基于GRNN神经网络模型的交通运输能力预测研究

The Research on Traffic Capacity Prediction Based on GRNN Neural Network Model

为得到交通运输能力的合理预测结果,文章在2010~2018年历史统计数据的基础上,通过建立GRNN广义回归神经网络模型,对交通运输能力进行合理预测,并将模型的预测结果与实际结果进行对比,分析误差率。结果表明:基于GRNN广义回归神经网络模型能作为一种新方法预测和研究运输能力,模型中光滑因子的设定大小与模型的逼近性能误差成正相关,与模型的预测性能误差成负相关。

In order to get a reasonable forecast result of transportation capacity,the paper establishes GRNN generalized regression neural network model which based on the historical statistical data of 2010-2018 to make a reasonable forecast of transportation capacity.The prediction results of the model are compared with the actual results,and the error rate is analyzed.The results show that the generalized regression neural network model based on GRNN can provide a new method for the prediction and analysis of transport capacity,and the set magnitude of the smoothness factor in the model is positively correlated with the approximation performance error of the model,and negatively correlated with the prediction performance error of the model.