Multi-scale linear programming support vector regression for ethylene rectification modeling
Standard support vector regression has difficulty in solving the complex problem of nonlinear modeling. However, multi-scale linear programming support vector regression can better deal with it. This method constructs the objective decision function by a linear combination of the multi-feature spaces, and applies the linear programming instead of quadratic programming to solve the regression problem. Consequently it reduces the computing complexity. Furthermore particle swarm algorithm is used effectively for the regression parameters selection. Ultimately a mixing function experiment has demonstrated the great effectiveness of this method, and it is also applied to the soft sensing modeling of ethylene rectification.
support vector regression multi-scale,linear programming particle swarm optimization
YU Yan-fang QIAN Feng
State-Key Laboratory of Chemical Engineering ,East China University of Science and Technology,Shanghai 200237,China
国际会议
西安
英文
2007-08-15（万方平台首次上网日期，不代表论文的发表时间）