反馈混沌遗传算法及其在约束优化中的应用

Chaotic Genetic Algorithm with Feedback and Its Applications to Constrained Optimzation

针对现有遗传算法中普遍存在的早熟与收敛慢的问题,将混沌映射和后天强化学习策略引入到标准遗传算法中,提出了带反馈的混沌遗传算法.该算法通过混沌映射来保持演化群体良好的多样性;通过基于Baldwin效应的后天强化学习来克服纯粹的随机演化.对复杂约束优化问题——基准问题的数值实验验证了文中算法的高效性及鲁棒性.

The existing genetic algorithms are generally lost in a dilemma between prematunty and stow-convergence. In order to solve this problem, a new chaotic genetic algorithm with feedback is proposed by introducing the chaotic mapping and the posterior reinforcement learning in the standard genetic algorithm. In this algorithm, the evolution population maintains a good diversity via the chaotic mapping, and the stochastic evolution is overcome by the posterior reinforcement learning based on Baldwin effect. Numerical experiments are finally carried out aiming at the complex constrained optimization problems, namely the benchmark problems. The results show that the proposed algorithm is effective and robust.