Soft Sensor for the Mooney-Viscosity Based on PCA-LSSVM

Mooney-viscosity is the dominate quality index for synthetic rubber. Monitoring the Mooney-viscosity effectively and realizing automatic quality control of the production process is an urgent problem in the rubber industry. This paper proposes a soft sensor model based on PCA-LSSVM to predict the Mooney-viscosity of styrene butadiene rubber (SBR). First, major parameters affecting the Mooney-viscosity were chosen based on mechanism analysis. The principal components were extracted by PCA and used as the secondary variables of SVM. Then a soft sensor model for the Mooney-viscosity was established by LSSVM. The simulation results show that the maximum relative error of Mooney-viscosity was low and acceptable at 5.78%. This data may be used to efficiently guide production.