A New Immune Algorithm and Its Application

The traditional single clonal selection algorithm has a lot of disadvantages, for example, it is easy to be trapped into local optima, and it has a lot of massive redudacy iteration in its later period and inferior global search ability and so on. In this paper, a new artificial immune algorithm is proposed based on the clonal selection theory and the structure of anti-idiotype(IAAI), which is improved as follows: firstly, IAAI constructs a dynamic clonal expansion, secondly clonal mutation, thirdly dislocated line clonal recombinant, lastly clonal selection. Through the above mentioned several key steps, IAAI is improved in order to achieve the evolution of the whole antibody population, then the new algorithm can have strong search capabilities which make it reach better performance by perfoming global search and local search in many directions in the solution space. Then the global convergence of the new algorithm is analyzed from the test of several typical complex functions. The result shows that the algorithm can effectively overcome the premature problem, and improve the ability of global optimization, the speed of convergence. Key-Words: immune algorithm, clonal selection, dynamic clonal expansion, clonal mutation, dynamic adjustment, anti-idiotype