An Improved Genetic Algorithm Based on Variable Step-Size Search

Abstract

Genetic algorithms (GAs) are global optimization algorithms which can be used to solve different kinds of problems. However, in the situation that the size of feasible solution space is far less than that of search space, GAs may degrade to random searches. This paper presents an improved genetic algorithm, which adopts variable step-size algorithm to obtain a feasible solution, and reduce search space during the same process. The theoretical analysis and experiments in comparison with the random search are also presented,which indicate that this improved algorithm would reduce the search time in solving problems that have a large search space.