Evolutionary Algorithm Based On Immune Strategy

Lei Wang,Licheng Jiao
IF: 1.019
2001-01-01
Chinese Journal of Electronics
Abstract:A novel evolutionary algorithm, evolution-immunity strategies(EIS), is proposed with reference to tire immune theory in biology, which constructs an immune operator accomplished by two steps, a vaccination and an immune selection. The aim of introducing the immune concepts and methods into ES is for finding the ways and means obtaining the optimal solution of difficult problems with locally characteristic information. The detail processes of realizing EIS are presented which contain G steps. EIS is analyaed with Markovian theory and it is approved to be convergent with probability 1. In EIS, an immune operator is an aggregation of specific operations and procedures, and methods of selecting vaccines and constructing an immune operator are given in this paper. It is shown with an example of tire 442-city TSP that the EIS can restrain the degenerate phenomenon during the evolutionary process by simulated calculating result, improve the searching capability and efficiency, and therefore, increase the convergent speed greatly.
What problem does this paper attempt to address?