Formaldehyde production process control performance improvement using model predictive control

Abstract

This article demonstrates the improvement of control performance in formaldehyde production process using model predictive control (MPC) in comparison between conventional proportional-integral control. MPC is an advance process control which can improve the performance of a control process in terms of time delay, open loop instability, constraints, and thereof combinations. MPC will reduce the variance in the control variable that affects the process to operate closer to physical constraints. The empirical model of the MPC controller is based on the process reaction curve (PRC) by using the first order plus dead time (FOPDT) approach. Four controllers which were flow control (FIC-102), temperature control (TIC-101), pressurce control (PIC-101), and liquid level control (LIC-101) were tested by changing the set points (SP) and giving disturbances. The performance indicator for the controllers are shown by their value of integral of absolute error (IAE) and integral of square error (ISE). The results show that the MPC improved the controllers’ performance either tested by changing SP or giving disturbance and are better in terms of IAE or ISE.