基于风特征分析的风电机组异常数据识别算法
发布时间：2018-09-17 17:46
【摘要】：对风电的研究往往要依托于历史功率数据,而风电机组采集到的历史数据中往往含有大量的异常数据,这严重影响了对风电功率规律特性的分析。针对风电机组的实测功率数据进行研究,分析风速升降特征与风向特征对风电机组输出功率的影响。将不同的风特征的数据分开讨论,分别利用Copula函数得到概率功率曲线,结合异常数据的时序特征归纳出三类异常数据,建立异常数据识别模型。利用风电机组的实际数据和人工生成数据进行仿真分析,结果表明,该方法能够高效地识别各类异常数据,对风电研究有着重要的意义。
[Abstract]:The research of wind power often relies on historical power data, and the historical data collected by wind turbines often contain a large number of abnormal data, which seriously affects the analysis of wind power characteristics. The influence of different wind characteristics on power is discussed separately. The probabilistic power curves are obtained by Copula function, and three kinds of abnormal data are summed up according to the time series characteristics of abnormal data. It is of great significance for wind power research to identify all kinds of abnormal data efficiently.
【作者单位】： 东北电力大学;
【基金】：国家重点研发计划项目课题(2016YFB0900101)~~
【分类号】：TM315
本文编号：2246658
[Abstract]:The research of wind power often relies on historical power data, and the historical data collected by wind turbines often contain a large number of abnormal data, which seriously affects the analysis of wind power characteristics. The influence of different wind characteristics on power is discussed separately. The probabilistic power curves are obtained by Copula function, and three kinds of abnormal data are summed up according to the time series characteristics of abnormal data. It is of great significance for wind power research to identify all kinds of abnormal data efficiently.
【作者单位】： 东北电力大学;
【基金】：国家重点研发计划项目课题(2016YFB0900101)~~
【分类号】：TM315
【相似文献】
相关期刊论文 前3条
1 陈亚红,穆钢,段方丽;短期电力负荷预报中几种异常数据的处理[J];东北电力学院学报;2002年02期
2 王雷;张瑞青;盛伟;徐治皋;;基于支持向量机的回归预测和异常数据检测[J];中国电机工程学报;2009年08期
3 蔡建壮,白同朔,侯志俭;电力系统偶然事故选择中异常数据的鉴别与处理[J];中国电机工程学报;2002年08期
相关会议论文 前1条
1 孙舒;刘莉;张瑛;;基于典型负荷模式的异常数据处理[A];中国高等学校电力系统及其自动化专业第二十四届学术年会论文集（中册）[C];2008年
，本文编号：2246658
本文链接：https://www.wllwen.com/kejilunwen/dianlidianqilunwen/2246658.html
