摘要
核覆盖算法融合了SVM中的核函数法与构造性学习的覆盖算法中的优点,将该方法用于电力系统负荷的预测,先将负荷数据映射到一个核空间,然后在核空间中利用一般的覆盖算法进行预测。实际数据验证表明:该方法与单一的SVM预测相比具有预测精度高、支持向量少和计算量小等优点。
Kernel covering algorithm fused advantages of SVM and constructive machine learning method. In this paper the method is used to predict the short - term Load, first, mapping the load data to a nuclear space, and then use coverage algorithm to predict in nuclear space, the result shows that comparing to single SVMs models the proposed algorithm have three advantages,i, e. more accurate forecasting, fewer support vector machines and smaller calculation complexity.
出处
《农机化研究》
北大核心
2008年第9期206-207,211,共3页
Journal of Agricultural Mechanization Research
关键词
覆盖算法
核函数
负荷预测
covering algorithm
kernel function
prediction of short - term