Research Of Multi-Modal Optimization Algorithm Based On Immune Evolution And Excellent Sub-Population Migrating

The multi-objective optimization has some apply value, especial in economy, management, military fields, society and so on. But this traditional method has the limitation very much, often only can search the partial solution, not the global optimal solution. The evolution algorithm use natural organism's principle and mechanism, having the auto-adapted environment ability. It searches the solution quickly, easy to apply. The massive facts indicated the mechanism of evolution algorithm suits solves the multi-objective optimization question mostly.The genetic algorithms and the immunity algorithm apply extremely widespread. Genetic algorithms have many merits specially: the intelligence, the global convergence property, hides parallelism, robustness strongly and so on. Memories, Recognition, Feature Extraction, Dynamic protection, Self-regulation are the special mechanism of the immune system, which has good application value. These two algorithms can overcome in other algorithms effectively the precocious phenomenon, the community multiple insufficiencies and the search speed slow.In this paper, a new multi-modal immune and genetic optimization algorithm based on immune mechanism and genetic algorithm with strategy of excellent sub-population migrating is developed. The immune mechanism is inspired by the idea of affinity maturation of immune cells in the evolutionary process. The main operators of the algorithm included hyper-mutation, positive selection to produce memory cells and excellent sub-population, similar cell suppression, reorganization operator. Used to optimize different multi-modal testing functions and compared with some similarity algorithms, testified that it has both validity and less computational magnitude.