A Cloud-based Artificial Immune Network for Optimization

This paper proposes an artificial immune network based on the cloud model (AINet-CM) for complex optimization problems. By introducing the cloud model to evaluate the candidate antibodies, three major immune operators are redesigned to enhance the convergence performance of AINet-CM. These operators are the increasing half cloud-based cloning operator, the asymmetrical cloud-based mutation operator and the normal similarity cloud-based suppression, respectively. Also, a dynamic searching step length is considered. A series of numerical simulations are arranged for technical investigations and three artificial immune systems (i.e., opt-aiNet, IA-AIS and AAIS-2S) are selected for comparison with AINet-CM. The experimental results suggest that the proposed AINet-CM algorithm outperforms the other three immune algorithms in convergence speed and solution accuracy.