摘要
支持向量机(SVM)已经成功地应用于解决非线性回归和时间序列问题,并且已经开始用于中长期负荷预测。提出了一种基于鲁棒支持向量回归机RSVR(Robust Support Vector Regression)的中长期负荷预测的新方法。给出利用粒子群优化算法对鲁棒支持向量机系数优化选择的方法。建立基于此原理的中长期负荷预测模型,算例分析比较验证本文方法具有预测精度高、计算量小等特点和优势。
Support Vector Machine （SVM） has been successfully applied to solve non-linear regression and time series problem, and has already been used for the medium and long-term load forecasting. This paper brings forward a new load forecasting method based on Robust Support Vector Regression （RSVR）. It gives the use of particle swarm optimization for the robust support vector machines coefficient optimization and constructs the forecasting model.A case is presented to verify the method possessing high precision, small amount of computation and so on.
出处
《电力系统保护与控制》
EI
CSCD
北大核心
2009年第21期77-81,共5页
Power System Protection and Control
关键词
中长期负荷预测
鲁棒性
支持向量机
回归估计
粒子群优化算法
medium and long-term load forecasting
robust
support vector machine （SVM）
regression
particle swarm optimization
