基于 SSA-SVR 模型的城市轨道 交通短时进站客流预测

针对现有城市轨道交通短时客流量预测单一模型可能存在预测不稳定的问题，提出一种基于奇异谱分析 (singular spectrum analysis，SSA)和支持向量回归(SVR)相组合的预测模型。该组合模型利用奇异谱分析(SSA)将轨 道交通原始时间序列客流数据进行分解和重构，对重构后的时间序列按奇异值从大到小进行排序，得到含有原始 时间序列数据主要信息成分的重构序列，将重构后的时间序列作为支持向量回归模型(SVR)的输入条件，最后进 行各站点的短时进站客流预测。采集 2015 年 11 月北京市全网的城市轨道交通进站客流数据，对提出的短时客流 预测模型进行验证和对比分析。结果表明，组合模型预测精度相比 ARIMA、SVR、CNN-LSTM 和 T-GCN 模型具 有更高的预测精度和更稳定的预测表现，具有一定的实际意义。

Prediction of Short-term Inbound Passenger Flow of Urban Rail Transit Based on the Singular Spectrum Analysis and Support Vector Regression Model

Considering the existing forecasting model of short-term traffic in the urban rail transit to predict instability problems, this paper proposes a new model based on singular spectral analysis (SSA) in combination with the support vector regression (SVR) forecasting model. The combined model uses SSA to decompose and reconstruct the original time-series passenger flow data of rail transit. In addition, this model sorts the reconstructed time series by singular values (ranked from large to small) to obtain the main information containing the original time series data. The reconstructed sequence of the components uses the reconstructed time series as the input of the SVR, and the short-term inbound passenger flow prediction of each station is finally performed. This paper collects the urban rail transit passenger flow data of the entire network in Beijing from November 2015 and validates and compares the proposed short-term passenger flow prediction model. The results show that the prediction accuracy of the combined model proposed in this paper is higher and more stable than those of the ARIMA, SVR, CNN-LSTM, and T-GCN models; the improved accuracy characteristics have practical significance.