Soft sensor modeling based on rough set and least squares support vector machines

Authors:
Li Chuan;Wang Shilong;Zhang Xianming;Xu Jun
Affiliations:
Institute of Separation Machinery, Chongqing Technology and Business University, Chongqing, China and College of Mechanical Engineering, Chongqing University, Chongqing, China;College of Mechanical Engineering, Chongqing University, Chongqing, China;Institute of Separation Machinery, Chongqing Technology and Business University, Chongqing, China;Institute of Separation Machinery, Chongqing Technology and Business University, Chongqing, China
Venue:
IMCAS'07 Proceedings of the 6th WSEAS International Conference on Instrumentation, Measurement, Circuits and Systems
Year:
2007

Citing 4
Cited 0

Least Squares Support Vector Machine Classifiers

Neural Processing Letters
Feature extraction using rough set theory and genetic algorithms--an application for the simplification of product quality evaluation

Computers and Industrial Engineering
Formulations of Support Vector Machines: A Note from an Optimization Point of View

Neural Computation
An overview of statistical learning theory

IEEE Transactions on Neural Networks

Quantified Score

Hi-index	0.00

Visualization

Abstract

Soft sensor is an effective tool to estimate industrial process variables which are hard to be measured online for the technical or economical reasons. The modeling methods of the sensor are related to the approximating precision and speed. A soft sensor model with rough set and Least Squares Support Vector Machines (LSSVM) is presented in the paper. The rough set is employed to compress the data for preprocessing, which can get rid of the multicollinearity and reduce the dimension of input variables for the model. To solve the nonlinear and multiple input characteristics of industrial process, the LSSVM is delivered for model regression. The model is applied for moisture content soft sensing of vacuum oil purifier. The result shows that the proposed method features high speed and precise approximation ability, which has better performance of generalization for tracking the trend of the moisture content variety during oil purification.