Application of artificial immune algorithm to multimodal function optimization

It is difficult for traditional search methods and simple genetic algorithm to solve multimodal function optimization problems. Based on the idea of affinity maturation of immune cells in the germinal centers (GCs) of natural immune systems, a new parallel optimization algorithm is proposed. The algorithm consists of a set of operators such as hypermutation, selection, memory and similarity suppression of antibodies. The simulation results show that the proposed algorithm not only can obtain a set of solutions including the global optimum and multiple local optima, but also has much less computational cost than other algorithms. Finally, the features of the algorithm are analyzed, and the user-defined parameters´ influence on the algorithm performance is discussed in detail.