| 引用本文: | 刘子文,曹博源,崔晓丹,周海强,王溢,许剑冰,杨忠光.基于等效电势动态相似性的直驱式风电场分群等值方法[J].电力自动化设备,2024,44(11):79-87. |
| LIU Ziwen,CAO Boyuan,CUI Xiaodan,ZHOU Haiqiang,WANG Yi,XU Jianbing,YANG Zhongguang.Clustering equivalence method for direct-drive wind farm based on dynamic similarity of equivalent potential[J].Electric Power Automation Equipment,2024,44(11):79-87. |
|
| 摘要: |
| 针对大规模直驱式风电场并网系统暂态电压等值建模困难的问题，考虑了直驱风机全功率变流器不同控制环节的作用机理，提出了一种直驱风机的暂态电压等值模型与动态聚类方法。建立计及Chopper电路动作过程的直驱风机暂态模型，构建了适用于暂态电压分析的等效电势指标。该指标能够反映暂态过程中直驱风机并网功率和端口电压的动态过程，并根据该指标的动态变化来反映系统故障下的暂态电压特性。针对风电场内各直驱风机物理位置、行为特征的差异性，提出了一种基于约束动态时间规整的相似度分析方法，构建了能够表征不同机组等效电势指标差异性的相似度矩阵。在此基础上，采用层次凝聚聚类算法实现了不同直驱风机的动态聚合。在PSCAD/EMTDC中建立了直驱式风电场多机等值模型，通过对比不同故障场景下的等值精度验证了所提方法的有效性和优越性。 |
| 关键词: 直驱式风电场 变流器 暂态电压 等效电势 约束动态时间规整 层次凝聚聚类算法 |
| DOI：10.16081/j.epae.202407007 |
| 分类号: |
| 基金项目:国家电网有限公司总部科技项目(5108-202218280A-2-67-XG) |
|
| Clustering equivalence method for direct-drive wind farm based on dynamic similarity of equivalent potential |
|
LIU Ziwen1, CAO Boyuan2, CUI Xiaodan3, ZHOU Haiqiang1, WANG Yi1, XU Jianbing3, YANG Zhongguang2
|
|
1.School of Electrical and Power Engineering, Hohai University, Nanjing 211100, China;2.State Grid Shanghai Electric Power Company, Shanghai 200437, China;3.NARI Group Corporation(State Grid Electric Power Research Institute),Nanjing 211106, China
|
| Abstract: |
| Aiming at the problem of difficulty in transient voltage equivalence modeling for the system connected with large-scale direct-drive wind farms, considering the operational mechanism of different control links for the direct-drive wind turbine based on full-power converter, a transient voltage equivalence model and a dynamic aggregation method for direct-drive wind turbine are proposed. Taking into account the operational process of Chopper circuit, a transient model for direct-drive turbine is formulated. The equivalent potential indicator suitable for transient voltage analysis is established, which can reflect the dynamic process of grid-connected power and terminal voltage during transient process, and the transient voltage characteristic under system faults is analyzed according to the variations of the indicator. In view of the differences of physical location and behavior characteristics of each direct-drive wind turbine in wind farm, a similarity analysis method based on canonical dynamic time warping is proposed, and a similarity matrix is constructed to characterize the difference of equivalent potential indexes of different units. Based on this, the hierarchical agglomerative clustering algorithm is adopted to realize the dynamic aggregation of different direct-drive wind turbines. A multi-turbine equivalence model of direct-drive wind farm is established in PSCAD/EMTDC, and the effectiveness and superiority of the proposed method are verified by comparing the equivalence accuracy under different fault scenarios. |
| Key words: direct-drive wind farm electric converters transient voltage equivalent potential canonical dynamic time warping hierarchical agglomerative clustering algorithm |
