Simulation & Control of Crude Oil Desalter of Khartoum Refinery

Abstract

A desalting unit of crude oil is required to reduce the salt content in the crude to an acceptable value. Then the control technique is required to change the desalting process properly in order to eliminate the unfavorable effect of the disturbances which is reflected on the amount of salts that existed in the desalted crude. The concentration of calcium salts that exist in the crude oil of Khartoum refinery is about 450 – 600 ppm to be reduced to about 11 ppm, hence different control mechanisms were proposed for controlling the concentration of calcium salts in the desalted crude oil of Khartoum refinery. The flow rate of the decalcium agent (decalcinizer) was used to manipulate the deviation in concentration from the set point, then the mathematical model was determined using Matlab software and the transfer function was estimated using the system identification toolbox. Simulink feature of Matlab was used to draw the control loop and the optimum settings of controller parameters were estimated by tuning the controller under PID-action, P-action, and PI-action, using two different methods, which are PID tunner method and Ziegler-Nichols (Z-N) tuning method. These settings were used for overshoot and peak deviation investigation, then the response curve was drawn for three types of the controller and the proportional integral derivative controller (PID Controller) resulted in highest value of controller gain lowest overshoot and peak deviation, and shortest settling time. Root Locus Niquest and bode criteria were used to investigate the system stability and the last two methods resulted in a stable system. Another mechanism for concentration control using the flow rate of the inlet water as a manipulated variable was proposed, the results were compared with that obtained from the flow rate of decalcinizer, and the decalcinizer flow rate gave better results than the inlet water flow rate.