Optimal Design Of The Impeller Of The Fog Sprayer Axial Flow Fan Based On Surrogate Optimization Algorithm

In today’s economic globalization and increasingly fierce market competition,in order to improve the global competitiveness,enterprises must shorten the product design cycle,reduce the production cost and improve the product quality.Nowadays,high-precision simulation models are used in the optimization design of complex electromechanical products and systems,so as to improve the design quality and reliability.Because of the black-box nature of simulation model and the complexity of calculation,the heuristic optimization method based on swarm is greatly limited,and the optimization method based on explicit function and gradient information is also invalid.Therefore,the surrogate optimization method based on the surrogate model has been widely concerned and valued by scholars at home and abroad.Surrogate model is an approximate mathematical model,which is usually used to replace complex and time-consuming simulation model in the process of analysis and optimization design.The surrogate optimization method guides the new test sample points to converge to the global optimal solution quickly by constructing the surrogate model and designing a reasonable optimization strategy,and finally shortens the product design cycle,reduces the production cost and improves the product quality.To solve these problems,this paper proposes self-adapting surrogate optimization algorithm.The algorithm includes global exploration stage and local exploration stage.In the global exploration stage,IPI infilling criterion and parallel infilling strategy are proposed.In the stage of local exploration,new sample points are obtained by MP infilling criterion.Each infilling criterion is optimized by differential evolution algorithm,and according to the relationship between new sample points and known sample points,the adaptive switching between global exploration and local exploration is realized until the optimal solution is found.Then,the proposed algorithm,EI,PI,and MP infilling criterion are respectively used to optimize the three typical examples.Through comparative analysis,it proves that the proposed algorithm improved the optimization efficiency and accuracy,and at the same time,the robustness of solving the problem became better.In order to further verify the engineering applicability of the algorithm proposed in this paper,the impeller of the fog sprayer axial flow fan is selected as the research object.Firstly,the structural parameters of impeller which have a great influence on the range of the fog sprayer axial flow fan are determined.Then,these parameters are substituted as design variables into the proposed self-adapting surrogate optimization algorithm for optimization.The optimal solution is obtained by iteratively increasing sample points until the optimization termination condition is reached.The final optimization results show that the range of the fog sprayer axial flow fan is increased by24.77%.Therefore,the algorithm in this paper provides a new idea and reference for the field of surrogate optimization methods.