基于在线序列极限学习机的风电场短期风速预测研究  

Short-Term Wind Speed Forecasting of Wind Field Based on Online Sequential Extreme Learning Machine

作　　者：覃永杰 QIN Yongjie(Longtan Hydropower Plant of Longtan Hydropower Development Co.,Ltd.Tian′e,547300 China)

机构地区：[1]龙滩水电开发有限公司龙滩水力发电厂,广西天峨547300

出　　处：《红水河》2023年第3期60-65,74,共7页Hongshui River

摘　　要：为了对风电场的风速进行比较准确的预测,提高风电的稳定性,减轻风电对整个电网的影响,针对风速时间序列的混沌特性,笔者运用相空间重构理论对风速时间序列数据进行相空间重构,提出一种运用在线序列极限学习机算法的风速预测理论。通过与BP神经网络算法相比较,在线序列极限学习机算法的预测精度和预测时间都有一定的提高,说明该算法在短期风速预测上是有效的和可行的。In order to predict the wind speed of wind farm accurately improve the stability of wind power and reduce the influence of wind power on the whole power grid aiming at the chaotic characteristics of wind speed time series the phase space reconstruction theory is used to reconstruct the phase space of wind speed time series data and a wind speed forecasting theory using online sequential extreme learning machine algorithm is proposed.Compared with the BP neural network algorithm the prediction accuracy and prediction time of the online sequential extreme learning machine algorithm are improved to a certain extent which shows that the algorithm is effective and feasible in short-term wind speed forecasting.

关 键 词：短期风速预测 风电场 在线序列极限学习机 相空间重构 

分 类 号：TM614[电气工程—电力系统及自动化] TP181[自动化与计算机技术—控制理论与控制工程]