Hydrocarbon Processing Copying and distributing are prohibited without permission of the publisher
Email a friend
  • Please enter a maximum of 5 recipients. Use ; to separate more than one email address.



Optimize desulfurization of gasoline via advanced process control techniques

10.01.2012  |  Yadav, V.,  Indian Oil Corp., Ltd., Mathura, Uttar Pradesh, IndiaDube, P.,  Indian Oil Corp., Ltd., Mathura, Uttar Pradesh, IndiaShah, H.,  Indian Oil Corp., Ltd., Mathura, Uttar Pradesh, IndiaDebnath, S.,  Indian Oil Corp., Ltd., Mathura, Uttar Pradesh, India

This case history describes the development of the inferential models used in open-loop and closed-loop applications, laboratory and analyzer update mechanisms, and APC model generation.

Keywords: [process control] [APC] [models] [analyzers] [predictive models] [inferential control] [PID] [inferential property prediction] [clean fuels]

At Indian Oil Corp.’s (IOC’s) Mathura refinery, a selective desulfurization unit was commissioned to reduce the sulfur content of fluidized catalytic cracked (FCC) gasoline—a blending component for finished motor spirit (MS). The objective of this new unit was lowering the sulfur content of FCC gasoline from 500 ppmw to 100 ppmw, thus meeting Euro IV product specifications for the refinery-gasoline blending pool. However, along with desulfurization, some undesirable olefin saturation reactions occurred, resulting in octane losses for the product gasoline. As per design, the octane loss in the desulfurization reactors is 1.3 units. With Euro IV specifications in place, the octane loss negatively impacted the refinery’s economics.

This refiner applied an advanced process control (APC) solution to minimize octane loss. The objective of the desulfurization unit’s APC program is to maximize sulfur content in the gasoline while still complying with Euro IV specifications and other process operating constraints. The control philosophy depended on sulfur estimations of the stabilizer-bottom product. An inferential property was developed for online estimation of the sulfur content, and it was used as a controlled variable in the multivariable predictive controller (MVPC).

This case history describes the development of the inferential models used in open-loop and closed-loop applications, laboratory and analyzer update mechanisms, and APC model generation. With APC, it was possible to increase the sulfur content in product gasoline by 10 ppm–12 ppm, along with an average octane gain of 0.11 units; all improved the refinery’s bottom line.

FCC GASOLINE DESULFURIZATION PROCESS

IOC’s Mathura refinery implemented a new gasoline desulfurization process. It is a two-step selective hydrotreating method. This processing unit consists of three major operations:

  • Selective hydrogenation unit (SHU)
  • FCC-gasoline splitter (FCCGS) unit
  • Hydrodesulfurization (HDS) unit.

In the first step, FCC gasoline is treated in the SHU, which selectively converts di-olefins into olefins and light mercaptans into heavier sulfur-containing compounds. In the second step, the SHU reactor effluent is separated into light-cut naphtha (LCN), heart-cut naphtha and heavy-cut naphtha (HCN) in the FCCGS unit. In the third step, the heavy fraction from the splitter bottom, containing high-sulfur content material, is processed in the HDS unit. This processing step converts heavy sulfur compounds into hydrogen sulfide (H2S). In addition, significant saturation of olefins occurs along with the HDS reactions. Saturating olefins reduces the final research octane number (RON) and is an undesirable condition.

ADVANCED APC OBJECTIVES AND DESIGN

In the Mathura refinery application, the control objectives are achieved by utilizing MVPC in conjunction with supporting predictions provided by an inferential property prediction package (IPPP). Supporting calculations are required to supplement existing process measurements. MVPC applications incorporate process models that permit forward-feed disturbance rejection and intermediate variables feedback, as well as constraint control. In configuring the controller, there is one main controller. The objectives for the main controller are:

  • Maximizing stabilizer-bottom product sulfur level within permissible limits, so that the upper limit of the total rundown sulfur for the desulfurization unit is maintained per Euro IV gasoline blending. Minimizing RON loss is also achieved.
  • Minimizing steam consumption by the stabilizer section
  • Maintaining safe unit operations.

To achieve these objectives, a main controller (MAINCON) and two sub-controllers are used:

  • Selective hydrogenation unit—SHUCON
  • Hydrodesulfurization unit—HDSCON.

Note: The stabilizer section of the HDS unit is considered part of HDSCON.

Sub-controller objectives

Before the APC installation, the SHU was operated to maintain stable flow to the reactor. Flow from the FCCU debutanizer (hot feed—70% of total) was routed to a feed-surge drum. A recycle stream (HDS stabilizer bottom stream) from a nitrogen-blanketed storage (cold feed—30% of total) was also sent to the feed-surge drum. Unit operators manually controlled the level of the SHU feed-surge drum by adjusting the recycle stream.

To maintain the SHU reactor inlet temperature, feed from the surge drum is heated by the SHU feed-effluent exchanger on the tube side by exchanging heat from the SHU effluent. The resulting mixture is heated in the SHU preheater using steam.

After the APC installation, the control objective was to keep steady flow to the SHU feed and maintain the surge-drum level by adjusting the FCCU debutanizer flow as a disturbance variable (DV) and adjusting the recycle stream. The control objective is to maintain a stable SHU RIT, by manipulating the effluent exchanger bypass flow and steam flow to the SHU pre-heater under allowable limits. The process equipment to be managed via the APC included:

  • SHU feed-surge drum (306-V-01)
  • SHU feed-effluent exchanger (306-E-01A/B)
  • SHU preheater (306-E-02).

Table 1 summarizes the sub-controller design for the selective hydrogenation unit. The SHUCON sub-controller was designed to manage steady flow to the SHU reactor while considering the debutanizer flow (hot feed) as a DV. The SHU feed/effluent exchanger bypass flow, along with steam to the SHU preheater, is used to control the SHU reactor-inlet temperature. Fig. 1 shows the same sub-controller (SHUCON) for the SHU.



 

  Fig. 1. Block diagram of the sub-controller for the selective
  hydrogenation unit.



HDS unit sub-controller

Before the APC implementation, the HDS unit was operated by controlling the severity conditions of the reactors. The unit operator controlled HDS reaction (first-bed inlet temperature and second-bed inlet temperature) based on daily sulfur levels in the stabilizer bottom product and rundown product. Sulfur levels were determined by analyzers and lab testing. The fuel gas was cascaded with first-bed inlet temperature, and the quench flow was cascaded with second-bed inlet temperature. To maintain stable reflux flow to the stabilizer, unit operators adjusted the stabilizer reboiler temperature and reflux pressure by continuous monitoring of the light-end flow to the column.

Post-APC operations

The HDS reactor is set by the APC based on sulfur levels of the stabilizer bottoms. The IPPP estimation is done on a 15-second basis. Also, the APC will maximize the sulfur level within given operator limits, thereby by adjusting the reactor severity. The stabilizer-bottom reboiler temperature is controlled by APC and facilities minimizing the steam consumption by the reboiler. However, the reflux flow to the stabilizer is also controlled by APC, along with stabilizer-bottom re-boiler temperature. The process equipment managed via APC includes:

  • HDS reactor (307-R-01)
  • HDS heater (307-F-01)
  • HDS feed-effluent exchanger (307-E-01 A/B/C/D)
  • Stabilizer section (307-C-02).

Table 2 summarizes the sub-controller design for the HDS unit. Fig. 2 shows the same sub-controller (HDSCON) for the HDS unit.



 

  Fig. 2. Block diagram of the sub-controller for the
  hydrodesulfurization unit.



Models

As shown in Fig. 3, the simple first-order process models were not providing tight control on the HDS reactor-inlet temperatures. In response, a ramp transfer function block was added into the model, along with the first-order transfer function block. The exothermic reaction in the reactor behaves in a “ramp” manner (unbounded runaway even in the case of a bounded input disturbance). Due to “ramp” behavior of the process, fast action is required in manipulated variables (MVs), such as fuel-gas flow and quench flow, to quickly control the exotherm (by controlling the first-bed and second-bed inlet temperatures) before they rise too high. The inherent instability of the reactor was countered via a ramp block, plus the normal first-order block, to relate the MVs and DVs with the inlet temperatures. For a step change in DVs, this combination predicts an unbounded response in the inlet temperatures—thus, moving the MVs quickly to reject the disturbance.

 

  Fig. 3. First-order process model response
  to reactor inlet temperature control.



SUPPORTING CALCULATIONS IPPP DEVELOPMENT

To calculate the sulfur content of the FCC feed inlet, several predicted values were considered. By using the flowrate and sulfur quantity of all streams listed in Table 3, the total sulfur value can be calculated at the FCCU feed inlet. The calculation used to estimate the sulfur content is:

=(79FC803.PV 3 S1 3 density) + (79FC802.PV 3 S2 3 density) + (79FC801.PV 3 S3 3 density) + (7FC6701.PV 3 S4 3 density) + (12FIC100.PV 3 S6 (if crude_select.op=1) 3 density)

or (MRA.12FIC100.PV 3 S7 (if crude_select.op = 2) 3 density)

or (MRA.12FIC100.PV 3 S8 (if crude_select.op = 3) 3 density) + ((2FC0708.PV 3 S5)/1000) / (79FC803.PV + 79FC802.PV + 79FC801.PV + 7FC6701.PV + 12FIC100.PV + 2FC0708.PV)

where S1–S8 are sulfur values that are entered by the operator.



Sulfur content of FCC gasoline splitter

Feed to FCCGSU is compensated by two streams—hot feed from the debutanizer (306FI0105.PV) and cold feed from recycle (306FIC0101.PV). Calculations to estimate sulfur at FCCGSU feed are:

= ((DSU_SULFUR.PV 3 306FI0105) + (STABBTM_SULFUR 3 306FIC0101.PV-5.5*)) / {(306FI0105)
+ (306FIC0101-5.5*)}

where DSU_SULFUR.PV and STABBTM_SULFUR are the IPPP sulfur estimations.

*5.5 is the flow correction since the control valve has a zero error.

IPPP applications

Several IPPP models were developed for the FCC gasoline desulfurization unit and include:

  • FCCDSU hot feed sulfur estimation
  • HDS feed sulfur estimation
  • Stabilizer bottom sulfur estimation.

FCCDSU feed sulfur

This model used several inputs:

Tag name Tag description
FCCUFD_SULFUR.PV Sulfur at FCCU (calculation)
19TRC153.PV FCCU main fractionator top temperature
20TI99.PV FCCU debutanizer bottom temperature.

To estimate the sulfur content of DSU feed, the following linear equation is used:

P = Ax1 + Bx2 + Cx3 + Bias

where: P = DSU_SULFUR.PV
(FCCDSU feed sulfur in hot feed)
A = Coefficient 0.041417
x1 = FCCUFD_SULFUR.PV
B = Coefficient 1.6497
x2 = 19TRC153.PV
C = Coefficient 5.736500
x3 = 20TI99.PV
Bias = –1067.4

HDS feed sulfur

This model used several inputs:

Process inputs used
Tag name Tag description
GSUFD_SULFUR.PV Feed to FCCGSU (calculation)
20PI0802.PV FCCGSU top pressure
20FC0306.PV FCCGSU light cut draw flow
20FC0404.PV FCCGSU heart cut draw flow

The following linear equation is used:

P = Ax1 + Bx2 + Cx3 + Dx4 + Bias

where: P = HDSFD_SULFUR.PV
A = Coefficient 1.097890
x1 = GSUFD_SULFUR.PV
B = Coefficient –272.28299
x2 = 20PI0802.PV
C = Coefficient 7.0273
x3 = 20FC0306.PV
D = Coefficient 3.291770
x4 = 20FC0404.PV
Bias = 598.81

Stabilizer-bottom sulfur

This model used several inputs:

Process inputs used
Tag name Tag description
HDSFD_SULFUR.PV HCN sulfur (HDS feed sulfur IPPP estimation)
307TI0642.PV HDS reactor 1st bed inlet temperature
307TI0630.PV HDS reactor 2nd bed bottom temperature
307TI1014.PV Stabilizer bottom temperature.

To estimate the sulfur content of HDS feed, the following linear equation is used:

P = Ax1 + Bx2 + Cx3 + Dx4 + Bias

where: P = STABBTM_SULFUR.PV
A = Coefficient 0.115679
x1 = HDSFD_SULFUR.PV
B = Coefficient –3.90
x2 = 307TI0642.PV
C = Coefficient –3.59673
x3 = 307TI0630.PV
D = Coefficient –0.341067
x4 = 307TI1014.PV
Bias = 1067.5

From Fig. 4, the quality estimation using the IPPP has good agreement with the actual sulfur content as measured from unit and lab analyzers. Table 4 summarizes the economic functions and RON improvement possible with APC.

 

  Fig. 4. Quality and process improvement
  achieved through APC IPPP.




PROJECT MILESTONES

Implementing APC on the HDS unit has yielded substantial tangible and intangible benefits. While the annual monetary gain is of the order of Rs. 39 lakhs, significant improvement via process control and optimization was achieved as measured through tighter control of the SHU and HDS reactor inlet temperatures. More accurate estimation of the stabilizer-bottom sulfur inferential was possible, which facilitated proper control action via the APC. With tighter control and action via APC, adjusting and preferentially lowering the reactor-inlet temperatures were possible. The effect of crude changes in the atmospheric and vacuum distillation unit is also incorporated into the model. The resultant sulfur changes in the FCC feed are transmitted via means of intermediate calculations and inferential estimations to the final stabilizer-bottom sulfur prediction. Operators now have more confidence when implementing control and optimization strategies. This has resulted in better operations of the refinery. Accordingly, APC was successfully implemented and is yielding expected benefits. HP

LITERATURE CITED

1 Perry, R. H., Chemical Engineers Handbook, Sixth Ed., New York, McGraw Hill, 1984.
2 Levenspiel, O., Chemical Reaction Engineering, Third Ed., Singapore, John Wiley and Sons, 1999.
3 Stephanopoulos, G., Chemical Process Control, Dorling Kindersley (India) Pvt. Ltd., 2007.

The authors

Shyamal Debnath is the chief technical services manager at Indian Oil Corp. (IOC) Ltd.’s Mathura refinery. His primarily responsibilities include providing technical services for strategic initiatives and advanced process control (APC). Mr. Debnath has more than 25 years of experience in unit operations, strategic initiatives (process and projects), research, troubleshooting and APC for all the major process units at various IOC refineries. He holds an MS degree in chemical engineering from Indian Institute of Technology, Kharagpur, India.

Hitesh Shah is a senior technical services manager with Indian Oil Corp. (IOC) Ltd.’s Mathura Refinery. His primary responsibilities include providing technical services for strategic initiatives and APC. Mr. Shah has more than 14 years of experience in strategic initiatives, planning and coordination, and APC. At present, he is working as a senior technical services manager at IOC’s Gujarat refinery. Mr. Shah holds an MS degree in chemical engineering from Indian Institute of Technology, Bombay, India.

Prashat Dube is a senior process engineer at Indian Oil Corp. (IOC) Ltd.’s Mathura Refinery. He is primarily responsible for providing technical services for APC implementation and maintenance. Mr. Dube has five years of experience in APC for all major process units at the Mathura Refinery and holds a BS degree in chemical engineering from Indian Institute of Technology, New Delhi, India.


Ms. Varsha Yadav is a senior process engineer at Indian Oil Corp. (IOC) Ltd.’s Mathura refinery. She is primarily responsible for providing technical services for APC implementation and maintenance. Ms. Yadav has three years of experience in APC for all major process units at the Mathura Refinery and holds a BS degree in chemical engineering from Regional Institute of Technology, Raipur, India.

 



Have your say
  • All comments are subject to editorial review.
    All fields are compulsory.

B.Barati
12.24.2013

Hi
Dear Sir or Madam,
I would be thankful if you reply me. Can you cooperate for desulfurization process with industrial company in Iran?

Related articles

FEATURED EVENT

GasPro North America

Sign-up for the Free Daily HP Enewsletter!

Boxscore Database

A searchable database of project activity in the global hydrocarbon processing industry

Poll

Should the US allow exports of crude oil? (At present, US companies can export refined products derived from crude but not the raw crude itself.)


66%

34%




View previous results

Popular Searches

Please read our Term and Conditions and Privacy Policy before using the site. All material subject to strictly enforced copyright laws.
© 2014 Hydrocarbon Processing. © 2014 Gulf Publishing Company.