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) ›› 2012, Vol. 31 ›› Issue (5): 997-1001.DOI: 10.3969/j.issn.1674-0696.2012.05.19

Previous Articles     Next Articles

Pavement Performance Combining Forecasting Based on BP Neural Network and Markov Model

Zhou Pengfei,Wen Shengqiang,Kang Haigui   

  1. Faculty of Infrastructure Engineering,Dalian University of Technology,Dalian 116024,Liaoning,China
  • Received:2011-12-26 Revised:2012-03-29 Online:2012-10-15 Published:2015-01-22

基于马尔可夫链与神经网络组合的路面使用性能预测

周鹏飞,温胜强,康海贵   

  1. 大连理工大学建设工程学部,辽宁大连11602
  • 作者简介:周鹏飞(1977—),男,河南卫辉人,讲师,博士,主要从事交通规划与管理方面的研究。E-mail:pfzhou@yeah.net。
  • 基金资助:
    国家自然科学基金项目( 71101014) ; 河南交通厅科技计划项目( 200912)

Abstract: Based on the analysis of important effects of the pavement performance,a pavement performance combining method is proposed with neural network and Markov process. The model combining forms are discussed and the parameters optimization is suggested. Then using actual survey datum of a highway in Henan,experiments are studied and the results show that the proposed combining model is capable to improve pavement performance forecasting by choosing suitable combining forms and their parameters.

Key words: road engineering, pavement performance, combining forecasting, Markov model, BP neural network

摘要: 在分析路面使用性能影响因素的基础上,结合马尔可夫过程链预测与BP 神经网络预测的优势,提出一种组 合预测模型; 探讨了模型组合形式及参数确定方法; 结合河南某高速公路工程实测资料进行了实证分析。结果表 明: 通过预测模型的合理组合与参数的优选,能够有效地提高预测精度。

关键词: 道路工程, 路面使用性能, 组合预测, 马尔可夫过程链, BP 神经网络

CLC Number: