Application of SVM based on Miztures of Kernels in soft-sensor for Rare Earth Countercurrent Eztraction Process
In this paper, in virtue of the problem of rare-earth counter-current extraction separation process, in which the real-time online measuring for component content is very difficult, a modeling method of Support Vector Machine (SVM) based on mixtures kernels for rare-earth counter-current extraction separation process is proposed. The model makes use of the character of mixture kernel by more global and local ability and the influence of difference kernels which can be turned by weight factor in the determination of the kernels. According to the results of application, it indicates that the method based on mixtures kernels has both better fitting output and satisfied prediction output, and meets the modeling and control for rare-earth extract process.
Component Content Miztures of Kernels SVM Modeling
Lu Rongxiu Yang Hui Zhong Lusheng
School of Electrical and Electronics Engineering, East China Jiaotong University, Jiangxi, Nanchang 330013
国际会议
2009年中国控制与决策会议(2009 Chinese Control and Decision Conference)
广西桂林
英文
5761-5764
2009-06-17（万方平台首次上网日期，不代表论文的发表时间）