基于学习的并行免疫量子进化算法及收敛性

Parallel Immune Quantum Evolutionary Algorithm Based on Learning Mechanism and Its Convergence

提出了基于学习的多宇宙并行免疫量子进化算法,算法中将种群分成若干个独立的子群体,称为宇宙。宇宙内采用免疫量子进化算法,宇宙间采用基于学习机制的移民、模拟量子纠缠的种群交叉等信息交互方式,使得进化算法具有更好的种群多样性,更快的收敛速度和全局寻优能力。不仅从理论上证明了该算法的收敛,而且通过仿真实验表明了该算法的优越性。

A Multi-Universe Parallel Immune Quantum Evolutionary Algorithm based on learning mechanism （MPMQEA） is proposed,in the algorithm,all individuals are divided into some independent subcolonies ,called universes. Each universe evolving independently uses the immune quantum evolutionary algorithm ;Information among the universes is exchanged by adopting emigration based on the learning mechanism and quantum-cross simulating entanglement of quantum. It not only can maintain quite nicely the population diversity,but also can help to accelerate the convergence speed. The convergence of the MPMQEA is proved and its superiority is shown by some simulation experiments in this paper.