Immune Algorithm in Non-stationary Optimization Task

Optimization task in the non-stationary environment is a specific kind of challenge. Not many algorithms well cope with this issue. One of these algorithms is the immune algorithm. The behaviour of clonal selection algorithm is discussed in the paper. Basing on it the algorithm modification for working in the non-stationary environment and the possibilities of its work monitoring were proposed as well. The algorithm modification in the behaviour of antibodies implements the mechanism of autoimmunization. Issues of group immune system and possibilities of genetic material exchange between individuals in group were also discussed in the paper. This discussion was exemplified by experimental results which confirm the considered assumptions.