• DocumentCode
    1896096
  • Title

    Research of Groundwater Environment Early Warning Based on Intelligent Algorithm

  • Author

    Wu, Yuchun ; Long, Xiaojian

  • Author_Institution
    Sch. of Electron. & Inf. Eng., Jinggangshan Univ., Ji´´an, China
  • Volume
    1
  • fYear
    2012
  • fDate
    23-25 March 2012
  • Firstpage
    536
  • Lastpage
    540
  • Abstract
    According to the need for groundwater health evaluation and its non-linear computation difficulty, on the basis of the problem effectively solved by Intelligent algorithm, this paper focused on analyzing calculation principle and characteristics about the BP neural network and support vector machine algorithm in groundwater health prediction model, designed the calculation model of support vector machine based on multi-level classifier, rough set theory was introduced to support vector machine calculation process optimization. It solved complex non-linear relationship between the small sample and groundwater health degreed, and speeded up the convergence speed, effectively improved the prediction accuracy and stability.
  • Keywords
    backpropagation; condition monitoring; environmental science computing; groundwater; neural nets; pattern classification; rough set theory; support vector machines; BP neural network; backpropagation; calculation principle; groundwater environment early warning; groundwater health evaluation; groundwater health prediction model; intelligent algorithm; multilevel classifier; rough set theory; support vector machine algorithm; Artificial neural networks; Biological neural networks; Classification algorithms; Computational modeling; Kernel; Neurons; Support vector machines; BP neural network; Rough set theory; Support vector machine; early warning;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Computer Science and Electronics Engineering (ICCSEE), 2012 International Conference on
  • Conference_Location
    Hangzhou
  • Print_ISBN
    978-1-4673-0689-8
  • Type

    conf

  • DOI
    10.1109/ICCSEE.2012.321
  • Filename
    6187903