Computer Engineering and Applications ›› 2008, Vol. 44 ›› Issue (14): 205-207.

• 工程与应用 • Previous Articles     Next Articles

SSVR algorithm and time series prediction of furnace combustion state

ZHANG Xin1,WANG Bing2,ZHAO Pu1   

  1. 1.College of Electronics and Information Engineering,Hebei University,Baoding,Hebei 071002,China
    2.College of Mathematics and Computer Science,Hebei University,Baoding,Hebei 071002,China
  • Received:2007-12-24 Revised:2008-03-03 Online:2008-05-11 Published:2008-05-11
  • Contact: ZHANG Xin


张 欣1,王 兵2,赵 璞1   

  1. 1.河北大学 电子信息工程学院,河北 保定 071002
    2.河北大学 数学与计算机学院,河北 保定 071002
  • 通讯作者: 张 欣

Abstract: The Support Vector Regression(SVR) is used for the time series analysis and prediction to resolve the complex nonlinear system modeling problems.The smooth method is introduced to improve the standard SVR algorithm in order to reduce calculation complexity.In this paper,the Smooth Support Vector Regression(SSVR)is applied for the time series prediction of furnace combustion states.By clustering and analyzing the furnace flame images,calculated the state coefficient denoting the furnace combustion states,constructed the state coefficient time series,and built the prediction model using the SSVR algorithm.The experimental results show that SSVR has faster convergence speed and higher fitting and prediction precision,which effectively extends the application of SVR.

Key words: smooth support vector regression, furnace combustion state, time series prediction

摘要: 用支持向量回归(SVR)的方法分析和预测时间序列,可解决复杂非线性系统的建模问题。采用光滑化方法对SVR的基本算法进行改进,可降低计算的复杂度。将光滑支持向量回归(SSVR)算法应用于炉膛燃烧状态时间序列预测。对炉内火焰图像进行聚类分析,计算表征炉膛燃烧状态的状态指数,建立状态指数时间序列,并利用光滑支持向量回归算法构建预测模型。实验结果表明,SSVR方法具有更快的收敛速度、更好的拟合精度和良好的预测性能。

关键词: 光滑支持向量回归, 炉膛燃烧状态, 时间序列预测