摘要
从最小二乘支持向量机(LS-SVM)的原理出发,从本质上阐明了LS-SVM在短期风速预测中的可行性与优越性。在对LS-SVM在应用中存在的包括数据预处理、核函数构造及选取以及参数优化等问题进行分析后,归纳了现行主要解决方法,从而全面总结了LS-SVM在短期风速预测中的应用概况。最后对基于LS-SVM的短期风速预测所存在的问题进行总结,并提出建议。
Based on the principle of LS-SVM, the feasibility and superiority of the LS-SVM method of Short term wind speed forecasting are essentially clarified. Some problems about the application of LS-SVM, including data pre-processing, the constructing and current solutions are provided respectively. For a series of LS-SVM-based improvements and some combination forecasting methods consisting of LS-SVM with other algorithms, a comprehensive summary is given, from the perspective of the mechanism about LS-SVM algorithm being applied to Short term wind speed forecasting, and the elevation of prediction accuracy and speed. Finally, some key issues about LS-SVM-based Short term wind speed forecasting are summarized and some recommendations are given.
出处
《电气技术》
2013年第6期22-25,共4页
Electrical Engineering
关键词
最小二乘支持向量机
短期风速预测
数据预处理
核函数
参数优化
组合预测
least squares support vector machine
Short term wind speed forecasting
datapre-processing
kernel function
parameter optimization
combination forecasting