An adaptive mutation operator for artificial immune network using learning automata in dynamic environments

Many real world problems are mostly time varying optimization problems, which require special mechanisms for detection changes in environment and then response to them. This paper proposes a hybrid optimization method based on learning automata and artificial immune network for dynamic environments, in which the learning automata are embedded in the immune cells to enhance its search capability via adaptive mutation, so they can increase diversity in response to the dynamic environments. The proposed algorithm is employed to deal with benchmark optimization problems under dynamic environments. Simulation results demonstrate the enhancements of our algorithm in tracking varying optima.