Title: Modified Method of Nonlinear PLS and its Application ——Based on Chebyshev Polynomial
Abstract: In practice, there is abundance correlativity with process variables, even nonlinear correlativity. The model parameter using Partial Least Square (PLS) is calculated, which could eliminate the correlativity. However, the general PLS fails to eliminate the nonlinear correlativity because it contacts the input and output variables using linearity relation. A new method of nonlinear PLS is developed, which is based on Chebyshev Polynomial. The new method is a nonlinear method, which makes use of the orthogonal character of Chebyshev Polynomial, first orthogonalize and linearizated process variables, then using PLS method to calculate model parameters and get accurate model. This method considers selfcorrelative and jointcorrelative relations among input variables, and can be used to express the nonlinearity more effective than general PLS and provide new thoughtaway to study nonlinear process.
Publication Year: 2003
Publication Date: 2003-01-01
Language: en
Type: article
Access and Citation
Cited By Count: 1
AI Researcher Chatbot
Get quick answers to your questions about the article from our AI researcher chatbot