Study of Monitoring the Real Liquid Level of the SBR Discharge System Accurately Based on Soft-sensing

Abstract

The false liquid level of discharge system is an important factor affecting the safety of SBR production process. A soft-sensing model is proposed to monitor the real liquid level of the discharge system. As the auxiliary variables, the feed pressure of manifold PJ, the inlet temperature of the discharge chute T and the pressure of the butadiene separation system PC are selected. PSO-SVM method is used in the soft-sensing model to predict the liquid level of the discharge chute. The strong predictive performance of the model will have a significant effect on the process safety of SBR.