| 引用本文: | 潘学萍,戚相威,梁伟,雍成立,丁新虎,李威,朱玲.综合模型聚合和参数辨识的风电场多机等值及参数整体辨识[J].电力自动化设备,2022,42(1): |
| PAN Xueping,QI Xiangwei,LIANG Wei,YONG Chengli,DING Xinhu,LI Wei,ZHU Ling.Multi-machine equivalence and global identification of wind farms by combining model aggregation and parameter estimation[J].Electric Power Automation Equipment,2022,42(1): |
|
| 摘要: |
| 参数的准确获取是风电场多机等值时的难题之一。提出了综合解析方法和辨识方法的风电场动态等值建模框架。基于解析方法获得各等值机参数的估计值，将其作为初始值，进一步基于实际受扰轨迹进行参数辨识；创新性地提出了综合稳态特性与动态特性的风电场分群方法，并将动态时间规整方法引入受扰轨迹的相似度分析；研究了多台等值机参数的可辨识性，针对多机等值时存在参数多、部分参数无法区分辨识的难题，提出了分类辨识和重点辨识相结合的参数整体辨识策略。最后，基于粒子群优化算法进行参数辨识，并对参数辨识误差进行了分析。 |
| 关键词: 风电场 分群 多机等值 参数辨识 可辨识性 粒子群优化算法 |
| DOI：10.16081/j.epae.202109004 |
| 分类号: |
| 基金项目:国家自然科学基金资助项目(52077061) |
|
| Multi-machine equivalence and global identification of wind farms by combining model aggregation and parameter estimation |
|
PAN Xueping1, QI Xiangwei1, LIANG Wei1, YONG Chengli1, DING Xinhu1, LI Wei2, ZHU Ling2
|
|
1.College of Energy and Electrical Engineering, Hohai University, Nanjing 211100, China;2.NARI Group Corporation/State Grid Electric Power Research Institute, Nanjing 211106, China
|
| Abstract: |
| Obtaining parameter accurately is a difficult task for multi-machine equivalence of wind farm. A dynamic equivalent modeling framework of wind farms by combing analytical method and identification method is proposed. The estimated values of parameters for equivalent WTGs(Wind Turbine Generators) are attained based on analytical method, and these results are used as initial values in parameter estimation by fitting the real disturbed trajectories. Then, the grouping method by combining steady-state and dynamic characteristics is proposed innovatively, and DTW(Dynamic Time Warping)-based method is introduced in the similarity analysis of WTGs’ disturbed trajectories. The identifiability of multiple equivalent WTGs’ parameters is studied. Since there are too many parameters and some parameters cannot be identified simultaneously in multi-machine equivalence, a global parameter estimation strategy for multi-WTG is proposed by combing classification identification and key identification. Finally, the PSO(Particle Swarm Optimization) algorithm is used in parameter estimation, and the identification accuracy is analyzed. |
| Key words: wind farms grouping multi-machine equivalence parameter estimation identifiability PSO algorithm |
