Research On Artificial Immune Algorithm And Its Application In Optimization Problems

Artificial immune algorithm is a kind of new intelligent optimization algorithm which is inspired by biological immune system. Because this algorithm combines the prior knowledge and the adaptive ability of immune system, it has some characteristics as follow: robust in information processing; not requiring derivable additional information of the objective function in solving optimization problem; be able to find better global optimal solution in the process of searching; Now this kind of algorithm has been used in many fields in which showing excellent performance and efficiency, such as machine learning, unconventionality and malfunction diagnosis, simulation of the behavior of robots, control of robots, intrusion detection of networks, function optimization and so on, so it is considered to be one of the most potential intelligent search algorithms.Based on the proposed artificial immune algorithm and the principle of the immune system, we have researched on artificial immune system and its application in optimization problems in this paper, and the major contributions are showed as follows:(1) According to the characteristic that the immune system can accomplish quickly the purpose of identifying the antigen by injecting vaccine, the immune genetic algorithm founded on dynamic vaccine (IGAB) which is applied in TSP is proposed. The simulation show that the IGAB can prevent the algorithm degenerative effectively during the process of optimization of the genetic algorithm, and improve the convergent speed of the algorithm.(2) Based on the mechanism that the Clonal selection mechanism of B cells is one of the important approaches to generates diversity of antibodies, the improved adaptive clonal selection algorithm (IACSA) is proposed. We applied the IACSA in multidimensional function optimization problems, and the experiment results show that the IACSA can restrain the stagnation at the end of iterative optimization, and improve the quality of the solutions. (3) According to the principle that the B cells can produce kinds of immune cells which have the stronger identical ability by mutation and reproduction of the gene in its body, the clonal selection algorithm with hyper mutation and spatial clone extension (HSCSA) is proposed. We applied the HSCSA in multimodal function optimization, and the experiment results show that the HSCSA can not only improve the quality of the solutions, but also make the antibody escape from local optima, and find the global optimum in the whole solution space finally.At last, summarized the research results and looked forward application foreground for artificial immune algorithm.