风光互补发电系统的分级递阶分布式预测控制  

Hierarchical distributed model predictive control of hybrid wind and solar generation system

作　　者：孔小兵[1] 刘向杰[1] 韩梅 Xiaobing KONG;Xiangjie LIU;Mei HAN(State Key Laboratory of Alternate Electrical Power System With Renewable Energy Sources,North China Electric Power University,Beijing 102206,China;Beijing Goldwind Science ＆ Creation Windpower Equipment Co.,Ltd,Beijing 100176,China)

机构地区：[1]华北电力大学新能源电力系统国家重点实验室,北京102206 [2]北京金风科创风电设备有限公司,北京100176

出　　处：《中国科学：信息科学》2018年第10期1316-1332,共17页Scientia Sinica(Informationis)

基　　金：国家自然科学基金(批准号:61603134;61673171;61533013);中央高校基本科研业务费专项基金(批准号:2017ZZD004;201-7MS033)资助项目

摘　　要：现代电力系统是可再生能源、清洁能源与信息互联网高度融合的分布式能源网络系统,这其中,风力发电、太阳能发电系统具有间歇性且受到较多外部条件约束.协同优化风力发电、太阳能发电控制可以在满足负荷需求的同时,减少机械损耗,延长机组使用寿命,保证电网的安全经济运行.针对大规模且地理分散的风光互补发电系统,本文提出一种分级递阶分布式预测控制策略.上层优化控制器采用迭代分布式预测控制策略,不仅实现功率优化分配,而且能够实现减少风机低速轴机械损耗、降低系统发电成本的经济性目标.下层控制器采用监督预测控制算法可同时保证子系统的跟踪性能和经济性能.所构造的分级递阶分布式预测控制策略通过各子系统间的协同优化,实现了分布式能源的"即插即用".仿真实验证明,本文提出的分级递阶分布式预测控制策略能够有效地实现微网的安全可靠、优质高效、灵活互动等要求.Distributed networked generation （DNG）; which incorporates renewable energy and clean energy into information networks; has become increasingly important for modern power systems. In DNG; wind power and solar power generation are considered as intermittent systems with multiple constraints. The coordinated optimization of the wind power and solar power generation can effectively meet the load demand while guaranteeing the safety of power grids by reducing the wear and tear of generating units and prolonging the lifetime of grids. This study aims to construct a hierarchical distributed model predictive control （HDMPC） for large-scale; geographically dispersed wind-solar power generation systems. In HDMPC; the upper layer utilizes an iterative distributed control strategy to coordinate the power dispatch. Thus; it attains economic objectives; including the reduction in the torsional shaft torque transmitted to the gearbox in the wind turbine system and the generation cost. The lower layer utilizes the supervisory predictive control to realize economic and tracking properties. HDMPC enables tile plug-and-play of distributed energy through the coordinated optimization of subsystems. Simulation experiments validated the advantages of the proposed method; which can meet the demands of safe; reliable; highly efficient; flexible; and interactive microgrid controlling.

关 键 词：风光互补发电系统 分级递阶控制 分布式预测控制 风力发电 太阳能发电 

分 类 号：TM61[电气工程—电力系统及自动化]