Optimizing the shape of an impeller using the differential ant-stigmergy algorithm

Abstract

A metaheuristic optimization algorithm for solving multi-parameter optimization problems is presented. The algorithm is applied to a real-world problem, where the aerodynamic power efficiency of the radial impeller of a vacuum cleaner is optimized. Here, the radial impeller is presented using parametric modeling. Due to the large number of parameters and, consequently, the enormous search space, an efficient metaheuristic approach is inevitable. Therefore, the so-called Differential Ant-Stigmergy Algorithm, which is an extension of the Ant-Colony Optimization for a continuous domain, is applied. The result of this is that the aerodynamic power of the radial impeller is increased by twenty percent. © 2008 Springer-Verlag Berlin Heidelberg.