地铁进站客流预测：基于GM-RBF组合模型的高效方案

**作者：**杨婷婷、张亚男、张涛、张晓明

**摘要：**地铁进站客流预测是地铁运营管理的重要组成部分，对于提高地铁运营效率、优化运营方案具有重要意义。本文提出了一种基于GM-RBF组合模型的地铁进站短时客流预测方法。该方法首先利用灰色理论对原始数据进行预处理，然后采用GM(1,1)模型对预处理后的数据进行建模，得到GM模型的预测结果。接着，利用RBF神经网络对GM模型的预测结果进行修正，得到RBF模型的预测结果。最后，将GM模型和RBF模型的预测结果进行组合，得到最终的预测结果。本文以北京地铁13号线为例进行了实证分析，结果表明，所提出的方法在地铁进站客流预测方面具有较高的准确性和可靠性。

**关键词：**地铁进站客流预测；GM-RBF组合模型；灰色理论；GM(1,1)模型；RBF神经网络

Abstract: Subway entrance flow prediction is an important part of subway operation management, which is of great significance for improving the efficiency of subway operation and optimizing operation plans. In this paper, a short-term subway entrance flow prediction method based on GM-RBF combination model is proposed. Firstly, the original data is preprocessed by using grey theory, and then the GM (1,1) model is used to model the preprocessed data to obtain the prediction results of the GM model. Then, the RBF neural network is used to modify the prediction results of the GM model to obtain the prediction results of the RBF model. Finally, the prediction results of the GM model and the RBF model are combined to obtain the final prediction results. Taking Beijing Subway Line 13 as an example, this paper conducts empirical analysis. The results show that the proposed method has high accuracy and reliability in subway entrance flow prediction.

Keywords: Subway entrance flow prediction; GM-RBF combination model; Grey theory; GM (1,1) model; RBF neural network