Upload
brenno-menezes
View
353
Download
6
Embed Size (px)
Citation preview
Brenno C. Menezes
Postdoctoral Fellow
Technological Research Institute
São Paulo, SP, Brazil
Jeffrey D. Kelly
CTO and Co-Founder
IndustrIALgorithms
Toronto, ON, Canada
Ignacio E. Grossmann
R. R. Dean Professor
Carnegie Mellon University
Pittsburgh, PA, US
Lincoln F. L. Moro
Senior Consultant
PETROBRAS
São Paulo, SP, Brazil
October 6th, New Orleans, LA, United States
AI L
Page 2 2015 [Q&A and Technology Forum]PD -15-08
Objectives
• To improve efficiency, effectiveness and
economy of mixing/blending, reacting/converting
and separating/fractionating inside the oil-
refinery.
• To integrate blending of several streams’
distillation curves with also shifting or adjusting
cutpoints of distilled streams (i.e., initial and/or
final boiling-points, IBP and FBP) in order to
manipulate their TBP curves in an either off- or
on-line environment (Kelly et al, 2014).
Page 3 2015 [Q&A and Technology Forum]PD -15-08
Perfect/Ideal/Sharp
Separation
Distillation Curves
From Other
Units
From CDU
Kerosene
Light Diesel
ATR
C1C2
C3C4
N
K
LD
HD
Naphtha
Heavy DieselCrude
CDU
Temperature
Yie
ld (
%)
N
K
LD
HD
N K LD HD ATRFeed
Page 4 2015 [Q&A and Technology Forum]PD -15-08
• Li, Hui, Li, A. Integrating CDU, FCC and Blending
Models into Refinery Planning. Comput. Chem.
Eng. 2005.
• Alattas, Grossmann, Palou-Rivera, Integration of
Nonlinear Crude Distillation Unit Models in Refinery
Planning Optimization. Ind. Eng. Chem. Res. 2011.
• Mahalec, Sanchez, Inferential Monitoring and
Optimization of Crude Separation Units via Hybrid
Models. Comput. Chem. Eng. 2012.
• Ali, Yusoff, Determination of Optimal Cut Point
Temperatures at Crude Distillation Unit using
Taguchi Method, Int. J. Eng. & Tech., 2012.
Page 5 2015 [Q&A and Technology Forum]PD -15-08
• Cutpoint temperatures (IBP and FBP) are strong
functions of both the light and heavy draw flows
and temperatures for the distillates leaving the
column, tower or side-strippers.
• Other causal factors or process variables
affecting cutpoint temperatures include draw tray
technology, location, efficiency and pressure,
pump-around flows (intermediate refluxes), duty
and return temperature as well as stripping steam
flow which all influence and contribute to the
column’s internal vapor and liquid traffic defining its
separation/fractionation operation.
Page 6 2015 [Q&A and Technology Forum]PD -15-08
• Four (4) principles for Nelson (1945) (cf. Kaes,
2009):
‒The yield of a given product or fraction is
primarily a function of the composition of
the feed mixture, not the degree of
separation.
‒The number of trays only slightly alters the
boiling range of the products as defined by
ASTM initial and final boiling points.
Page 7 2015 [Q&A and Technology Forum]PD -15-08
• Four (4) principles for Nelson (1945) (cf. Kaes,
2009):
‒The initial boiling point of side draw products is
always low, and must be corrected by either
steam stripping or reprocessing (refluxing).
‒The final boiling point of a side draw
product is primarily controlled by opening
or closing the draw valve to change the
yield.
Page 8 2015 [Q&A and Technology Forum]PD -15-08
T01 T05 T10 T30 T50 T70 T90 T95 T99
Temperature
Yie
ld (
%) Back-end:
Front-end:
New Temperature: NT
Old Temperature: OTNew Yield: YNT
Page 9 2015 [Q&A and Technology Forum]PD -15-08
Page 10 2015 [Q&A and Technology Forum]PD -15-08
Yie
ld (
%) Back-end:
Front-end:
Temperature (⁰F)
T99: 230→214
T01: 91→85
Page 11 2015 [Q&A and Technology Forum]PD -15-08
Yie
ld (
%) Back-end:
Front-end:
95.87%
-1.45%
Temperature (⁰F)
T99: 230→214
T01: 91→85
Page 12 2015 [Q&A and Technology Forum]PD -15-08
Yie
ld (
%)
Temperature (⁰F)
Difference in Yield: DYNT DYNT99
DYNT01
Page 13 2015 [Q&A and Technology Forum]PD -15-08
DBCTO
• Conservative and adaptive TBP cutpoint method
to marginally/incrementally adjust/shift distillate
yields when “non-sharp” (non-ideal, imperfect)
separation exists i.e., without an “infinite” number
of stages and/or reflux.
• Uses ASTM D86 cutpoint temperatures that are
not as suitable as TBP temperatures given that
they over-predict the initial boiling-point (IBP) and
under-predict the final boiling-point (FBP) – but,
they are quicker to measure in the lab and field!
Page 14 2015 [Q&A and Technology Forum]PD -15-08
DBCTO
From Other
Units
From CDU
Temperature (⁰F)
Yie
ld (
%)
ASTM D86
TBP
Inter-conversion
Evaporation
Curves
Interpolation
Ideal Blending
Evaporation
Curve
Multiple
Components
Final
Product
ASTM D86
Interpolation
Inter-conversion
TBP
Page 15 2015 [Q&A and Technology Forum]PD -15-08
Cutpoints.+Process Variables
• Step1: Collect and Convert ASTM D86
laboratory or field analyzer historical data to TBP
temperatures using API TDB 3A1.1.
• Step2: Collate process variable data properly
synchronized with ASTM D86 laboratory or field
data.
• Step3: Regress selected TBP cutpoint
temperature (typically 90%, 95% or 99%) with
causal process variables previously mentioned
such as draw flow, temperature, pressure, steam,
pump-around duty, etc.
Page 16 2015 [Q&A and Technology Forum]PD -15-08
Curves Interpolation
What if I need T85%?
Use the interpolated
curve to extract any point
Page 17 2015 [Q&A and Technology Forum]PD -15-08
Example
• Maximize flow of DC1 and DC2 ($0.9 for DC1 and
$1.0 DC2) with lower and upper bounds of 0.0 and
100.0 each (for fixed DC3 and DC4 at 1). The ASTM
D86 specifications for their blend are D10 ≤ 470, 540
≤ D90 ≤ 630 and D99 ≤ 680. Only DC1 is an CDU stream.
DC1 DC2 DC3 DC4
1% 305.2 (353) 322.2 (367) 327.0 (385) 302.4 (368)
10% 432.9 (466) 447.1 (476) 405.2 (435) 369.7 (407)
30% 521.6 (523) 507.1 (509) 457.1 (462) 441.0 (449)
50% 565.3 (551) 549.5 (536) 503.3 (492) 513.8 (502)
70% 606.4 (581) 598.4 (573) 551.1 (528) 574.3 (550)
90% 668.3 (635) 666.1 (634) 605.8 (574) 625.4 (592)
99% 715.7 (672) 757.7 (689) 647.0 (608) 655.2 (620)
Table. Inter-Converted TBP (ASTM D86) Temperatures in Degrees F.
Page 18 2015 [Q&A and Technology Forum]PD -15-08
Example
Figure. ASTM D86 distillation curves, including the final blend, which
is determined by the blended TBP interconversion to ASTM D86.
The new and
optimized TBP curve
for DC1 given its
front and back-end
shifts is now:
[(1.053%,312.8),
(10.015%,432.9),
(31.188%,521.6),
(52.361%,565.3),
(73.534%,606.4),
(94.707%,668.3),
(98.995%,689.3)]
Yie
ld (
%)
Temperature
For DC3 and DC4 fixed at 1,
DC1 takes a new profile to match D90 of
the blend.
Page 19 2015 [Q&A and Technology Forum]PD -15-08
CDU and Blender integration
DBCTO in Scheduling Oper.
FCC
DHT
NHT
KHT
REF
DC
B
L
E
N
SRFCC
Fuel gas
LPG
Naphtha
Gasoline
Kerosene
Diesel
Diluent
Fuel oil
Asphalt
Crude-Oil
ManagementCrude-to-Fuel Transformation Fuel Blending
and Distribution
VDU
Menezes et al. (2015)
CDU and Blender integration
Page 20 2015 [Q&A and Technology Forum]PD -15-08
Summary
• Use TBP temperatures instead of ASTM D86.
• Fit a regression model i.e., MLR, PCR, PLS or DR
(Dynamic Regression), of how causal process
variables affect the selected TBP cutpoint
temperature (at 90%, 95% or 99%) and not the
distillate yield explicitly.
Page 21 2015 [Q&A and Technology Forum]PD -15-08
Novelty
Page 22 2015 [Q&A and Technology Forum]PD -15-08
Impact in the Industry
• Integration of the distillation unit and fuel blending
using IBP and FBP (cutpoints) of the streams.
• Based on daily experimental data run gathered in the
field (ASTM D86) instead of blending indices
correlations.
• Reduce losses by off-spec fuels.
• Reduce the blender RTO efforts to get on-spec fuels.