Immune Evolutionary Algorithms And Its Applications In Function Optimization Problems

Function optimization problems (FOPs) belong to a kind of mathematical programming problems, which are frequently encountered in the disciplines of science and engineering application. So the search on it is important in theoretics and practice. Immune Algorithms (IAs) are a kind of simulated nature algorithms (SNAs), which developed byArtificial Immune Systems------a kind of computation intelligence thatrecently new developed.Evolutionary algorithms (EAs) are a kind of randomly search algorithm, which simulates the process of natural evolution. During the past decade, Immue evolutionary algorithms (IEAs) have been broadly applied to solve FOPs, and considerable function optimization algorithms (FOAs) have been proposed.Firstly, this dissertation introduces the biology background, relative theory and applications of IAs. Then, this description of the origin, advance, three main branches (i.e., genetic algorithms, evolutionary programming and evolutionary strategies) and applications of EAs is given.Afterwards, two immune evolutionary algorithms proposed by author. The main contribution and work are described as following:1) The algorithm, which considering the drawback of easily being trapped in a local optimal solution and slow convergence velocity of clone selection of immune algorithm, was named new immune evolutionary algorithm(NIEA) which proposed by the basement on the interior study mechanism of biological immune system and evolutionary algorithm. The new algorithm included orthogonal crossover, simplex crossover, clone, multipolar-mutation and selection. The idea of evolutionary computation was integrated into clone selection and a new mutation operator was proposed. This new algorithm can guarantee the diversity of the population and improve the global search ability. Theoretical analysis proved that NIEA converges to the global optimum. Different functions were utilized to test this method and the simulation results suggest that this algorithm has good performance.2) Proposed novel immune evolutionary algorithm based on local search mechanism (IEALSN) based on immune algorithm and differential evolution algorithm. Firstly, initial population is randomly generated. Secondly, simplex crossover operator is used to population to generate offspring. After that a local search mechanism is designed to improve the ability of local search and the speed of converge, which contains clustering and a new differential evolutionary (NDE). Theoretical analysis proved that IEALSN converges to the global optimum. Then, the updating method of immune algorithm is used to update population. Different functions were utilized to test this method and the simulation results suggest that this algorithm has good feasibility and performance.