采用正则化极限学习机的短期风速预测

发布时间：2018-04-27 09:44

本文选题：ACF + 风速　；参考：《电网与清洁能源》2016年11期

【摘要】：高效、准确的风速预测是风电场功率预测的基础,对风力发电控制和风电场并网运行等具有重要意义。针对风速时间序列具有强烈的非线性和波动性,且难以精准预测的特点,提出一种基于正则化极限学习机(regularized extreme learning machine,RELM)的风电场短期风速预测新方法。首先,采用自相关函数(ACF)对风速时间序列的相关性进行分析,得到预测模型输入属性集合;其次,确定预测网络的输入、输出等参数,并建立RELM模型;再次,利用训练集在训练过程中确定网络参数,构建RELM预测模型;最后,以RELM预测模型开展短期风速预测,得出预测结果。采用美国风能技术中心的实测风电场风速数据开展实验证明,相对于标准的ELM和BP神经网络,新方法具有更好的预测精度。
[Abstract]:Efficient and accurate wind speed prediction is the basis of wind farm power prediction, which is of great significance to wind power generation control and grid operation of wind farm. In view of the strong nonlinearity and volatility of wind speed time series and the difficulty of accurate prediction, a new method of wind farm short-term wind speed prediction based on regularized extreme learning machine (RELM) is proposed. First, the correlation of wind speed time series is analyzed by the autocorrelation function, and the input attribute set of prediction model is obtained. Secondly, the input and output parameters of prediction network are determined, and the RELM model is established. The training set is used to determine the network parameters during the training process, and the RELM forecasting model is constructed. Finally, the short-term wind speed prediction is carried out by using the RELM forecasting model, and the prediction results are obtained. The wind speed data of wind farm measured by the Wind Energy Technology Center of the United States show that the new method has better prediction accuracy than the standard ELM and BP neural networks.
【作者单位】：东北电力大学电气工程学院;国网杭州供电公司;
【基金】：国家高技术研究发展计划(863计划)项目(SS2014AA052502) 吉林省科技发展计划项目(20160411003XH) 吉林省社科基金(2015A2) 吉林市科技发展计划项目(20156407)~~
【分类号】：TM614

【相似文献】