基于季节性灰色Fourier模型的短时交通流预测

针对短时交通流数据的周期性、非线性和随机性的特点,提出一种基于复化Simpson公式的季节性灰色Fourier模型.在季节性GM(1, 1)模型的基础上,首先利用复化Simpson公式对背景值进行优化,然后用Fourier级数方法修正预测结果.将新模型应用于加拿大Whitemud Drive高速公路的交通流预测,数值计算结果表明:新模型的预测平均绝对值百分比误差为1.54%、拟合度为0.996 0,均优于传统的GM(1, 1)模型、季节性GM(1,1)模型和Fourier优化的季节性GM(1, 1)模型.

Seasonal Grey Fourier Model with Applications to Short-Term Traffic Flow Forecasting

According to the characteristics of the periodic fluctuation, nonlinearity and randomness of short-term traffic flow datum, a seasonal grey Fourier model based on compound Simpson formula is proposed. In the new model, the background value is first reconstructed by using the compound Simpson formula and the Fourier series method is then used to extract the periodic information to improve the residual error of the original seasonal GM(1,1). Finally, the model is applied to traffic flow prediction of Canadian Whitemud Drive expressway. Numerical results show that the mean absolute percentage error and the fitting degree of the new model are 1.54% and 0.996 0, respectively. Both of them are better than those of the traditional GM(1,1) model, the seasonal GM(1,1) model and the seasonal GM(1,1) model with Fourier optimization.