Founded in 1971 >
Chinese Sci-tech Core Periodicals >
British Science Abstracts (SA, INSPEC) Indexed Journals >
United States, Cambridge Scientific Abstract: Technology (CSA: T) Indexed Journals >
United States, Ulrich's Periodicals Directory(UPD)Indexed Journals >
United States, Cambridge Scientific Abstract: Natural Science (CSA: NS) Indexed Journals >
Poland ,Index of Copernicus(IC) Indexed Journals >
International Standard Serial Number:
ISSN 1001-4551
Sponsor:
Zhejiang University;
Zhejiang Machinery and Electrical Group
Edited by:
Editorial of Journal of Mechanical & Electrical Engineering
Chief Editor:
ZHAO Qun
Vice Chief Editor:
TANG ren-zhong,
LUO Xiang-yang
Tel:
86-571-87041360,87239525
Fax:
86-571-87239571
Add:
No.9 Gaoguannong,Daxue Road,Hangzhou,China
P.C:
310009
E-mail:
meem_contribute@163.com
Optimization of soft sensor modeling methods of generator sets based on data-driven
CHEN Kai-liang, CHEN Jian-hong, SHENG De-ren, LI Wei
(Institute of Thermal Science and Power System, Zhejiang University, Hangzhou 310027, China)
Abstract: In order to solve the problems existing in generating sets system,such as soft sensing of some inaccurate measured parameters,accuracy test of sensor measurement data,a new modeling method based on generalized regression neural network(GRNN)and partial-least squares regression(PLSR)was proposed. Firstly,a GRNN model was built with variables after the mechanism analysis to assess the average contribution rate of each independent variable on the dependent variable in the model and filter out the main modeling parameters. Secondly,the relevant parameters were modeled with a PLSR method transformed by the cubic B-spline transformation,which is an effective solution to the nonlinear modeling and multicollinearity problems. Then,the final simplified and reliable model was founded. The computational result of the project example shows that this modeling approach to sensor measurement data under different conditions fits well on both accuracy and generalization ability. Importantly,only a few parameters need to be saved,so the model is more suitable to solve the above-mentioned online problems.
Key words: generating sets system; data-driven; soft sensor modeling; generalized regression neural network(GRNN); partial-least squares regression(PLSR); NN-PLS