多智能体量子多目标进化算法及其在EELD问题中的应用

Multi-agent Quantum Multi-objective Evolutionary Algorithm and Its Application to EELD Problem

环境经济负荷分配问题是电力系统中重要的多目标优化问题。求解多目标优化问题的关键在于找到尽可能多的Pareto最优解。在基于量子进化理论,智能体的竞争、学习能力和生物的进化策略的基础上,提出了一种用于求解多目标优化问题的量子编码的多智能体进化算法。该方法将智能体分布在多智能体网络环境中,智能体之间通过量子进化来生成问题的可行解。将该算法应用于经济环境负荷分配的两目标(燃料成本和NOx排放)与三目标(燃料成本,NOx排放和SO2排放)优化问题,通过与经典多目标优化算法进行比较,表明了该算法的有效性。

Economic/emission load dispatch （EELD） is an important multi-objective problem in electric system. The target of solving multi-objective optimization problems （MOOP） is to find as many Pareto optimal solutions as possible. A new algorithm aimed to solve MOOP was proposed—multi-agent quantum multi-objective evolutionary algorithm （MAQMOEA） based on quantum mechanics theory,the study and competition ability of multi-agent system and organic evolutionary strategy. In the multi-agent system,agents interactive with neighborhood are to produce solutions under the quantum evolution mechanism. The new algorithm was applied to EELD,including bi-objective （fuel cost and NOx emission） and three-objective （fuel cost,NOx emission and SO2 emission） optimization problems,and the effectiveness of the new algorithm was demonstrated compared with the classical multi-objective optimization algorithm.