基于支持向量机的网格负载信息预测模型

Grid Load Forecasting Based on Least Squares Support Vector Machine

提出了采用小波分析和最小二乘支持向量机(LS-SVM)混合模型对网格负载信息进行预测。该模型首先基于小波多分辨率分析对非平稳的网格负载样本做序列分解,得到不同尺度下的负载分量,然后利用LS-SVM对不同尺度的分量进行预测,最后通过对各分量预测信息进行重构得到相应的预测值。实验结果表明,使用本模型进行短期负荷预测比传统小波神经网络方法可以获得更好的预测精度。

An algorithm for grid load forecasting based on wavelet analysis and least squares support vector machine, was introduced. Begin with discussion of decomposition of serial signal of grid load and then get the forecasts of each sub- signa/s by LS- SVM. The third step is combination of these forecasts. This method was successfully achieved on forecasting of memory load. The experiment result shows that it can get better forecasting accuracy to traditional wavelet neural network in short - term load forecasting.