基于大语言模型的交通拥堵预测研究

打开文本图片集

关键词：大语言模型；实时交通信息；交通拥堵预测；时间序列模型；城市交通中图分类号： 文献标志码：A DOI： 10.13714/j.cnki.1002-3100.2025.17.024

Abstract：Withtherapiddevelopmentof China’seconomy，urbanpopulationandvehiclenumbershavesurged，leadingtoincreasinglyserioustraffccongestionissues.Efectivelypredictingurbantraffccongestioniskeytoallviatingthseprobles.Tis paper selectsARIMA，SARIMA，andLSTMmodelsfortraffcflowpredictionandprovidesadetaledintroductiontothemathematicalalgoritmsandpredictionprocesesofthesemodels.TakingtheadministrativeditrictsofSuzhouasanexample，real-time trafic dataandroadconditiondata wereobtainedusingtheBaidu Maps trafic big dataplatformandtheAmapopen platform. Thispapercomparesthepredictionefectivenessofthethreemodelsfromfiveaspects：erorvaluesresidualdistributiontraining los，time-seriespredictionerorndmodelperformance，ultimatelydemonstratingtattheLSTmodelhasbeterpredictiveac curacy and practicality compared to the ARIMA and SARIMA models.

Keywords：largelanguagemodels;real-timetraffcinformationtrficcongestionprediction;tieseriesmodels;urbantransport

0引言

截至2024年6月底，公安部发布最新统计数据，我国机动车保有量达4.4亿辆，其中汽车3.45亿辆，新能源汽车2472万辆]。（剩余5963字）