中文核心期刊
CSCD来源期刊
中国科技核心期刊
RCCSE中国核心学术期刊

Journal of Chongqing Jiaotong University(Natural Science) ›› 2018, Vol. 37 ›› Issue (06): 81-86.DOI: 10.3969/j.issn.1674-0696.2018.06.13

• Traffic & Transportation Engineering • Previous Articles     Next Articles

Prediction Method of Dynamic Stochastic Effective Parking Space

DUAN Manzhen,CHEN Guang,ZHANG Lin,MI Xueyu   

  1. (School of Civil and Architectural Engineering,North China University of Science and Technology, Tangshan063009,Hebei,P.R.China)
  • Received:2016-09-29 Revised:2016-11-15 Online:2018-06-12 Published:2018-06-12

动态随机有效停车泊位预测方法

段满珍,陈光,张林,米雪玉   

  1. (华北理工大学 建筑工程学院,河北 唐山 063009)
  • 作者简介:段满珍(1974—),女,河北滦县人,副教授,博士,主要从事城市智能交通方面的研究。E-mail:1033838477@qq.com。
  • 基金资助:
    国家自然科学基金项目(51378171;61374157);河北省软科学计划项目(16456232)

Abstract: In order to satisfy the demand for the personalized parking guidance,an effective parking space forecast method under the dynamic stochastic condition was researched.Utilizing the C-C algorithm in a nonlinear relational model,the integration process of the C-C algorithm and the Elman recursive network were researched.During the research process,the chaotic characteristics of the phase space reconstruction were verified using Small Data.A dynamic stochastic parking forecast model was built,and a model algorithm was designed.The integration algorithm was carried out by MATLAB with real data,and the results show that the predictions are fairly consistent with the actual value and the maximum relative error.The average relative error and mean absolute error are both less than the results of the forecast method based on linear assumption.This illustrates that the integration of the C-C algorithm and network technology in the dynamic stochastic parking space forecast is effective.

Key words: traffic engineering, dynamic stochastic parking space, parking, C-C algorithm, phase space reconstruction, Elman network

摘要: 为满足个性化停车诱导需求,研究动态随机条件下有效停车泊位预测方法。利用C-C算法在求解非线性关系模型方面的优势,研究C-C算法与Elman递归网络技术的融合过程,采用小数据量法验证重构相空间的混沌特性,在此基础上研究动态随机泊位预测模型和求解算法。利用MATLAB对融合算法进行仿真实验分析。结果表明:模型预测结果与实际值一致性较好,最大相对误差、平均相对误差和平均绝对误差均小于线性假设的预测方法。说明C-C算法与网络技术的融合算法在动态随机泊位预测方面的有效性。

关键词: 交通工程, 动态随机泊位, 停车, C-C算法, 相空间重构, Elman网络

CLC Number: