基于季节波动序列的灰色预测模型及其应用

Grey Forecasting Model Based on Seasonal Fluctuation Sequence and Its Application

文章针对复杂系统行为序列中的季节性波动特征,提出了基于季节因子及傅里叶优化的灰色季节预测模型。首先,该模型提出包含年度作用系数的季节因子,其通过年度作用系数的改变,可转化为均值季节因子、新信息季节因子和关联季节因子;其次,考虑行为序列受时间变化作用的影响,在预测模型中加入线性修正项,以提高模型预测精度,并利用傅里叶级数来拟合模型预测残差序列中的周期波动特征。最后,将模型用于郑州站点降水的模拟与预测,研究结果证明了构建的模型具有更高的预测精度。

Aiming at the seasonal fluctuation characteristics of the complex system behavior sequence,this paper proposes a grey seasonal forecasting model based on seasonal factors and Fourier optimization.Firstly,the model proposes seasonal factors including annual effect coefficient,which can be transformed into mean seasonal factor,new information seasonal factor and correlation seasonal factor through the change of annual effect coefficient.And then,considering the influence of time variation on behavior sequence,the paper adds linear correction terms to the prediction model to improve the prediction accuracy of the model,and also uses the Fourier series to fit the model to predict the periodic fluctuation characteristics in the residual series.Finally,the model is applied to the simulation and prediction of precipitation in Zhengzhou,and the research results prove that the constructed model has higher prediction accuracy.