基于神经网络和支持向量机的电力系统负荷预测方法

An electricity load forecasting method based on neural networks and SVM

提高电力系统负荷预测的精确度是当前负荷预测工作的难点。考虑到神经网络可以逼近任意的非线性关系,而支持向量机能够将约束问题转化,容易地找到全局极小。本文提出了一种基于神经网络和支持向量机的混合负荷预测方法,此方法能通过支持向量机消除了神经网络的总和较小,但单点误差较大的不利现象,而神经网络消除了支持向量机对于模型的简单化问题。最后,负荷预测结果表明本文的方法非常有效。

The improvement of accuracy of the electricity load forecasting was the most difficult point. Neural networks can simulate any non-linear function, while the SVM can transform the limited problems. This paper proposed a method that was based on neural networks and SVM. SVM can dispel the less summation and the larger single point errors of neural networks, and the neural networks can dispel the simplification of the model of the vector support machine. The result indicated that the method was effective.