Optimization of mechanical characteristics of short glass fiber and polytetrafluoroethylene reinforced polycarbonate composites using the neural network approach

This study analyzed variations in mechanical properties that depend on the injection molding process during the blending of short glass fiber (SGF) and polytetrafluoroethylene (PTFE) reinforced polycarbonate (PC) composites. Experiments were planned according to a d-optimal mixture design (DMD) method. A hybrid method integrating back-propagation neural networking (BPNN), genetic algorithm (GA), and simulated annealing algorithm (SAA) is proposed for use in determining the optimal mixture ratio settings. The results of a DMD experimental run were used to train the BPNN in predicting mechanical properties and then the GA and SAA approaches were applied to individual searches to find the optimal mixture ratio settings. In addition, analysis of variance (ANOVA) was applied to identify the effect of mixture ratio of SGF and PTFE reinforced PC composites for the ultimate strength, flexural strength and impact resistance. The results show that the combinations of BPNN/GA and BPNN/SAA methods are effective tools for the optimization of the reinforced process.