基于两步处理的大型海上风电场分布式运行功率优化方法  

Distributed Optimization Method for Operation Power of Large-scale Offshore Wind Farm Based on Two-step Processing

作　　者：杨建[1] 黄朝能 宋冬然[1] 董密[1] 陈思范 胡阳[3] 房方[3] YANG Jian;HUANG Chaoneng;SONG Dongran;DONG Mi;CHEN Sifan;HU Yang;FANG Fang(School of Automation,Central South University,Changsha 410083,China;Mingyang Smart Energy Co.,Ltd.,Zhongshan 528437,China;School of Control and Computer Engineering,North China Electric Power University,Beijing 102206,China)

机构地区：[1]中南大学自动化学院,湖南省长沙市410083 [2]明阳智慧能源集团股份公司,广东省中山市528437 [3]华北电力大学控制与计算机工程学院,北京市102206

出　　处：《电力系统自动化》2023年第7期94-104,共11页Automation of Electric Power Systems

基　　金：国家自然科学基金资助项目(62211540397);湖南省自然科学基金资助项目(2021JJ30875);华能集团总部科技项目(海上风电与智慧能源系统科技专项(一期),HNKJ20-H88-06)。

摘　　要：随着海上风电场逐步走向大型化,尾流效应对风电场发电效率的影响加剧。已有风电场尾流控制方法的计算负荷较大,优化效果有限。对此,基于风电场网络的更新与分簇,提出一种两步功率优化方法。基于海上风电场尾流效应完善网络化定义,提出将优化分为两个阶段以处理优化动作对风电场网络和功率的影响;引入谱聚类方法实现网络分簇,结合改进的平衡优化算法进行高效求解,并通过分簇更新与优化迭代的结合实施两步优化策略。案例仿真结果表明,相对于集中式优化,所提出的方法优化效果接近,求解时间显著缩短;与单步优化相比,在具有复杂尾流关联的场景下,能以增加一定的求解时间为代价,获得更大的功率增益。最后,对不同规模风电场和分簇数量进行了探究,所提出方法对大型海上风电场具有较好的应用前景。As offshore wind farms gradually become larger,the wake effect has intensified the impact of wind farm power generation efficiency.Existing wake control methods for wind farms have large computational load and limited optimization effect.In this regard,according to the update and clustering of wind farm networks,a two-step power optimization method is proposed.Based on the improved network definition considering the wake effect of offshore wind farms,the optimization is divided into two stages to deal with the impact of optimization actions on the network and power of wind farms.The spectral clustering method is applied to achieve network clustering.Combined with improved equilibrium optimizer to achieve efficient solution,a two-step optimization strategy is implemented by the combination of cluster update and optimization iterations.The results of case simulation show that,compared with centralized optimization,the proposed method has significant optimization effect and significantly shortens solving time.Compared with direct optimization,larger power gain is obtained at the cost of increasing the limited solving time in complex wake correlation scenarios.Finally,the wind farms of different scales and the number of clusters are explored,and the proposed method has good application prospects for large-scale offshore wind farm.

关 键 词：海上风电场 分布式优化 尾流效应 网络分簇 

分 类 号：TM614[电气工程—电力系统及自动化]