A new tool in electrostatics using a really-coded multipopulation genetic algorithm tuned through analytical test problems

We describe a new genetic algorithm (GA) to optimize multimodal continuous functions. It is based on a splitting of the traditional GA into a sequence of three processes. The first process creates several sub-populations using the information entropy theory. The second process applies the genetic operators on every sub-population. We then determine the best point s* among the best solutions issued from each of the preceding sub-populations, In the neighborhood of this point s* is generated a population used to initialize a traditional CA in the third process. Other features of our program must be pointed out: in particular, we use a real-value coding, more adapted to optimization handling continuous variables; the variety of the initial population is ensured by controlling its entropy. A comparison of performances with competitive metaheuristics is presented, using analytical test functions of which local and global minimums are known. Finally the tool is successfully applied to an electrode benchmark problem. (C) 2001 Elsevier Science Ltd. All rights reserved.