Artificial immune algorithm based on biological immune clone selection and immune network theory

The paper proposed an immune algorithm based on the mechanism exhibited in biological immune clone selection and the immune network theory. The algorithm eliminates antigen by clone selection and mutation, accumulates knowledge by selection of antibody memory cells which inspired by the immune network theory, and realizes immune homeostasis. The function of antibody memory cells is immune surveillance. The self-adaptation of antibody is realized by the immune mutation mechanism, which makes the algorithm self-learning and self-adaptive. The results of an application study on the pattern recognition of heat furnaces shows that the algorithm has good abilities of pattern recognition and data compress.