基于组合灰色神经网络模型的电力远期价格预测

针对电力远期价格受多种因素影响，变化趋势复杂，难以通过建立准确的数学模型进行预测，提出了采用灰色动态模型对电力远期价格进行预测，并在此基础上构造了组合灰色神经网络预测模型。该模型有效地将灰色理论弱化数据序列波动性的优点和神经网络特有的非线性适应性信息处理能力相融合。研究结果表明，本模型能在小样本、贫信息的条件下对电力远期价格做出比较准确的预测，为电力市场的参与者能更好地利用电力远期合约进行套期保值提供了有效的工具。

Electricity Forward Price Forecasting Based on Combined Gray Neural Network Model

Due to the fluctuation of electricity forward price affected by various factors, it is difficult to establish an accurate math model to describe its movement. Gray dynamic models were used to predict electricity forward price and a combined gray neural network (CGNN) model was proposed on the basis of those models. The fluctuation of data sequence is weakened by the gray theory and the neural network is capable of processing non-linear adaptable information, and the CGNN model is a combination of those advantages. The results reveal that electricity forward price can be accurately predicted through this model by reference to small sample and information. It was concluded that this CGNN model serves as an important role in helping the electricity market participants to hedge through the electricity forward contract more effectively.