基于小波变换和支持向量机的电力系统短期负荷预测

提出一种改进的基于离散小波变换和支持向量机的短期负荷预测方法.运用离散小波变换将负荷时间序列分解为高低频子序列,采用目前较为成熟的支持向量机方法,选择适当的参数对每个序列进行滚动式的单支预测,最后将各分支预测结果相加最终实现负荷预测.实例中负荷数据采用四川省某地区的历史负荷,对该地区的日96点负荷进行全年预测,并将该算法与支持向量机算法进行比较,结果表明,该算法具有较高预测精确性.

Short-term load forecasting based on wavelet transform and support vector machines

This paper presents an improved technique in short-term load forecasting based on discrete wavelet transform (DWT) and support vector machines (SVM). The DWT splits up load time series into low and high frequency components which are to be the features for the SVMs. The SVMs which is improved forecast each components by adopting the appropriate parameters, at the end, all forecasted components are summed up to produce a final forecasted load. The data from an area in Sichuan province is used to verify ninety-six points load forecasting,the performance of algorithm is compared with that of the SVM without DWT,the experimental results show that the proposed algorithm can improve the calculation accuracy.