Jan-2015

Dynamic simulation of process controls

Simulation of a catalytic naphtha reformer confirmed the accurate performance of control instrumentation

SEPEHR SADIGHI and S REZA SEIF MOHADDECY
RIPI

Article Summary

Catalytic reforming is used to upgrade a significant portion of the low octane heavy naphtha that is unsuitable as a high octane gasoline blending component. A control system is essential to guarantee high productivity whilst the safety aspects of the process should be considered. In process control, it is often difficult to measure some important process variables such as composition of product or yield of process because of the limitations of measurement. This task is usually carried out by simulating the process in dynamic mode.

For almost 20 years, dynamic simulation has been used as an effective tool for DCS operator training and design verification. Dynamic simulation is useful throughout the entire lifetime of a plant: from conception to decommissioning. Some of the benefits of dynamic simulation for chemical processes are:
• Identify unit or process operating constraints at the conceptual design phase before capital expenditures are committed
• Improve the process
• Identify and correct weaknesses in the automation system design before commissioning and operation
• Demonstrate control applications before deployment
• Begin start-up of a new or modernised unit or process sooner, and complete it faster
• Train and develop operators on procedures and systems that are safe, optimised, and environmentally friendly
• Provide simulation applications for improving real-time operation and control.

Modern dynamic simulators can be used to provide this type of dynamic information. These flow sheet simulators are supported by strong databases, complete sets of modules and flexible simulation tools, so the user is capable of developing process-wide simulators, and studying the performance of the whole plant.

Most of the studies have focused on steady state modelling and the reaction section, that is fixed bed catalytic reactors in series. In the field of dynamic simulation, Hu et al developed a dynamic mode for the catalytic reforming plant and tried to study the effect of the feed flow rate and reactor temperatures on the yield of the process. Arani et al developed a dynamic model for the catalytic naphtha reformer which focused on the reaction section of this unit. The action of the controllers and dynamic responses were not studied in this work.

Therefore, the purpose of this research is to develop a dynamic simulator to study the action of 
the main controllers in the plant. In this way, a whole commercial fixed bed catalytic reforming process was selected, and was simulated using the Aspen-Hysys platform. After validating the simulation, the action of the controllers is studied in the dynamic mode using the set point tracking or servo method.

Process description
A commercial fixed bed catalytic naphtha reforming unit, called Platformer licensed from Chevron Research Corporation, was chosen as a case study. The feed of the plant prior to entering catalytic reforming should undergo hydrodesulphurisation (HDS) in the hydrotreating unit. Then the hydrotreated naphtha, called Platcharge, is introduced to the catalytic reforming process. The specification of the naphtha feed is shown in Table 1. This feed is treated in a Unifiner (naphtha hydrotreating) plant; therefore, the sulphur, nitrogen, olefin and metals contents are negligible.
Figure 1 shows that Platcharge is first preheated by the first furnace (Heater No. 1), and then enters the first reactor (Reactor No. 1) where naphthenes are dehydrogenated to aromatics. Then the product stream from the first reactor passes through the second reactor (Reactor No. 2), and the outlet stream enters the third reactor (Reactor No. 3). The overall reforming reactions are endothermic; therefore, a preheater (heaters Nos. 1, 2 and 3) should essentially be provided before each reforming reactor.

Next, the product stream from the last reactor enters the separator, wherein the hydrogen produced during the reforming process (the gas stream) is recycled, and is then mixed with the naphtha. Finally, the liquid product leaving the separator is introduced to the stabiliser in which LPG and light gases are separated from the gasoline. Hence the vapour pressure of the gasoline can be set according to market requirements.

The properties of the catalyst and its distribution for the naphtha reforming process are shown in Tables 2 and 3, respectively. Moreover, the normal operating conditions of the unit are shown in Table 4. The selected operating conditions of the process depend strongly on the life (activity) of the catalyst.

Process simulation
Catalytic reforming is often modelled and simulated based on the number of reactive species, 
and the type of kinetic model used. The presence of many components as reactants or intermediate products in the reactive mixture, as a result of the presence of new reactions, leads to a sophisticated challenge in modelling the process. To decrease these complications, reactants in the mixture are 
classified in certain and limited groups called pseudo-components. Additionally Arrhenius and Langmuir–Hinshelwood kinetics are widely used for kinetic-based catalyst modelling and simulation of catalytic naphtha reforming.

For this study, the kinetic constants presented in previous work were used. The feed of the catalytic reformer (see Table 1) was lumped according to the structural components using a discrete 
lumping procedure with 26 
pseudo-components (see Table 5).

Results and discussions Steady state simulation results
From start of run to end of run, actual test results were gathered from the target reforming unit. Because for a specified catalyst life the simulator was used for predicting the steady state condition of the Platformer, the data used for simulation should be selected from the normal condition when no abnormalities, such as tower flooding, emergency depressurisation, or pump or compressor shutdown, occurred. Before using these data to estimate the tuning parameters, it was necessary to validate them. If a reasonable overall mass balance (±5%) cannot be calculated, the validity of the test run is compromised.

After validating the data, the average absolute deviation (AAD%) of the simulation for the product flow rate, outlet temperatures of the first, second and third reactors, and the volume yield were 2.4%, 0.98%, 0.1%, 0.04% and 2.39%, respectively. A comparison between the simulated paraffinic, naphthenic and aromatic contents of the product against actual data is shown in Table 6.

 From the simulation results, it can be concluded that the simulation program was reliable enough for predicting the behaviour of the catalytic reforming unit being studied. It was assumed that the main source of deviations was the possibility of errors in gathering data through faults such as signal transmission, calibration and power fluctuation of instruments which could not be excluded from the actual data.

Dynamic simulation results
Before starting the dynamic simulation, it was necessary to define the size of equipment, according to its design values, in the Aspen-Hysys simulator. Moreover, the degree of freedom (DOF) of the dynamic simulation should be set to zero. This task was performed by setting the flow rate of the naphtha feed and recycle streams (F streams), and also the pressures of all product streams, leaving the tower (P) streams as fixed variables. In 
addition, the controllers of the catalytic reforming plant were the proportional-integral-differential (PID) type. The corresponding control constants are shown in Table 7. Figure 2 shows the simulation after these values are applied.