计算机工程与应用 ›› 2009, Vol. 45 ›› Issue (1): 242-244.DOI: 10.3778/j.issn.1002-8331.2009.01.073

• 工程与应用 • 上一篇    下一篇

基于智能支持向量回归的瓦斯涌出量预测

戴宏亮1,2   

  1. 1.广东商学院 数学与计算科学学院,广州 510320
    2.中山大学 数学与计算科学学院,广州 510275
  • 收稿日期:2008-06-02 修回日期:2008-07-28 出版日期:2009-01-01 发布日期:2009-01-01
  • 通讯作者: 戴宏亮

Forecasting gas pushing based on intelligent support vector regression

DAI Hong-liang1,2   

  1. 1.Department of Mathematics and Computational Science,Guangdong University of Business Studies,Guangzhou 510320,China
    2.Department of Mathematics,Sun Yat-Sen (Zhongshan) University,Guangzhou 510275,China
  • Received:2008-06-02 Revised:2008-07-28 Online:2009-01-01 Published:2009-01-01
  • Contact: DAI Hong-liang

摘要: 进行瓦斯涌出量预测是保障安全生产的一个很重要步骤。由于瓦斯涌出量的非线性、不确定性,其预测是很复杂的一个问题。提出一种新的RGASVR智能模型,即基于实值遗传算法参数优选的支持向量回归模型,并且将提出的模型应用于瓦斯涌出量预测。实验结果表明,所提出的模型比BP神经网络、标准支持向量回归有更高的预测精度,具有较强的实用价值。

关键词: 支持向量回归, 实值遗传算法, 瓦斯涌出量, 预测

Abstract: Forecasting gas pushing is an important step to ensure production safety.It is a complicated problem due to its nonlinearity and uncertainty.In this study,a novel RGASVR model is proposed.The model is based on real-valued genetic algorithm to optimize the parameters of Support Vector Regression(SVR).In addition,the model is applied to forecast gas pushing.Experimental results show that RGASVR model performs better than BP neural networks and standard SVR,implying that RGASVR is very practical.

Key words: Support Vector Regression(SVR), real-valued genetic algorithm, gas pushing, forecasting