A danger theory based immune network dynamic optimization algorithm

作者：Zhang Ruirui^*; Li Tao; Xiao Xin; Shi Yuanquan

来源：Journal of Information and Computational Science, 2013, 10(6): 1829-1838.

DOI：10.12733/jics20101624

• Artificial immune
• Classical dynamics
• Danger theories
• Dynamic functions
• Dynamic optimization
• Environmental change
• Optimization algorithms
• Population diversity

Although the existing artificial immune dynamic optimization algorithms have dynamic tracking abilities, but the errors are large. This paper proposes a danger theory based immune network dynamic optimization algorithm, named ddt-aiNet. By defining the danger zone to calculate danger signals for each antibody, the algorithm adjusts antibodies' concentrations through their own danger signals, which triggers immune responses of self-regulation. So the population diversity can be maintained. And the algorithm sets detection antibodies in the solution space. Through monitoring the danger signals of these detection antibodies, the algorithm can perceive the environmental changes, and then re-initialize the population in proportion. Experimental results show that the algorithm is better than the classical dynamic optimization algorithm dopt-aiNet. It has smaller errors, and can better track the environmental changes.

• 出版日期2013
• 单位四川大学