改进的花朵授粉算法在微网优化调度中的应用

摘要 研究了用于解决微网优化调度问题的群智能算法.针对微网优化调度问题的多目标、多约束条件等特点，对微网优化调度问题建模;提出了改进的花朵授粉算法，并将其应用到微网优化调度问题.在初始化时，采用对立点方法增加种群多样性和优化搜索空间;局部更新时，使用一种新的局部更新算子提高算法收敛速度;此外，为了减少计算量和避免陷入局部最优，定义了是否使用遗传操作的判断条件.仿真结果表明，该算法性能优于原始花朵授粉算法和遗传算法等其他算法.
Abstract：The swarm intelligence algorithm for solving the problem of optimization dispatch of microgrid was investigated. A model for optimization dispatch of microgrid was established under the consideration of the process characteristics， such as multi-objective and multi-constraints. A modified flower pollination algorithm (MFPA) was proposed and applied to the optimization dispatch of microgrid. During initialization， opposition method was utilized to improve the diversity of the population as well as fully explore the space. During local updating， the new operation can accelerate the convergence. In addition， the condition for using genetic operations was defined in order to reduce the calculation and avoid the local optimal solution. Simulation results demonstrated that the performance of MFPA was better than those of FPA， GA and several other algorithms.