An environment-gene double evolution immune clone algorithm for constrained optimization

By referencing the effect of environment to the biologic evolution, an environment-gene double evolution immune clone algorithm (EGICA) is proposed based on normal immune clone algorithm. This algorithm can avoid blind search effectively and enhance the convergence speed since an environment-gene double evolution mutation operator is introduced, which can accumulate the experience of evolution process. In other words, it means that EGICA has self-learning capability. Moreover, a new sequencing strategy is used for design cost function to solve constrained Optimization. Then, the convergence of EGICA is proved with probability 1. At last, by the experiments of testing 13 classical benchmarks of constraint optimal problem, it shows that EGICA has good capability.