (SLS) process. In this study, optimum values of process parameters in selective laser sintering polystyrene to achieve minimum warpage are determined. A predictive model for warpage is created using back propagation artificial neural networks (ANNs) with the process parameters of laser power, scanning speed, hatch spacing, layer thickness and powder temperature. Neural network model validated using experimental data for predictive capability and then integrated with an effective genetic algorithm (GA) to find the optimum process parameters. Based on the simulation results, the influences of process parameters on warpage are investigated. The results showed that laser power affects warpage most followed by powder temperature. The layer thickness and hatch spacing have less influence on warpage. Scanning speed shows the least impact.