Open AccessMulti-objective Optimization of Injection Process Parameters Based on EBFNN and NSGA-II
Author(s) -
Yao Lu,
He Qing Huang
Publication year - 2020
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1637/1/012117
Subject(s) - latin hypercube sampling , shrinkage , multi objective optimization , volume (thermodynamics) , process (computing) , molding (decorative) , mathematical optimization , set (abstract data type) , computer science , reliability (semiconductor) , mathematics , engineering , mechanical engineering , statistics , monte carlo method , power (physics) , physics , quantum mechanics , programming language , operating system
Taking the process parameters and quality index data obtained from the optimal Latin hypercube test of a front hood inner panel as sample data, this paper constructs an EBFNN approximate model by combining the NSGA-II multi-objective optimization algorithm to perform multi-objective optimization of injection molding process parameters, obtained Pareto solution set for volume shrinkage and warp during ejection. Comparing the Moldflow numerical simulation results of the four relatively optimized Pareto solutions, the optimal combination of process parameters is obtained. The optimized volume shrinkage and warp deformation reduced by 50.518% and 39.845% respectively, which verified the reliability and practicability of this method, and had certain practical reference significance for the optimization of injection molding process of other plastic parts.