论文部分内容阅读
针对机场货运量预测不能满足机场实际运行精度等缺点,提出一种季节性ARIMA和RBF神经网络集成模型预测机场货运量,该模型首先利用季节性ARIMA模型预测机场货运量线性部分,然后用RBF神经网络模型预测机场货运量非线性部分,最后把非线性部分预测结果作为线性部分预测结果的补偿,得到最终预测结果。实验结果表明,新模型可以有效结合季节性ARIMA和RBF神经网络各自的优点;相对单一季节性ARIMA模型和单一RBF神经网络模型预测精度分别提高了6.30%和3.32%,预测精度满足机场实际运行要求。
Aiming at the defects that the airport cargo volume forecast can not meet the actual operation accuracy of the airport, a seasonal ARIMA and RBF neural network integrated model is proposed to forecast the airport cargo volume. The model first uses the seasonal ARIMA model to predict the linear part of the airport cargo volume, The network model predicts the non-linear part of the airport cargo volume. Finally, the non-linear part of the forecast result is used as the compensation of the linear part of the forecast result to get the final forecast result. The experimental results show that the new model can effectively combine the advantages of seasonal ARIMA and RBF neural networks. The prediction accuracy of ARIMA model and RBF neural network model increases by 6.30% and 3.32% respectively, and the prediction accuracy meets the requirements of the actual operation of the airport .