计算机工程与应用 ›› 2019, Vol. 55 ›› Issue (20): 216-221.DOI: 10.3778/j.issn.1002-8331.1903-0262

• 工程与应用 • 上一篇    下一篇

基于EEMD-GWO-LSSVM的公共交通短期客流预测

王盛,杨信丰   

  1. 兰州交通大学 交通运输学院,兰州 730070
  • 出版日期:2019-10-15 发布日期:2019-10-14

Short-Term Passenger Flow Forecasting of Public Transport Based on EEMD-GWO-LSSVM

WANG Sheng, YANG Xinfeng   

  1. School of Traffic & Transportation Engineering, Lanzhou Jiaotong University, Lanzhou 730070, China
  • Online:2019-10-15 Published:2019-10-14

摘要: 为了提高大型公共交通短期客流预测精度,提出了一种在利用集成经验模态分解原始数据的条件下,采用灰狼优化算法优化最小二乘支持向量机(EEMD-GWO-LSSVM)的算法,利用该算法实现城市大型公共交通短期客流预测。该模型采用EEMD分解原始数据,将分解后的各个本征模函数(IMF)分量运用最小二乘支持向量机进行回归预测,最小二乘支持向量机的预测参数由灰狼算法进行优化。通过对西安地铁二号线北客站一个月进出站人数进行训练预测,将预测结果和支持向量机(SVM),自回归移动平均模型(ARIMA),仅利用灰狼优化参数的最小二乘支持向量机(GWO-LSSVM)算法以及基于交叉检验进行参数优化的最小二乘支持向量机进行对比,分析得出该算法具有更加精确的预测结果。

关键词: 公共交通, 短期预测, 灰狼优化, 最小二乘支持向量机

Abstract: In order to improve the accuracy of short-term passenger flow forecasting for large-scale public transport, an optimization algorithm of Least Squares Support Vector Machine(EEMD-GWO-LSSVM) based on Grey Wolf Optimization algorithm is proposed under the condition of decomposing the original data with integrated empirical mode. The algorithm is used to realize short-term passenger flow forecasting for large-scale public transport. The model uses EEMD to decompose the original data, and uses LSSVM to predict the decomposed IMF components. The prediction parameters of LSSVM are optimized by grey wolf algorithm. By training and forecasting the number of people entering and leaving the North Passenger Station of Xi’an Metro Line 2 in a month, the forecasting results and Support Vector Machine(SVM), Autoregressive Integrated Moving Average Model(ARIMA) are predicted, and the Least Squares Support Vector Machine(GWO-LSSVM) algorithm based on Grey Wolf Optimization parameters and cross-checking are used only. The Least Squares Support Vector Machines(LS-SVMs) with row parameter optimization are compared and the results show that the proposed algorithm has more accurate prediction results.

Key words: public transportation, short term prediction, grey wolf optimization, least square support vector machine