【摘要】 提出了一种基于外部档案(EA)的多目标人工蜜蜂群优化算法(MOABC)。MOABC算法通过帕累托支配(Pareto Domination)的概念比较解的优劣,并通过建立EA以保存找到解集。同时为了使解集分布更均匀,采用广泛学习策略保证算法的多样性。通过3个多目标测试函数上的试验结果,验证了MOABC优于多目标粒子群优化算法。然后将MOABC应用于求解两目标的微电网调度问题。对多微电源的微电网调度进行实例仿真。结果显示,MOABC在获得最优Pareto解集、降低计算复杂度、提高收敛效率等方面具有很强的优越性。更多还原

【Abstract】 This paper presented an external archive based multi-objective artificial bee colony optimization algorithm( MOABC). The proposed MOABC algorithm uses the concept of pareto domination tocompare the advantages and disadvantages of pareto solutions,and through the establishment of an external archive( EA) to save the solution set. In order to make more evenly distributed solution set,the MOABC model uses comprehensive learning strategy to ensure the algorithmic diversity. The experimental results on three multi-objective benchmarks verify the superior performance of MOABC compared to multi-objective particle swarm optimization( MOPSO).Then the work applied MOABC to solve two-objective schedule dispatch problem in Microgrid. The simulation results on microgrid system show that MOABC can obtain optimal Pareto set,reduce the computational complexity,and improve the convergence efficiency.更多还原