Improving the Performance of a Continuous Stirred Tank Reactor Using Moving Horizon State Estimation and Model Predictive Control

CATEGORY: REACTORS
Improving the Performance of a Continuous Stirred Tank Reactor Using Moving Horizon State Estimation and Model Predictive Control Journal Advanced Materials Research (Volumes 403 - 408), Volume MEMS, NANO and Smart Systems, Pages 3454-3460 (2011)
Fazlollah Armoon, Hooshang Jazayeri-Rad
Abstract
Chemical reactors are used to produce specific products.  Consequently, concentration of products should be regulated at a specified level. As a disturbance input, a change in the inlet concentrations can vary the product concentration.  In order to regulate the product concentration, the inlet concentrations and the product concentration should be measured.
However, measurement of concentration encounters such problems as high cost and time delay. Authors propose estimation of concentration to compensate for these failures. The inlet concentration and the product concentration of a continuous stirred-tank reactor (CSTR) are estimated based on the moving horizon state estimation (MHSE). The product concentration is regulated based on the model predictive control (MPC). Simulation results indicate that the proposed strategy improves the performance of the CSTR compared with the method in which the inlet concentration is not estimated.
Full Text Source (Subscription or Fee): http://www.scientific.net/AMR.403-408.3454