The paper addresses the relevant issue of enhancing the efficiency and reliability of mold design for polymer injection molding. Traditional engineering calculation methods, relying on sequential parameter search, often prove labour-intensive and unable to achieve a global optimum of the design, particularly when dealing with multiple conflicting criteria such as minimal weight, maximal rigidity, uniform temperature distribution, and minimization of casting defects. The aim of the study is to improve productivity and enhance the quality of polymer injection molding production. The objective is to optimize the effectiveness criteria of mold design for polymer injection molding. The research methods rely on the principles of fundamental theories of geometric modeling, bio-inspired optimization algorithms, genetic algorithms, and particle swarm optimization. The novelty of the work lies in adapting hybridization operators, enabling the incorporation of the directed movement mechanism of particle swarms into genetic search to overcome premature convergence and facilitate abrupt exploration of the design variable space. The study results demonstrate that the proposed hybrid algorithm is 22% faster in convergence speed compared to classical genetic algorithms and yields Pareto-optimal front solutions, achieving a 15% reduction in tool mass and simultaneously decreasing maximum thermal deformations by 8% compared to the original design. The findings state that the developed hybrid algorithm of bio-inspired optimization proves to be an effective tool for supporting decision-making during the conceptual design of complex casting equipment.
Yarovoy et al. (Mon,) studied this question.