灰色GM(1,1)和神经网络组合的能源预测模型

Combined energy forecasting model of gray GM(1,1) and neural networks

能源消费预测是制定能源规划的重要组成部分。鉴于能源消费系统复杂性和非线性的特征,文章结合某省能源消费的历史数据,首先用灰色预测和神经网络建立了单项预测模型,并对单项预测模型的优缺点进行了分析,然后采用最优组合权重的方法进行优化组合,从而获得更为精确的预测模型和预测值。实例的预测结果表明该模型可以作为能源消费预测的有效工具。

The energy consumption forecast is an important part for establish energy programming?In view of the complexity and non-linearity of energy system,the respective model of gray forecast and neural networks are established firstly,then analyze and compare their advantages and disadvantages?Lastly,the method of optimized combination is applied to allocate the weights for the accurate forecast model and forecast value?The results of Shaanxi show that it is a effective tool for energy consumption forecast.