文章摘要
叶林,路朋,滕景竹,等.考虑风电功率爬坡事件的功率预测-校正模型[J].电力系统自动化. DOI: 10.7500/AEPS20180321004.
YE Lin,LU Peng,TENG Jingzhu, et al.Power Prediction and Correction Model Considering Wind Power Ramping Events[J].Automation of Electric Power Systems. DOI: 10.7500/AEPS20180321004.
考虑风电功率爬坡事件的功率预测-校正模型
Power Prediction and Correction Model Considering Wind Power Ramping Events
DOI:10.7500/AEPS20180321004
关键词: 风电爬坡事件  最优旋转门算法  模糊极端学习机网络  风电功率预测
KeyWords: Wind power ramp event  Optimized swinging door algorithm  FS-ELM neural network  wind power forecast
上网日期:2019-01-31
基金项目:国家电力公司科技项目,国家自然科学基金
作者单位E-mail
叶林 中国农业大学信息与电气工程学院 YL@cau.edu.cn 
路朋 中国农业大学信息与电气工程学院 lupeng@cau.edu.cn 
滕景竹 国网冀北电力有限公司 yby@cau.edu.cn 
翟丙旭 国网冀北电力有限公司
国网冀北电力有限公司 
yby@cau.edu.cn 
吴林林 国网冀北电力科学研究院 yby@cau.edu.cn 
蓝海波 国网冀北电力公司 yby@cau.edu.cn 
仲悟之 中国电力科学研究院 yby@cau.edu.cn 
摘要:
      随着大规模风电接入电力系统,风电爬坡事件对电网的安全稳定运行带来一定的影响。研究爬坡事件发生时的功率预测已越来越迫切。基于极限学习机理论,提出了一种考虑风电场爬坡事件的超短期功率预测和校正模型。首先,利用最优旋转门算法对当前爬坡事件进行识别,提取爬坡事件特征值,建立模糊C均值聚类模型以得到同类数据,在此基础上,采用极限学习机算法对上述数据进行训练、预测,通过元组向量时间扭曲法在历史风电功率预测爬坡事件库中寻找与当前风电功率预测结果相似的爬坡事件,得到功率预测历史相似爬坡事件。最后,利用功率预测历史匹配值与实际值之间的特征值误差,对风电功率预测结果进行修正。算例表明,所提方法可准确识别风电爬坡事件、有效提高风电功率超短期预测精度。
Abstract:
      With the continuously increasing integration capacity of wind power into power system, ramping events of large scale highly-concentrated wind power generation have brought series of challenges on power system operation. A novel method of multi-time scale power prediction and correction model considering the characteristic of wind power ramping events is employed based on extreme learning machine (ELM). Firstly, the optimized swinging door algorithm (OpSDA) is utilized to identify the wind power ramp events (WPRE). Then, the fuzzy C-means clustering model is established to describe the characteristic of the WPRE as well as to get the congeneric data. ELM is used as a predictor to forecast the same type of data separately. Wind power forecasting result shows a good agreement with historic event database by use of the tuple vector time warp (TVTW) algorithm. Finally, the wind power prediction result is corrected corresponding to the difference between the predicted power history value and the actual value. Case studies show that the proposed approach significantly improves the precision of wind power forecasting in wind power ramping periods.
查看全文   查看/发表评论  下载PDF阅读器