1 Active constraint regions for optimal operation of a simple LNG process Magnus G. Jacobsen and Sigurd Skogestad Department of Chemical Engineering NTNU.

Slides:



Advertisements
Similar presentations
A series of heat exchangers with each stage using a different refrigerant. Tailored to take advantage of different thermodynamic properties of the refrigerants.
Advertisements

1 Optimal Control of Chiller Condenser Sub-cooling, Compressor Speed, Tower Fan and Pump Speeds, and IGV Omer Qureshi, Hassan Javed & Peter Armstrong,
Exercise 3. Solutions From Fujio Kida, JGC Co. All cases
Dynamic modeling, optimization and control of a CO 2 stripper Marie Solvik Supervisors : Sigurd Skogestad and Marius Støre Govatsmark.
Refrigeration Cycles Chapter 11.
Self-Optimizing Control of the HDA Process Outline of the presentation –Process description. –Self-optimizing control procedure. –Self-optimizing control.
AIR CONDITIONING (COOLING) UNIT 40 TYPICAL OPERATING CONDITIONS
1 Single-cycle mixed-fluid LNG (PRICO) process Part I: Optimal design Sigurd Skogestad & Jørgen Bauck Jensen Quatar, January 2009.
EXERCISE 1 CHAPTER 11.
Carnot Thermodynamics Professor Lee Carkner Lecture 12.
Introduction: What is LNG? When natural gas is cooled to a temperature of approximate (–160 C) at atmospheric pressure it condenses to a liquid,called.
Throttling Thermodynamics Professor Lee Carkner Lecture 22.
Verbal Modeling Anchors: Manganese  Matt Morabito Jamie Polan.
Department of Mechanical Engineering ME 322 – Mechanical Engineering Thermodynamics Lecture 29 The Vapor Compression Refrigeration (VCR) Cycle.
Product Introduction.
1 M. Panahi ’Plantwide Control for Economically Optimal Operation of Chemical Plants’ Plantwide Control for Economically Optimal Operation of Chemical.
HOT COLD HOT COLD The Ideal Vapor-Compression Refrigeration Cycle 1-2 Isentropic compression in a compressor 2-3 Constant-pressure heat rejection in.
1 Operation of heat pump cycles Jørgen Bauck Jensen & Sigurd Skogestad Department of Chemical Engineering Norwegian University of Science and Technology.
1 Single-cycle mixed-fluid LNG (PRICO) process Part II: Optimal operation Sigurd Skogestad & Jørgen Bauck Jensen Quatar, January 2009.
Department of Mechanical Engineering ME 322 – Mechanical Engineering Thermodynamics Lecture 35 Analysis of Air Conditioning Processes.
Jack Hinze Advisors Professor Sanford Klein Professor Gregory Nellis
GHGT-8 Self-Optimizing and Control Structure Design for a CO 2 Capturing Plant Mehdi Panahi, Mehdi Karimi, Sigurd Skogestad, Magne Hillestad, Hallvard.
Vapor and Combined Power Cycles (2)
Process Operability Class Materials Copyright © Thomas Marlin 2013
Optimal operation of distillation columns and link to control Distillation Course Berlin Summer Sigurd Skogestad. Part 3.
1 Coordinator MPC for maximization of plant throughput Elvira Marie B. Aske* &, Stig Strand & and Sigurd Skogestad* * Department of Chemical Engineering,
Process Operability Class Materials Copyright © Thomas Marlin 2013
1 Modelling, Operation and Control of an LNG Plant Jens Strandberg & Sigurd Skogestad Department of Chemical Engineering, Norwegian University of Science.
Partial Evaporation Example 1/31/2007. Shell and Tube Heat Exchanger.
1 Outline Skogestad procedure for control structure design I Top Down Step S1: Define operational objective (cost) and constraints Step S2: Identify degrees.
1 EFCE Working Party on Fluid Separations, Bergen, May 2012 New results for divided-wall columns Deeptanshu Dwivedi (PhD Candidate, NTNU) Ivar Halvorsen.
PSE and PROCESS CONTROL
A Vapor Power Cycle Boiler T Turbine Compressor (pump) Heat exchanger
1 Single-cycle mixed-fluid LNG (PRICO) process Part II: Optimal operation Sigurd Skogestad & Jørgen Bauck Jensen Qatar, January 2009.
1 Self-Optimizing Control HDA case study S. Skogestad, May 2006 Thanks to Antonio Araújo.
1 Active constraint regions for optimal operation of chemical processes Magnus Glosli Jacobsen PhD defense presentation November 18th, 2011.
1 Single-cycle mixed-fluid LNG (PRICO) process Part I: Optimal design Sigurd Skogestad & Jørgen Bauck Jensen Qatar, January 2009.
1 A Plantwide Control Procedure Applied to the HDA Process Antonio Araújo and Sigurd Skogestad Department of Chemical Engineering Norwegian University.
1 E. S. Hori, Maximum Gain Rule Maximum Gain Rule for Selecting Controlled Variables Eduardo Shigueo Hori, Sigurd Skogestad Norwegian University of Science.
ChemE 260 Heat Pump Systems Gas Refrigeration Systems May 31, 2005 Dr. William Baratuci Senior Lecturer Chemical Engineering Department University of Washington.
Håkon Dahl-Olsen, Sridharakumar Narasimhan and Sigurd Skogestad Optimal output selection for batch processes.
1 Active constraint regions for economically optimal operation of distillation columns Sigurd Skogestad and Magnus G. Jacobsen Department of Chemical Engineering.
1 Using Self-Optimizing Control on the Statoil Mongstad HEN Daniel Greiner Edvardsen May 27, 2010 NTNU Confidential.
3) OBJECTIVE FUNCTION & DISTURBANCES Objective function: Assuming product prices are the same, p D = p S = p B and (p-p F ) = p’, with F given and Q =
ChemE 260 The Brayton Power Cycle and Variations
1 Optimization of LNG plants: Challenges and strategies Magnus G. Jacobsen Sigurd Skogestad ESCAPE-21, May 31, 2011 Porto Carras, Chalkidiki, Greece.
1 Self-optimizing control From key performance indicators to control of biological systems Sigurd Skogestad Department of Chemical Engineering Norwegian.
1 PLANTWIDE CONTROL Identifying and switching between active constraints regions Sigurd Skogestad and Magnus G. Jacobsen Department of Chemical Engineering.
1 Unconstrained degrees of freedom: C. Optimal measurement combination (Alstad, 2002) Basis: Want optimal value of c independent of disturbances ) – 
1 Self-optimizing control Theory. 2 Outline Skogestad procedure for control structure design I Top Down Step S1: Define operational objective (cost) and.
1 Control of maldistribution of flow in parallell heat exchangers Magnus G. Jacobsen, Sigurd Skogestad Nordic Process Controi workshop, Porsgrunn
1 Self-optimizing control From key performance indicators to control of biological systems Sigurd Skogestad Department of Chemical Engineering Norwegian.
1 Combination of Measurements as Controlled Variables for Self-optimizing Control Vidar Alstad † and Sigurd Skogestad Department of Chemical Engineering,
1 An Extended Pinch Analysis and Design Procedure utilizing Pressure Exergy for Subambient Cooling A. Aspelund, D. O. Berstad, T. Gundersen The Norwegian.
Vapour Compression Cycle You will Learn: 1 Vapour Compression Cycle Actual Vapour Compression Cycle Components in a Vapour Compression Plant Multistage.
1 Visit at Perstorp Daniel Greiner Edvardsen (Master student) Johannes Jäschke (PhD) April 14th, 2010.
EXERCISE 3 ANIS ATIKAH AHMAD. 1. Consider the distillation column shown in Figure Q1.1: Figure Q1.1: Distillation column It would be reasonable to control.
by Lars Erik Øi and Vladyslav Shchuchenko
air cycle cooling systems
Dr Xuesong Zheng – Process Integration Ltd (PIL)
TOPIC:- VAPOUR CYCLES CREATED BY:
ES 211: Thermodynamics Tutorial 5 & 6
Self-optimizing control Theory
Changing between Active Constraint Regions for Optimal Operation: Classical Advanced Control versus Model Predictive Control Adriana Reyes-Lúa, Cristina.
Step 2. Degree of freedom (DOF) analysis
Process Operability Class Materials Copyright © Thomas Marlin 2013
ChemE 260 The Carnot Cycle Dr. William Baratuci Senior Lecturer
Air Treatment School Refrigerated Dryer Training
“THERMODYNAMIC AND HEAT TRANSFER”
Presentation transcript:

1 Active constraint regions for optimal operation of a simple LNG process Magnus G. Jacobsen and Sigurd Skogestad Department of Chemical Engineering NTNU Trondheim, Norway 2nd Trondheim Gas Technology Conference November 3rd, 2011

2 Active constraint regions for a simple liquefaction process. OUTLINE: The PRICO process Modelling Definition of optimal operation –Minimize compressor work –Degrees of freedom –Constraints Optimal operation –Active constraint regions as a function of disturbances in feedrate and ambient temperature Control

3 Simple LNG liquefaction process with one multi-stream heat exchanger and a mixed refrigerant 30 bar Gas 4.5 bar heat 5.5 bar condenser CW Gas, 30C saturation subcooled Liquid Natural gas (NG): 90% C1, 5% C2, 2% C3, 0.1% C4, 3% N2 (mole-%) Mixed refrigerant (MR): 33% C1, 34%C2, 0% C3, 23% C4, 10% N2 The PRICO process

4 Temperature profile in main heat exchanger Cold MR “Hot” MR Natural gas

5 Modelling Unisim flowsheet simulator (Honewell) = Hysys (Aspen) SRK equation of state Data from Jensen and Skogestad (Adchem 2006) –Natural gas feed (1517 kmol/h, 40 bar, 30C): 90% C1,5% C2, 2% C3, 3%N2 –Mixed refrigerant (6930 kmol/h): 33% C1, 34%C2, 0% C3, 23% C4, 10% N2 –Cycle pressures: About 30 bar and 5 bar –Compressor speed = 1000 rms –Given compressor characteristics

6 Compressor map as function of speed (N) and MR flowrate Pressure ratio Efficiency

7 Key variables (nominal conditions) * *gProms ΔT SUP heat

8 Optimal operation Equipment fixed (UA-values, compressor map)! Optimization objective: Minimize compressor work (cost J = W s ) Five degrees of freedom for operation: 1.Amount of cooling in condenser (assumed at max) 2.Compressor speed 3.Turbine speed 4.Choke valve opening 5.Active charge (level in liquid receiver) WsWs

9 Constraints for operation 1.Refrigerant to compressor must be superheated, ΔT sup > 5°C 2.Compressor must not operate in surge, ΔM surge > 0 3.Maximum compressor work, W s ≤ 132MW 4.Turbine exit stream must be liquid only, ΔP sat > 0 5.Maximum compressor speed: ω comp ≤ ω max = 1200rpm + Exit temperature for natural gas, T NG,out ≤ -157°C (always active) + constraints on fully closed or open valves (cooling assumed at max) -> 5 CONSTRAINTS + REMAIN TO BE CONSIDERED + 3 UNCONSTRAINED DOFs

10 Optimal operation with compressor speed as DOF (nominal feedrate) 1 (of 5) active constraint

11 Want optimal operation also for disturbances NG feed flow rate [mol/s], +- 20% Ambient temperature [°C] – or actually cold inlet temperature to HEX for NG and MR (nominally 30C)

12 How do active constraints change? Have 3 DOFs and 5 constraints –ΔT sup,, W s, ΔP sat, ω comp 26 possible constraint combinations (of 3 or less) Only 5 combinations (”regions”) found in our simulations: I.ΔM surge is active II. ΔT sup and ΔM surge are active III. ΔT sup, ΔP sat and ω max are active IV. ΔT sup and ω max are active V. ω max is active

13 Constrains regions for PRICO process * Upper blue line: Cannot exceed W s,max since we minimize J=W s Lower green line: Too cold to get supersaturation > 5C… cool less or adjust MR composition W s > W s,max

14 Control of PRICO process 5 regions  5 control structures? ΔP sat is optimally close to zero in all regions –Keep at zero to simplify Surge margin and max speed: closely connected (switch) –Use compressor speed to control ΔM surge when ω opt < ω max We can use two control structures! –One for Regions I and V (where ΔT sup is inactive) –One for Regions II, III and IV (where ΔT sup is active)

15 Control structure DPC DPSAT saturation

16 Control structure DPC DPSAT saturation But: Constant level assumed used as self-optimizing variable when constraint on DTSUP is not active

17 Control structure DPC DPSAT saturation Optimal: DPSAT ~ 0 But: DPSAT >0 (and valve) may be needed to make sure we have no vapor in turbine turbine

18 Conclusions Optimization of PRICO LNG-process (Unisim/Hysys) Optimal operation (given UA-values + compressor) is often not same as optimal design Propose simple control structure to achieve optimal operation for disturbances in feed rate and temperature Still remains: Find self-optimizing variable when constraint on ΔT SUP is not active