基于小波自适应支持向量机的短期电价预测

电价预测对于发电商、供电企业以及市场监管者都具有重要的意义。提出一种小波自适应支持向量机预测模型，先将电价时间序列作小波分解得到低频和高频分量，再采用自适应调整法，自动地为支持向量机选择较好的参数对电价小波分量逐一预测，最后通过小波重构得到电价最终预测结果。实例证明前述方法得到的预测精度高于BP、RBF、SVM等传统预测模型。

The Short-term Electricity Price Forecasting Based on Wavelet Adaptive Support Vector Machine

Electricity price forecasting is of important significance for electricity generators, power supply enterprises and market regulator. A wavelet adaptive support vector machine forecasting model is proposed in the paper. Firstly, the electricity price time series are decomposed to low frequency and high frequency wavelet components, then the adaptive adjustment method are adopted to select pa-rameters automatically for support vector machine to forecast electricity price wavelet components one by one, and the final prediction re-sult is achieved by wavelet reconstruction. The example proves that the proposed method behaves a higher prediction accuracy than tradi-tional forecasting models such as BP network.