To improve the accuracy of electricity price forecasting, the paper presents a optimal weighted geometric combined forecasting method, which makes use of the information of second-index flatness, adaptive neuron-fuzzy inference system and grey model of modification comprehensively.
针对电价预测的精确度问题,引入了最优加权几何平均组合预测方法,它综合利用了二次指数平滑、自适应模糊神经网络和修正的灰色模型三种方法提供的有用信息,并且该组合预测模型的误差平方和小于各单一预测方法的误差平方和,因此进一步提高了预测结果的准确性。