






摘 要:準確預(yù)測城市軌道交通短時客流量的變化,有助于運營部門做出決策,并幫助軌道交通集團提高服務(wù)水平和實現(xiàn)智慧化運營。然而,客流數(shù)據(jù)的動態(tài)性和隨機性使短時客流預(yù)測變得困難,因此,文章提出了一種組合預(yù)測模型,將Transformer模型中的位置編碼(Positional Encoding)層與長短期記憶(Long Short-Term Memory,LSTM)神經(jīng)網(wǎng)絡(luò)相結(jié)合,構(gòu)建了LSTM-Transformer預(yù)測模型。隨后以青島市的106個站點的進站客流數(shù)據(jù)為研究對象,并使用聚類算法對站點進行聚類分析。在10分鐘的時間粒度下,利用前四周的客流數(shù)據(jù)作為訓(xùn)練數(shù)據(jù),對未來一天的客流數(shù)據(jù)進行預(yù)測研究。同時,將差分自回歸移動平均模型(Auto-Regressive Integrated Moving Average,ARIMA)、LSTM、GA-SLSTM和Transformer作為對照模型進行驗證。通過多組實驗證明了文章提出的LSTM-Transformer模型相較于對照模型組具有更好的預(yù)測精度和實用性。
關(guān)鍵詞:智能交通;城市軌道交通;短時客流預(yù)測;聚類算法;LSTM-Transformer模型
中圖分類號:F570;U293.13 文獻標志碼:A DOI:10.13714/j.cnki.1002-3100.2024.14.020
文章編號:1002-3100(2024)14-0103-05
Short-Term Passenger Flow Prediction of Urban Rail Transit Based on LSTM-Transformer
(1. School of Traffic Engineering, Shandong Jianzhu University, Jinan 250101, China; 2. Institute of Road Traffic Safety, Shandong Police College, Jinan 250014, China; 3. Jinan Zhiye Electronics Co., Ltd., Jinan 250013, China)
Abstract: Accurately predicting changes in short-term passenger flow for urban rail transit is crucial for operational decision-making and improving service levels and intelligent operations within rail transit groups. However, the dynamic and stochastic nature of passenger flow data presents challenges in short-term prediction. To address this, the study proposes a combined prediction model, the LSTM-Transformer, which integrates the Positional Encoding layer from the Transformer model with the Long Short-Term Memory (LSTM) neural network. The paper focuses on the inbound passenger flow data from 106 stations in Qingdao and conducts clustering analysis using clustering algorithms to group the stations. Subsequently, based on a 10-minute time granularity, the paper utilizes passenger flow data from the preceding four weeks as training data to predict and analyze the passenger flow for the following day. Additionally, the paper compares LSTM-Transformer model with several control models, including the Differential Autoregressive Integrated Moving Average (ARIMA), LSTM, GA-SLSTM, and Transformer. Through multiple experiments, the study demonstrates that the proposed LSTM-Transformer model outperforms the control models in terms of prediction accuracy and practicality.
Key words: intelligent transportation; urban rail transit; short-term passenger flow prediction; clustering algorithm; LSTM-Transformer model
收稿日期:2023-11-29
基金項目:國家自然科學(xué)基金項目(71871130,71971125);山東省公安廳科技服務(wù)項目(SDGP370000000202202004905,SDGP370000000202202006498)
作者簡介:張思楠(1999—),男,陜西咸陽人,山東建筑大學(xué)交通工程學(xué)院碩士研究生,研究方向:智能交通;李樹彬(1977—),本文通信作者,男,山東聊城人,山東建筑大學(xué)交通工程學(xué)院,山東警察學(xué)院道路交通安全研究所,教授,博士,碩士生導(dǎo)師,研究方向:系統(tǒng)分析與集成、智能交通系統(tǒng)。
引文格式:張思楠,李樹彬,曹永軍.基于LSTM-Transformer的城市軌道交通短時客流預(yù)測[J].物流科技,2024,47(14):103-106,114.
隨著城市人口不斷增加,出行引起的環(huán)境污染、交通延誤和交通堵塞等問題頻繁出現(xiàn)。……