位置:成果数据库 > 期刊 > 期刊详情页
基于时间序列与支持向量机的风电场风速预测研究
  • ISSN号:1008-4835
  • 期刊名称:陕西电力
  • 时间:0
  • 页码:1-4
  • 语言:中文
  • 分类:TM614[电气工程—电力系统及自动化]
  • 作者机构:[1]新疆大学电气工程学院,新疆乌鲁木齐830008
  • 相关基金:本文受国家自然科学基金项目(50767003)、新疆自治区教育厅科研资助项目(XJEDU2007105)资助.
  • 相关项目:风力发电机系统智能故障诊断技术研究
中文摘要:

介绍时间序列法与支持向量机用于风速预测的理论基础,通过Matlab软件,利用风电场采集得到的风速数据,建立时间序列法与支持向量机模型,对这2种方法在风电场风速预测中的应用进行了研究和比较。仿真结果表明,这两种方法都有效,但支持向量机风速预测精度更高,预测结果更好,具有一定的实用价值。

英文摘要:

Two methods for predicting wind speed based on support vector machines and time series were proposed in this paper. Through Matlab software,using the wind speed data obtained from wind farm,the models were built respectively for every method. The comparison of the time series method and support vector machines in wind speed prediction application was discussed.The result shows that the validity of the proposed methods are verified by the results of the simulation but support vector machines is better than time series method under most circumstances by the results of the simulation.Support vector machines possess higher accuracy and it is of a certain practical value.The research provides reference for later wind speed prediction.

同期刊论文项目
同项目期刊论文