论文部分内容阅读
本文分析铁路客运量影响因素,利用主成分分析(PCA)消除原始铁路客运量影响因素之间的相关性,将主成分分析结果作为BP神经网络的输入,并通过增加动量项、输入数据处理、调整学习速率优化BP神经网络,提出基于PCA-BP神经网络的铁路客运量预测模型。实例研究表明,与BP神经网络相比,PCABP神经网络能有效提高铁路客运量预测精度。
This paper analyzes the influencing factors of railway passenger volume, uses the principal component analysis (PCA) to eliminate the correlation between the factors affecting the original railway passenger volume, and uses the result of principal component analysis as the input of BP neural network. By adding the momentum term, input data processing, Adjust the learning rate to optimize BP neural network, and propose the prediction model of railway passenger volume based on PCA-BP neural network. The case study shows that compared with BP neural network, PCABP neural network can effectively improve the prediction accuracy of railway passenger volume.