基于BKF-SVM电力短期负荷预测

Short-term Load Forecasting based on the BKF-SVM

支持向量机方法在负荷预测领域已经得到广泛应用,但它在训练数据时仍存在许多弊端,如数据处理量太大、处理速度慢等,针对这些缺点,本文提出了一种基于布尔核函数的SVM(BKF-SVM)预测系方法,为了确定SVM中直接影响其推广能力的超参数,提出了固定步长迭代法,实现了对超参数的自动选取。实际算例表明,将该系统应用于短期负荷预测中,与RBF-SVM方法作了比较,得到了较高的预测精度,具有结构简单,泛化性能好,不易发生过拟合现象等优点。

The SVM has been successfully applied to the load forecasting area, but it has some disadvantages of very large data amount and slow processing speed. Therefore,a SVM forecasting system based on boolean kernel function （BKF-SVM） was proposed. For confirm the super parameter impacting directly its generalizing capacity,it proposed an abbreviated algorithm by adopting constant iterative step length to optimize the parameter,which implemented automatically choosing.This approach has achieved greater forecasting accuracy comparing with the method of RBF-SVM throug the practical calculating example,and has some merits such as more simple structure,better capability and not likely occur over-fitting phenomenon.