论文部分内容阅读
首次将诱导有序加权平均(IOWA)算子应用到短时交通流预测中。建立了以整体预测误差平方和最小为目标的组合预测模型。在分析短时交通流预测模型的基础上,选取了指数平滑法、季节自回归求和移动平均模型(SARIMA)、BP神经网络模型对短时交通流进行预测;再用IOWA算子将这三种模型进行组合预测。最后进行实例验证。通过MAE、MSE和MAPE三项指标比较分析四种模型的预测效果。结果证明,IOWA算子组合预测模型明显优于其他的预测模型,有效地提高了短时交通流的预测精度。
For the first time, the induced ordered weighted average (IOWA) operator is applied to short-term traffic flow prediction. A combined forecasting model with the goal of minimizing the square sum of the overall forecasting error is established. Based on the analysis of the short-term traffic flow forecasting model, exponential smoothing method, seasonal autoregressive and moving average model (SARIMA) and BP neural network model are used to predict short-term traffic flow. Then IOWA operator The model is combined forecasting. Finally, verify the examples. The predictive effects of four models were compared by MAE, MSE and MAPE. The results show that the IOWA operator combination forecasting model is obviously superior to other forecasting models, which effectively improves the forecasting accuracy of short-term traffic flow.