欢迎访问《太阳能学报》官方网站,今天是 分享到:
ISSN 0254-0096 CN 11-2082/K

太阳能学报 ›› 2022, Vol. 43 ›› Issue (8): 391-398.DOI: 10.19912/j.0254-0096.tynxb.2021-0054

• • 上一篇    下一篇

基于风电机组状态的超短期海上风电功率预测

黄玲玲, 李锁, 符杨, 王振帅   

  1. 上海电力大学电气工程学院,上海 200090
  • 收稿日期:2021-01-13 出版日期:2022-08-28 发布日期:2023-02-28
  • 通讯作者: 黄玲玲(1982—),女,博士、教授,主要从事海上风电规划与并网方面的研究。linglinghuang82@126.com
  • 基金资助:
    国家电网公司总部科技项目(52090R180006)

ULTRA-SHORT TERM OFFSHORE WIND POWER PREDICTION BASED ON CONDITION-ASSESSMENT OFWIND TURBINES

Huang Lingling, Li Suo, Fu Yang, Wang Zhenshuai   

  1. School of Electrical Engineering, Shanghai University of Electric Power, Shanghai 200090, China
  • Received:2021-01-13 Online:2022-08-28 Published:2023-02-28

摘要: 提出一种基于风电机组状态的超短期海上风电功率预测模型。首先,综合考虑海上环境因素以及风电机组部件间的相互作用建立指标的预测模型,以长短期记忆神经网络的预测误差作为监测指标的动态劣化度;然后采用模糊综合评价法对风电机组的运行状态进行评估,依据评估结果对风电机组历史运行数据进行划分;最后根据分类后历史运行数据建立基于机组状态的超短期风电功率预测模型。结合国内某海上风电场实例数据进行分析,算例结果表明所提方法可有效提高风电功率预测精度。

关键词: 海上风电机组, 长短期记忆神经网络, 风电功率, 预测模型, 状态评估

Abstract: In this paper, an ultra-short term offshore wind power prediction model based on condition-assessment of wind turbines is proposed. Firstly, considering the influences of marine environmental factors and complex correlations between the various components of the wind turbines, an index prediction model based on the dynamic deterioration is proposed. The dynamic deterioration obtained by the relative errors of Long Short-term Memory (LSTM) . Then, the fuzzy comprehensive evaluation method is introduced to evaluate the operation conditions of wind turbines, according to which the historical operation data of the wind turbine are then categorized . Finally,an ultra-short term power prediction model is established. A case with the data from a domestic offshore wind farm is disscussed to illustrate the effectiveness and the prediction accuracy of the proposed methodology.

Key words: offshore wind turbines, long short-term memory, wind power, prediction model, condition assessment

中图分类号: