Muthanna Journal of Engineering and Technology
Volume (12), Issue (2), Year (30 December 2024), Pages (09-18)
DOI:10.52113/3/eng/mjet/2024-12-02/09-18
Research Article By:
Ashraf A. Salih, Thaer A. Abdulla and Ahmed S. Abdullah
Corresponding author E-mail: asraf.a.salih10479@st.tu.edu.iq
ABSTRACT
This research examines the process of choosing between the reliable proportional-integral-derivative (PID) controller and the recommended model predictive control (MPC) controller for the vacuum distillation unit (VDU) by analyzing its dynamic behavior. Aspen HYSYS V12.1 and MATLAB Simulink Environment (2021 A) software programs were used in this study to compare the performance of the PID and MPC controllers to analyze the overhead temperature response. The temperature response in the middle of the tower (stage No. 15) is both a basic factor in improving product quality and quantity through a set point (SP) step change in the mass flow rate of the feed stream, feeding temperature, and flow rate of stripping steam. In addition, to study the effect of a disturbance step change on the feed pump efficiency, as well as making a change in the overhead temperature and the temperature of stage No. 15 to clarify the performance difference between the (PID) and (MPC), The study result showed that MPC is better than PID in reaching the desired value and overcoming the turbulence effect, The overhead and (stage No.15) temperatures response time to reach steady state, it is found that the MPC response time shorter than the PID with respect to the efficiency disturbance by (81% and 293%), step change in feed flow rate by (43% and 47.2%), feed temperature by (440% and 158%), low pressure steam (LPS) (81% and 290%) and set point (86.6% and 218%), In the overshot and raising time, the MPC shows better temperature overshoot and raising time in all step change cases taken than the PID except the overhead temperature in feed temperature step change shows that PID is better in small temperature difference, and steady state error percentage for the MPC shows that it is zero or approximately zero in the inputs step change but for the PID the steady state error is not zero like in step change in feed flow rate and temperature which are 0.01 and 0.02 respectively.
This study shows how the efficiency of the distillation separation process can be increased through the prediction disturbance, overcoming it and high response time to reach the desired value, which increases the production quality, and quantity and reduces the energy associated with the production process (steam, fuel, electricity, etc.).
Keywords: Vacuum distillation unit, model predictive control, proportional integral derivative, LPS, vacuum diesel oil, heavy vacuum gas oil, light vacuum gas oil.