Abstract
From the hint of the chaotic system and immune system, a chaotic search-based adaptive immune genetic algorithm (CSAIGA) is presented to improve the genetic algorithm (GA). Taking advantage of the characteristics of chaotic system, the CSAIGA produces the initial population by chaotic iteration, and performs the chaotic local search in the antibody neighborhood of memory population to improve the local search ability and computation efficiency. Learning from the basic principles of the immune system, the CSAIGA employs the selection mechanism based on the affinity and concentration of antibody, and introduces the chaotic replacement operation to maintain the diversity of population and avoid the premature convergence. In addition, the CSAIGA adjusts the mutation probabilities adaptively in response to the affinities of antibodies, so as to improve the global convergence further. The experimental results show that the algorithm can stably converge to the global optimum with lower calculation cost. It is a fast and efficient global optimization algorithm.