中国电力 ›› 2021, Vol. 54 ›› Issue (12): 102-111.DOI: 10.11930/j.issn.1004-9649.202012112

• 发电 • 上一篇    下一篇

基于概率可靠性评估的永磁直驱风机低电压穿越控制模型参数辨识

乔腾1, 张益铭1, 曹一家1, 王力1,2, 袁清1   

  1. 1. 智能电网运行与控制湖南省重点实验室, 长沙理工大学 电气与信息工程学院, 湖南 长沙 410114;
    2. 可再生能源电力技术湖南省重点实验室, 湖南 长沙 410114
  • 收稿日期:2020-12-24 修回日期:2021-03-19 出版日期:2021-12-05 发布日期:2021-12-16
  • 作者简介:乔腾(1995-),男,硕士研究生,从事新能源并网控制技术研究,E-mail:orgqiao@163.com;王力(1990-),男,通信作者,博士,从事电力系统运行与控制研究,E-mail:wangli@csust.edu.cn
  • 基金资助:
    西藏自治区重大科技专项资助项目(高寒高海拔乡村高适应性独立风光柴耦合供能系统研发与示范,XZ201901-GA-09);湖南省研究生科研创新项目(新能源并网LVRT源网协调机制研究,CX20190684)。

Parameter Identification of Low Voltage Ride-Through Control Model for Permanent Magnet Direct-Drive Wind Turbine Based on Probabilistic Reliability Assessment

QIAO Teng1, ZHANG Yiming1, CAO Yijia1, WANG Li1,2, YUAN Qing1   

  1. 1. Hunan Province Key Laboratory of Smart Grids Operation and Control, School of Electrical and Information Engineering, Changsha University of Science & Technology, Changsha 410114, China;
    2. Hunan Provincial Key Laboratory of Renewable Energy Electric-Technology, Changsha 410114, China
  • Received:2020-12-24 Revised:2021-03-19 Online:2021-12-05 Published:2021-12-16
  • Supported by:
    This work is supported by the Major Science and Technology Project of the Tibet Autonomous Region (R&D and Demonstration of High Adaptive Independent Wind, Photovoltaic, and Diesel Coupled Energy Supply Systems in High-Altitude and Frigid Rural Areas, No.XZ201901-GA-09); Hunan Provincial Innovation Foundation for Postgraduate (Research on the Mechanism of LVRT Source-Network Coordination of New Energy Grid-Connection, No.CX20190684).

摘要: 参数的准确获取有助于提高永磁直驱风机低电压穿越控制模型的仿真精度。复杂模型不仅存在参数多、无法全部测试获取的问题,非目标参数的不精确状况也会影响到目标参数辨识的准确度。通过灵敏度分析判断参数获取的难易程度,作为选择辨识顺序的依据。将非目标参数随机设置在经验范围内,采用概率可靠性评估方法选取用于准确辨识目标参数的观测变量。针对目标参数,采取兼顾辨识优先顺序和对应观测变量选取的分步辨识策略,获取高概率可靠性的辨识结果作为最终参数值,消除非目标参数设置不准确的影响。基于所提方法,对永磁直驱风机低电压穿越控制模型中目标参数进行理论辨识,并通过实测数据进行验证。

关键词: 永磁直驱风机, 参数辨识, 灵敏度分析, 概率可靠性评估, 低电压穿越控制模型

Abstract: The accurate parameter acquisition is helpful to improve the simulation accuracy of the low voltage ride-through control model for permanent magnet direct-drive wind turbines. The parameters of a complex model are numerous and cannot be fully obtained through tests, and the inaccuracy of its non-target parameters will affect the accuracy of target parameter identification. Therefore, the sensitivity analysis is used to evaluate the difficulty of parameter acquisition, providing a basis for the identification order selection. The non-target parameters are randomly set within the empirical range, and the probabilistic reliability assessment method is employed to decide observed variables from them for the accurate identification of target parameters. A stepwise identification strategy that takes into account both the priority order for identification and the selection of corresponding observed variables is adopted for the target parameters, and the identification results with high probabilistic reliability are obtained as the final parameter values. In this way, the influence of inaccurate settings of non-target parameters is eliminated. The proposed method is applied to identify the target parameters of the low voltage ride-through control model for permanent magnet direct-drive wind turbines, and the results have been verified by measured data.

Key words: permanent magnet direct-drive wind turbine, parameter identification, sensitivity analysis, probabilistic reliability assessment, low voltage ride-through control model