Optimization of cooling system for plastic injection molding using artificial neural network and genetic algorithm

The widespread application of thermoplastics in almost every area of the modern industry results in an increasing requirement for injection molds that must satisfy the precise specification of high quality parts. In this study, an optimization approach of process parameters during plastic injection molding is presented in a unified way by using artificial neural network and genetic algorithm. First, a multilayer back-propagation artificial neural network model is developed to map the mathematical non-linear relationship between process parameters and quality characteristics of the molded parts. To build the model, training and testing databases were conducted utilizing finite element simulation software Moldflow. Afterwards, a multi-objective optimization model of a plastic injection molding system was established by adopting proposed network model and genetic algorithm, cooperatively. The optimization method is applied in the process optimization for an industrial component. The warpage of final part as well as solidification time during plastic injection molding are investigated as the optimization objectives. Additionally, mold temperature, melt temperature, holding pressure, cooling cannels diameter and their configurations, are considered to be the design variables. The case study demonstrates that the proposed optimization method can adjust the process parameters accurately and effectively to satisfy the demand of real manufacture.

Keywords:

Plastic injection molding , Optimization , Artificial neural network , Genetic algorithm , Moldflow