A Novel Cooperative Immune-based Algorithm for Optimization Problems

In this paper, a novel cooperative immune-based algorithm is proposed to optimize optimization problems. In contrast to other IBAs, which usually are based on one or two features of the immune system, it adopts almost all of the immune system’s features. These features include generating immune cells in bone marrow, immune memory, immune network theory, negative selection algorithm, and clonal selection algorithm, which led to the method’s high performance. In this algorithm, each dimension is evaluated separately and divided into several regions. This not only helps the algorithm efficiently locate the global optimum but also increases NCIBA’s convergence speed. To examine the algorithm’s efficiency, several experiments are conducted on CEC2020 competition on single objective bound constrained numerical optimization problems. The results show the high efficiency of NCIBA when the number of regions grows. Also, according to the experimental results, the algorithm has a high efficiency as the number of dimensions increases.

Keywords:

Artificial immune system , cooperative approach , global optimization , negative selection algorithm , clonal selection algorithm.