IBP1325_12 MAXIMISING PROFITABILITY THROUGH THE USE OF SIMULATION FOR NEW MARKET CRUDES EVALUATION Maryro P.

Mendez1

Copyright 2012, Brazilian Petroleum, Gas and Biofuels Institute - IBP

This Technical Paper was prepared for presentation at the Rio Oil & Gas Expo and Conference 2012, held between September, 1720, 2012, in Rio de Janeiro. This Technical Paper was selected for presentation by the Technical Committee of the event according to the information contained in the final paper submitted by the author(s). The organizers are not supposed to translate or correct the submitted papers. The material as it is presented, does not necessarily represent Brazilian Petroleum, Gas and Biofuels Institute opinion, or that of its Members or Representatives. Authors consent to the publication of this Technical Paper in the Rio Oil & Gas Expo and Conference 2012 Proceedings.

Abstract
There is nothing that contributes more to the cost of production for a refinery than the cost of crude oil and other feedstocks. The efficient and effective evaluation and selection of crude oil to a refinery will have a very significant impact on its profitability. The crude oil selected will impact the utilization and output from different process units, has the potential to impact maintenance and energy costs, and will impact product properties from the refinery. The tools and knowledge necessary for a good new crude evaluation program are addressed as part of this paper. The work process is outlined in detail. Finally, we address some of the risk management issues inherent in processing high margin crudes. Innovative simulation tools like Petro-SIM offer the platform to evaluate opportunities to capture the value that the sour/sweet crude differential and that the new crudes coming in to the market have to offer for a given refinery configuration. The paper provides comments on how KBCs methodology has helped refiners to check their ability to run new challenging crudes.

1. Introduction
Currently, one of the main themes around feedstock strategy worldwide is increasing flexibility to select the highest margin crude grades, heavy sour crudes being one of them. For the last couple of years, refiners have been evaluating a new market crude from Brazil, Peregrino. To determine its suitability as a refinery feedstock. The paper provides comments on how KBCs methodology has helped refiners to check their ability to run new challenging crudes like, Peregrino. In the case of Peregrino, no process data was available in the industry at that time and due to its quality, most refiners have been looking at different blends with diluting crudes as part of the evaluation process. As new crudes come in to the market refiners would like to be able to identify how the change in crude would impact refinery operations, product yields and identify potential constraints prior the test run. Process simulations provide significant data on optimum operating parameters for the refinery in order to achieve a successful test run. Innovative simulation tools like Petro-SIM together with the experience of KBC consultants offer the platform to evaluate opportunities to capture the value that the sour/sweet crude differential and that the new crudes coming in to the market have to offer for a given refinery configuration. This was key when evaluating the suitability and the economics of a crude where no processing experience was available.

2. Base Case Simulation

KBC has developed Petro-SIM software containing process unit models that can be linked together to produce a full refinery simulation model. For Crude and Vacuum Distillation Units, these models provide the capability for simulation, optimisation and design of distillation columns and associated fired heaters. Petro-SIM can be used to

______________________________ 1 Master of Engineering in Chemical Engineering, Senior Consultant - KBC ADVANCED TECHNOLOGIES, INC.

Rio Oil & Gas Expo and Conference 2012 represent both new and existing unit operations and will accurately predict intermediate and final product yields and qualities.
CADB Naphtha Hydrotreater Reformer

CDU Kero Hydro treater

Diesel Hydro treater Blend optimizer

Visbreaker FCC

HF Alky Unit

Figure. 1 Refinery Wide Simulation Petro-SIM interface For the refiner, KBC executes the base Petro-SIM case. This case identifies potential issues with the new crude as well as highlight areas where more detailed unit operation models will be required to properly identify issues and solutions. For example it may be necessary to convert some columns from short-cut methods to tray-to-tray models to identify flooding issues. Where such changes are necessary KBC implements them. In addition to the plant data collected and the column and reactor unit simulation, to allow understanding of the furnaces and transfer line performance KBCs Petro-SIM process simulation software has been used. The software contains a rigorous model for simulating the process side performance of fired heaters and transfer lines as part of the refinery wide flowsheet. The model performs detailed heat balance calculations and pressure drop calculations. The flow regimes in the furnace tubes and coking risk under the processing conditions are evaluated. Petro-SIM accurately simulates an existing operation since process models and even physical properties of feedstocks, intermediates, and products can be calibrated to actual lab data if assays do not accurately represent actual results. For the refiners evaluating the new crude, a refinery flowsheet including all process units was developed and this approach was used. Prior to the test run and after the base case is complete. The methodology for the subsequent runs is: Execute run and review results at high level to determine if there are significant issues associated with the case that could make it infeasible for further consideration. If the case looks reasonable at a high level, examine in more detail to identify process and mechanical issues that would need to be addressed. If the refiner decides to proceed with the analysis, then potential solutions to the issues will be identified. This was the worked process followed when evaluating Peregrino crude at a number of refineries to developed the processing strategy and risk mitigation plans.

3. New Crude to Market Simulation

3.1. Crude Assay characterization Validating the crude assay is the most important step in developing the economics and strategy for processing new crudes with existing assets. The most frequent response to process operations that do not perform according to current operation is that the crude quality changed. The KBC methodology to analyze and validate crude analyses is with assay databases and Petro-SIM analytical tools. Petro-SIM predicted yields from a complex refinery model where the crude slate changes considerably from the normal operation are only as accurate as the new crude assays. KBC crude assays are synthesized with Petro-SIM. During the process of generating user input files, assays missing lab results or questionable results for critical fractions and the whole crude analysis are identified and re-testing is requested. The first step in the validation process was to compare TBP curves of the proposed assays for consistency over the boiling range of the whole crude. The next step was to compare the following physical properties of assay cuts 2

Rio Oil & Gas Expo and Conference 2012 over the boiling range of the crude; molecular weight, density, sulfur, total nitrogen, nickel, vanadium, and conradson carbon. Assuming consistent TBP curves, assays were first evaluated for accuracy to predict gas oil contaminants and then distillate cold properties such as, kerosene freeze and diesel cloud point are reviewed.

Figure 2. TBP distillation plot generated by Petro-SIM Assays for each crude blend that met overall consistency criteria were first evaluated for accuracy in the gas oil boiling range and predictions of heavy contaminants. Criteria included Ni/Va distribution ratios, sulfur, nitrogen, and conradson carbon consistent with heavy sour crude oils. Some assays were rejected because the model would overpredict heavy vacuum gas oil Ni and Va concentrations, limiting vacuum tower cutpoints. Ni, Va, sulfur, nitrogen and concarbon contaminant curves were inspected in the HVGO range for accurate representation of operations. If crude assays met all of the above criteria, final assay selection was based on accuracy to predict distillate cold flow properties; kerosene freeze, diesel cloud and pour points. This is another area where incomplete analytical results eliminated assays which only report pour points for kerosene fractions and not freeze points.

Figure 3. Peregrino Assay Synthesized by Petro-SIM based on lab data An assay review would point out the viscosity, nitrogen, and bottoms yield as the initial concerns. Focus needs to be taken on the distribution to highlight any initial concerns before case work. In order to address some of the main concerns brought in this case by Peregrino, alternatives blends are evaluated where Peregrino is diluted with different crudes. The resulting blends are compared with blends where there is operating experience. The properties predicted by Petro-SIM oil Characterization for each crude, were used to estimate potential compatibility problems of each blend. 3

Rio Oil & Gas Expo and Conference 2012 Table 1. Feed properties predicted by KBCs Petro-SIM
CASTILLA PEREGRINO PEREGRINO BLEND BLEND 1 BLEND 2 28.7 0.8835 2.13 0.186 7.2 26.6 126.7 202.3 665.2 1269.2 1443.6 1585.6 24.9 0.9046 2.51 0.346 8.0 41.8 154.0 240.9 737.0 1352.6 1504.3 1617.2 26.7 0.8945 2.61 0.277 7.2 32.6 146.5 225.0 702.3 1316.3 1481.6 1607.9

Feed Properties
Crude Charge API Gravity SG Sulfur Nitrogen Concarbon TBP 1% 5% 10% 50% 90% 95% 99% F F F F F F F WT% WT% WT%

Base 30.1 0.8757 1.96 0.165 5.6 26.7 132.9 210.8 637.0 1231.3 1409.4 1574.9

A final further validation of the crude assay requires the crude to be processed in the refinery. Hence, as part of the new crude evaluation KBC has supported test runs and the Petro-SIM model was recalibrated. The procedure is to analyze the cuts and combine them with the yields and back calculate the crude feed. Knowing the crude recipe for the day and the individual crude assays a comparison of the back calculated to predict crude quality can be made. This allows determination if one of the assays is no longer representative of the actual crude being processed. 3.2. Prediction of properties and contaminants Simulation tools like Petro-SIM predict the changes in properties by separation and reactions that occur refinery wide and provide detailed component property curves.

Figure. 4. Nitrogen distribution Generated by KBCs Petro-SIM

Rio Oil & Gas Expo and Conference 2012

Figure.5. Generated by KBCs Petro-SIM The predicted properties and contaminant distribution and yield of the new crude allows the refiner to understand where the main constraints would be. For example for Peregrino it was clear that Nitrogen/Sulfur Content is much higher in the Kero range than the ones for the usual crudes normally processed at those facilities. Also, lower Smoke point/Cetane means more Aromatics. Hence Peregrino was not going to make Jet specs neat. This type of analysis was reached by doing a flowsheet review Also, the impact on catalyst life can be predicted, the higher amounts of sulfur and nitrogen translate into less catalyst life to meet similar targets for nitrogen and more hydrogen consumption. Table 2. Predicted properties for intermediate streams generated by refinery wide Petro-SIM flowsheet
Peregrino Blend 1 Peregrino Blend 2 Peregrino Blend 3 Peregrino Blend 4

HDS Feed Properties API Sulfur wt% Basic N2 ppm CCR wt% Refractive Index @20C Average BP F Product API Sulfur wt% Basic N2 ppm CCR wt% Refractive Index @20C Average BP F FCC High sulfur VGO feed Comb Feed Properties API Sulfur wt% Basic N2 ppm CCR wt% Refractive Index @20C Average BP F

Base case

16.628 2.548 1291 0.2428 1.5363 767 20.738 0.433 736 0.2428 1.51 749 10.085 19.92 0.6456 577 0.837 1.5147 801

16.584 2.467 1363 0.2217 1.5369 767 20.691 0.4195 777 0.2217 1.511 750 10.606 19.835 0.6364 629 0.8057 1.5155 802

16.519 2.3108 1528 0.173 1.5382 767 20.612 0.3928 871 0.173 1.512 750 11.643 19.713 0.6183 756 0.7297 1.517 801

16.7011 2.3945 1372 0.205 1.5368 766 20.8131 0.4071 782 0.205 1.5106 749 9.901 19.9205 0.6087 636 0.7965 1.5154 801

16.777 2.2199 1433 0.1784 1.5371 765 20.891 0.3774 817 0.1784 1.511 748 11.99 19.927 0.6015 763 0.4666 1.5168 799

3.3 Impact on Furnaces To allow understanding of the furnaces and transfer line performance KBCs Petro-SIM process simulation software has been used. The software contains a rigorous model for simulating the process side performance of fired heaters and transfer lines. The Furnace model simulates a general refinery furnace. The model performs detailed heat balance calculations and pressure drop calculations. The flow regimes in the Furnace tubes were checked and are as shown in the chart below. In the example below, KBC finds that the furnace tubes are seeing normal flow regimes (viz. spray or dispersed and bubble). 5

Rio Oil & Gas Expo and Conference 2012

Figure 6. Furnace Flow regimes generated by KBCs Petro-SIM The Oil film temperature is plot against the Oil Residence time in the furnace in both the modes in the chart below. The red line represents the limit of allowable range, and to its right is the coking zone. As can be seen, at a COT of 780 F and at a 58-60 TBPD rate, the vac furnace is operating very close to the coking zone. However this is not different the operation regime observed with the typical crude slate

Figure 7. Furnace residence time graph generated by KBCs Petro-SIM The furnace simulation allows the refiner to set optimum operation targets with the new crude while minimizing potential production loss due to coking and to understand potential operating costs of the new crude.

3. Challenges and Opportunities

Through simulation a operating challenges are identified for the specific refineries looking to process the new crude. At the same simulation work allows to develop and test operating strategies to address the identified challenges. 3.1 Peregrino Study Table 3. Challenges and Management WHOLE CRUDE KERO RANGE CHALLENGE Higher viscosity and pour point makes the crude logistically challenging Nitrogen/Sulfur Content is significantly high in Kero Range Lower Smoke point(15.7 mm)/Cetane means more Aromatics Would not make Jet specs neat MANAGEMENT Blending the crude with a less viscous stream/crude overcomes this hurdle The contaminants and the smoke point are blendable. By choosing the right crude blend the jet specs are met

Rio Oil & Gas Expo and Conference 2012 ULSD VGO Potentially Higher nitrogen will require more severe treating and increased H2 consumption Lower distillate yield high nitrogen content requires more severe treating and increased H2 consumption. High naphthenic acids in this cut but blend should be between the refinery limits High asphaltene content in resid Resid Viscosity extremely high Expect increased coker feed sales 6-oil spec on visc will be more difficult Nitrogen is higher but may not be a controlling specification when blended with the studied crudes Less ULSD yield volume may open up additional LCO material to be processed in the hydrotreaters By refinery wide simulation impact on FCC could be assess and take impact on life cycle taken into account. Concarbon and metal content of the blended Peregrino are predicted to be acceptable By doing a blending study with Petro-SIM cutterstock volumes required for 6-oil production or asphalt can be predicted

RESID

4. Black Oil Management

Heavy Sour crudes like Peregrino also come with high black oil yields, which is why; options for reducing black oil yields are underway in various design projects which involve the use of coking for residual destruction or other means. Capturing the Asphalt market may not be the first option refiners explore when trying to maximize the value of such crudes, however, a point of view can be adopted that sufficient coking or upgrading capacity has been implement in the last few years, which has removed a large part of the asphalt production capacity. Hence, with the use of simulation tools, Peregrino suitability as coker, visbreaker and asphalt feedstock has been included in the evaluation for the specific refineries.

5. Final Remarks

Most refiners are examining altering crude mix at existing refinery to capture the benefits of the emerging crudes and unconventional crudes Key concerns: Crude compatibility and handling feasibility Yield and property effects Maximizing economic facility utilization Solution: Utilize KBCs Petro-SIM to generate key properties for estimating compatibility Simulated in refinery-wide Petro-SIM model to account for yield, property, and utilization Develop ranking matrix Outcome: Highlighted several crude that meet overall processing and compatibility objectives Provided clear guidance on reasonable operating window for alternate crude blends Characterizing and tracking specific properties and contaminants are critical in modern refineries o Impacts revamp and new facility design o Drives selection and processing of crude slates in existing facilities