• DocumentCode
    3579820
  • Title

    Fuzzy Self-Adaptable Prediction Method for High-Fill Road Foundation Settlement

  • Author

    Xiangxing Kong ; Hongzhe Li

  • Author_Institution
    Northern Center of Bus. Manage., China Commun. Constr. Co. Ltd., Xi´an, China
  • Volume
    1
  • fYear
    2014
  • Firstpage
    216
  • Lastpage
    218
  • Abstract
    In the forecasting of road foundation settlement, different methods can often provide valuable information. When different forecasting methods combined properly, comprehensive forecast information are variously utilized. According to the principle of artificial intelligence and method of fuzzy mathematics, the fuzzy combination forecasting method of self-adaptively variable weight is proposed. Based on matching degree of past-recent period and actual value, the weights are automatically adjust, so that forecasting results will be accurate. The fuzzy combination forecasting method of self-adaptively variable weight is applied to the sub grade settlement in Hengzao expressway, which provides the prediction accuracy and reliable applicability. In order to improve the prediction accuracy, the measured sample points should increase, and with the passage of time, the recent settlement results should be set to the set of original sample.
  • Keywords
    artificial intelligence; forecasting theory; foundations; fuzzy set theory; fuzzy systems; prediction theory; road building; roads; Hengzao expressway; artificial intelligence; forecast information; forecasting methods; fuzzy combination forecasting method; fuzzy mathematics; fuzzy self-adaptable prediction method; high-fill road foundation settlement; self-adaptively variable weight; Forecasting; Load modeling; Mathematical model; Predictive models; Roads; Soil; different weight; fuzzy prediction method; road foundation settlement; selt-adaptable;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Computational Intelligence and Design (ISCID), 2014 Seventh International Symposium on
  • Print_ISBN
    978-1-4799-7004-9
  • Type

    conf

  • DOI
    10.1109/ISCID.2014.189
  • Filename
    7064176