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) ›› 2020, Vol. 39 ›› Issue (11): 99-108.DOI: 10.3969/j.issn.1674-0696.2020.11.15

• Highway & Railway Engineering • Previous Articles     Next Articles

Application of Series IGM-FM Combination Model in Prediction of Subgrade Settlement of High-Speed Railway

ZHANG Xianzhou1,2,XIA Chenxi1,CHEN Xiao1,JIANG Yinghao1,CHEN Jianying1   

  1. (1. Faculty of Geosciences and Environmental Engineering, Southwest Jiaotong University, Chengdu 611756, Sichuan, China; 2. National and Local Joint Engineering Laboratory for Safety Space Information Technology for High-Speed Railway Operation, Chengdu 611756, Sichuan, China)
  • Received:2019-06-24 Revised:2019-07-20 Online:2020-11-19 Published:2020-11-23

IGM-FM串联模型在高铁路基沉降预测中的应用

张献州1,2,夏晨翕1,陈霄1,蒋英豪1,陈建营1   

  1. (1. 西南交通大学 地球科学与环境工程学院,四川 成都 611756; 2. 高速铁路运营安全空间信息技术国家地方联合工程实验室,四川 成都 611756)
  • 作者简介:张献州(1962—),男,河南扶沟人,教授,博士,主要从事大地测量与测量工程的研究。E-mail: xzzhang@swjtu.edu.cn 通信作者:夏晨翕(1995—),男,四川成都人,硕士研究生,主要从事精密工程测量与变形监测的研究。E-mail:chenxixia9592@Foxmail.com
  • 基金资助:
    中央高校基本科研业务费专项资金项目(SWJTU10ZT02)

Abstract: The settlement of high-speed railway subgrade is characterized by small deformation and large fluctuation of data. Aiming at the problem that the single prediction models such as exponential curve method, three-point method and grey system cannot properly reflect the settlement and deformation law of high-speed railway subgrade in construction period, the improved grey prediction Fourier-Markov residual combination prediction model was used to predict the rule of series superposition of subgrade settlement deformation data, in order to improve the overall accuracy and robustness of the prediction. The proposed model was applied to the measured data of subgrade settlement of a high-speed railway through linear interpolation. The research results show that the combination prediction model is effective. By using Fourier series and Markov state transition matrix, the quadratic combination prediction of the primary residual is predicted, which can better fit the change trend of high-speed railway subgrade settlement monitoring data, and the prediction accuracy has been significantly improved.

Key words: railway engineering, monitoring of subgrade settlement and deformation, improved GM (1,1) model, Fourier series, Markov chains, series combination forecasting

摘要: 高速铁路路基沉降素有“变形量级小、数据波动大”的特点。针对现有对指数曲线法、三点法、灰色系统等单项预测模型不能恰当地反映高速铁路路基建设周期内沉降变形规律的问题。笔者利用改进灰色预测傅里叶-马尔科夫残差式组合预测模型,对路基沉降变形数据存在串联式叠加的规律进行预测研究,以期提高预测的整体精度和稳健性。应用该模型于经过线性插值补全的某高速铁路路基沉降实测数据中,研究结果表明:该组合预测模型构造有效,利用傅里叶级数和Markov状态转移矩阵对一次残差进行二次组合预测,更好地拟合高速铁路路基沉降监测数据的变化趋势,预测精度得到明显提高。

关键词: 铁道工程, 路基沉降变形监测, 改进GM(1, 1)模型, 傅里叶级数, Markov链, 串联式组合预测

CLC Number: