Abstract: Taking the central distance of cooling water channel, the distance from the water channel center to the die wall, and the cooling channel diameter of the plastic chain plate injection mold as the experimental factors, a second-order response surface model between the three experimental factors, the temperature difference of the model wall and the cooling time was constructed by a response surface method. The Pareto genetic algorithm was used to iteratively optimize the two response surface models, and a set of optimum solutions for design parameters of cooling water channel was obtained when the temperature differences of the die wall and cooling time were both small. The results indicated that the optimum design parameters of the cooling water channel were follows as: the central distance of cooling water channel was 30 mm, the distance from the center of the channel to the die wall was 15 mm, and the diameter of the channel was 10 mm. The corresponding temperature difference of die wall was determined as 7.61 ℃, and the cooling time was determined as 62.74 s. Such data was smaller than the experimental results of any groups of response surface. This analysis method was proved to be feasible.

Key words: injection mold, cooling water channel, optimization of design parameter, response sur face method, Pareto genetic algorithm