摘要
在考虑风功率/负荷预测误差以及常规机组随机停运的基础上,构建了电网日前风电接纳模型,在运行方式优化的基础上对风电接纳能力进行评估。模型是典型的多目标优化模型,具有风电接纳能力最大与发电成本最小两个互相冲突的优化目标。采用非支配分类遗传算法对其进行求解,获得了Pareto最优解集,给出了日前自然风电接纳电量与最大理论风电接纳电量。在Pareto最优解集的基础上,对不同风电接纳水平下的常规系统平均发电成本以及平均风电接纳成本进行了分析。基于IEEE 118节点系统的仿真实验验证了本文所提模型及算法的有效性。评估模型给出的风电接纳电量、接纳成本等信息可为调度决策提供有益的参考。
With considerations of the uncertainties on wind power/load prediction errors and random outages of generators, a day-a- head assessing model is presented to assess wind power integration capability on the basis of operation senario optimaztions. The model presented here is a typical multi-objectively optimization formulation, and its two optimization objects, i. e. , wind power integating capacties maximization and generating costs minimzation are inherently conflicting each other and can not get their optimal results sim- ultaneously. Non-dominated sorting genetic algorithm （NSGA） is utilized to obtain Pareto optimal result set of the formulation, from which, day-ahead minimum and maximum wind power injection amounts can be obtained. On the basis of Pareto optimal result set ob- tained by NSGA, average generation costs and average wind power integration costs with respect to different wind power injection a- mounts are calculated and analyzed. Simulation results on IEEE 118-node systems justified the formulation and solving technique pres- ented in this paper. Wind power injection amounts and their costs can provide supports for day-ahead dispatching decision.
出处
《南方电网技术》
北大核心
2016年第1期60-67,共8页
Southern Power System Technology
基金
国家自然科学基金(51407097)~~
关键词
风电接纳
不确定性
非支配分类遗传算法
平均风电接纳成本
发电成本
wind power integration
uncertainties
non-dominated sorting genetic algorithm
average wind power integration cost
generation cost
