Optimization of injection moulding conditions with user-definable objective functions based on a genetic algorithm

Abstract

This paper applies a Genetic Algorithm (GA) method to optimize injection moulding conditions, such as melt temperature, mould temperature and injection time. A GA is very suitable for moulding conditions optimization where complex patterns of local minima are possible. Existing work in the literature has limited versatility because the optimization algorithm is hard-wired with specific objective function. However, for most of the practical applications, the appropriateness of optimization objective functions depends on each specific moulding problem. The paper develops a multi-objective GA optimization strategy, where the objective functions may be defined by the designers, including using different criteria and/or weights. For parts with general quality requirements, an objective function is also recommended with some quality measuring criteria, which are either more accurately represented or cover more moulding defects than those from existing simulation-based optimization approaches. The paper also elaborates on the effective GA attributes suited to moulding conditions optimization, such as population size, crossover rate and mutation rate. A case study demonstrates the effectiveness of the proposed approach and algorithm. The optimization results are compared with those from an exhaustive search method to determine the algorithm's accuracy in finding global optimum. It is found to be favourable.

Acknowledgements

Work was supported by the Academic Research Fund, Ministry of Education, Singapore and Moldflow Corporation.