基于改进LSTM的城市轨道短时流量预测研究  

Research on Short-term Flow Prediction of Urban Rail Based on Improved LSTM

作　　者：魏化永[1] 李建华[2] WEI Huayong;LI Jianhua(Department of Urban Rail Transit and Information Engineering,Anhui Communications Vocational&Technical College,Hefei 230051,China;School of Computer Science and Information Engineering,Hefei University of Technology,Hefei 230022,China)

机构地区：[1]安徽交通职业技术学院城市轨道交通与信息工程系,安徽合肥230051 [2]合肥工业大学计算机与信息学院,安徽合肥230022

出　　处：《太原学院学报（自然科学版）》2024年第4期49-55,共7页Journal of TaiYuan University:Natural Science Edition

基　　金：安徽省教育厅质量工程项目(2021gkszgg018);安徽省高校自然科学研究项目(2023AH040384)。

摘　　要：针对LSTM应用于城市轨道短时流量预测存在的模型参数确定困难、对预测精度影响大的问题,采用改进的BA算法对模型参数进行优化。对传统BA算法,采用自适应策略来动态调整脉冲频率和蝙蝠速率,同时蝙蝠位置更新模型中引入随机扰动项,提高了BA的优化性能。采用BA对LSTM参数进行优化,提出了基于改进LSTM的城市轨道短时流量预测模型。将提出的模型应用于郑州地铁1号线,通过和BP神经网络预测模型、LSTM预测模型的对比,验证了所提出的改进LSTM预测模型具有更高的预测精度。Aiming at the difficulties in determining the model parameters of LSTM applied to urban rail short-term flow prediction,which has great influence on prediction accuracy,the improved BA algorithm is used to optimize the model parameters.For the traditional BA algorithm,the adaptive strategy is used to dynamically adjust the pulse frequency and bat rate,and the random disturbance term is introduced into the bat position update model,which improves the optimization performance of BA.Using BA to optimize LSTM parameters,a short-term urban rail flow prediction model based on improved LSTM is proposed.The proposed model is applied to Zhengzhou Metro Line 1,and compared with BP neural network prediction model and LSTM prediction model;the improved LSTM prediction model proves to have higher prediction accuracy.

关 键 词：改进蝙蝠算法 长短时记忆网络 城市轨道交通 短时客流量预测 

分 类 号：U293.13[交通运输工程—交通运输规划与管理]