摘要
Abstract
In order to address the problem like the inefficiency of distributed power co-optimization in mi-crogrid economic scheduling and the problem that traditional algorithms are prone to falling into local opti-mum,this study proposes an improved flower pollination algorithm for multi-objective optimal scheduling of microgrids.Firstly,it constructs an economic scheduling model including photovoltaic,wind turbines,energy storage and diesel generators,and establishes correlated constraint conditions.Secondly,aiming at the shortcomings of the standard flower pollination algorithm,such as slow convergence speed and low op-timization accuracy,this study designs an improved flower pollination algorithm.This algorithm adopts op-position-based learning to generate the initial population and an adaptive mechanism to dynamically adjust the conversion probability,and the influence of the parameters of the improved flower pollination algorithm is discussed.Finally,in the optimal scheduling of microgrids,the improved flower pollination algorithm,particle swarm optimization and the standard flower pollination algorithm are used to solve the problem simultaneously,and their optimal solutions are compared.The results show that the improved flower polli-nation algorithm is superior to the standard flower pollination algorithm and particle swarm optimization in terms of the convergence speed,the convergence accuracy and optimization results.Moreover,according to the output of each distributed power source,the improved flower pollination algorithm reduces the total grid cost in the microgrid optimal scheduling model,which verifies the effectiveness of the algorithm.关键词
花授粉算法/微电网/优化调度Key words
flower pollination algorithm/microgrid/optimal scheduling分类
信息技术与安全科学
