Steam Reformer Surveys - Techniques for Optimization of Primary Reformer Operation
37 likes6,328 views
The document discusses optimizing the operation of primary reformers through surveys that assess catalyst performance, reformer operation, and tube life estimation. It includes various techniques such as tube wall temperature measurement, process data collection, and computer modeling to identify issues and suggest improvements, with case studies demonstrating significant cost savings. The methodology has been employed in over 30 plants, emphasizing the importance of accurate data collection and analysis for effective reformer management.
![Reformer Surveys
TWT Survey (Optical Pyrometer)
•Correct to minimize background radiation
effects
•Use a Stefan-Bolzman Equation
Tt = {(Tm
4 - [1 -e] Tw
4)/e}0.25
• Tt : True temperature
• Tm : Measured temperature
• Tw : Background temperature
• e : emissivity](https://image.slidesharecdn.com/steamreformersurveys-130923165519-phpapp02/85/Steam-Reformer-Surveys-Techniques-for-Optimization-of-Primary-Reformer-Operation-12-320.jpg)
Steam Reformer Surveys - Techniques for Optimization of Primary Reformer Operation
- 1. Steam Reformer Surveys Gerard B. Hawkins Managing Director Techniques for Optimization of Primary Reformer Operation
- 2. Introduction Background Radiation and Temperature Measurement Reformer Survey Inputs Case Study 1 Case Study 2 Case Study 3 Conclusions
- 3. Reformer is at the heart of the plant ◦ Converts feed gas to Syngas ◦ Complex operation ◦ Integrated design ◦ Main energy consumer ◦ Most expensive single plant item Reformer is often a throughput constraint
- 4. Combination of techniques used Tube Wall Temperature measurement Plant heat & mass reconciliation Reformer simulations Output provides assessment of Catalyst performance Reformer operation Operating limits Tube life estimation
- 5. Introduction Background Radiation and Temperature Measurement Reformer Survey Inputs Case Study 1 Case Study 2 Case Study 3 Conclusions
- 6. Tube Wall
- 7. Background radiation affects readings Minimize errors when using IR pyrometer ◦ Use emissivity setting of 1.00 ◦ Use correction formula Post processing calculation Use Gold Cup pyrometer
- 8. Reformer Surveys TWT Survey (Optical Pyrometer) Gold Cup: •Most accurate temperature measurement •Eliminates the effects of background radiation •Limited number of tubes can be measured •Large cumbersome equipment •Significantly more readings on side fired furnaces
- 9. Reformer Surveys TWT Survey (Optical Pyrometer) Optical Pyrometer: •Good for taking 'lots' of readings •Most tubes are visible •Easy to use •Portable •Absolute figures not accurate •Relative figures are more accurate
- 10. Reformer Surveys TWT Survey (Optical Pyrometer) •Measures total radiation from target •Picks up radiation from •refractory •flue gas •other tubes •Can not distinguish between •radiation emitted and radiation reflected •Measured temperature is high •Typically 68-104°F (20-40°C)
- 11. Reformer Surveys TWT Survey (Optical Pyrometer) •Cyclops 52/153 has narrow bandwidth •0.8-1.1 micron •Reduces radiation from flue gas effect •Ensure that reading taken at 90° to tubes •Both vertically and horizontally •It is possible to correct for these radiation effects •Temperature to Fourth Power •Lots of data should eliminate random errors
- 12. Reformer Surveys TWT Survey (Optical Pyrometer) •Correct to minimize background radiation effects •Use a Stefan-Bolzman Equation Tt = {(Tm 4 - [1 -e] Tw 4)/e}0.25 • Tt : True temperature • Tm : Measured temperature • Tw : Background temperature • e : emissivity
- 13. Reformer Surveys TWT Survey (Optical Pyrometer) •Must correct measured temperatures •For background readings use temperatures from: •Refractory (walls, floor and roof) •Use following expression Tw = {1/N *( TW1 4 + TW2 4+ TW3 4 .…+ TWN 4)}0.25 •N is number of readings
- 14. Reformer Surveys TWT Survey (Optical Pyrometer) •Pyrometer used with an emissivity of 1 •Emissivity of 0.85 used in correction •Plant data reconciled and furnace modelled in ASPEN HYSYS V8 •Corrected temperature compared to simulated values
- 16. Introduction Background Radiation and Temperature Measurement Reformer Survey Inputs Case Study 1 Case Study 2 Case Study 3 Conclusions
- 17. Tube wall temperature survey ◦ Tube temperatures ◦ Background temperatures Process operating data collection ◦ Pressure, Temperature, Flows Chemical analysis of all streams Radiant and convection section data ◦ Geometry, Layouts …
- 18. VULCAN CERES - data fitting package used to reconcile data The use VULCAN REFSIM to model furnace Close H&M balance on process and flue gas using Aspen HYSYS V8 Allow certain values to float Wider data envelope = better fit
- 19. VULCAN REFSIM - fully coupled computer model ◦ Radiant heat transfer in flue gas ◦ Heat transfer inside tubes ◦ Reaction kinetics inside tubes Radiation based on proven theory Tubeside based on operating plant data
- 20. Heat Flows Radiation Convection Tube Fluegas Flame Wall
- 21. 680 700 720 740 760 780 800 820 840 860 0 0.2 0.4 0.6 0.8 1 1.2 Fractional Distance Down Tube Temperature(°C) Simulation Measured
- 22. Fire Extinguisher ◦ Inject via side peepholes or burner ignition port ◦ Check for flue gas maldistribution ◦ See case study 1 ◦ Can use K2CO3 Fuel gas pressures ◦ Check for fuel mal-distribution ◦ Use standard pressure gauge
- 23. Combustion air pressure ◦ Use standard manometer ◦ Check by row and then by burner Visual Inspection ◦ Look at tubes, refractory and burners ◦ Check for deviations from expectation Design Philosophy ◦ Check for deviations from expectation
- 24. Check wind box pressure ◦ Ensure even firing through out furnace Check oxygen levels ◦ Ensure even combustion air flow Thermal Imaging ◦ Check for refractory damage
- 25. Reformer Surveys Summary A Reformer Survey involves: •Collection and analysis of data from both the process and flue gas sides •Assess the performance of the reformer •Assess the performance of the catalyst Collecting data from the whole reformer minimizes errors.
- 26. Reformer Surveys Summary Typical outputs from a Reformer Survey includes: • Catalyst performance • Real tube skin temperature • Reformer balance • Efficiency gains • Benchmarking
- 27. Reformer Surveys Content • Introduction • Safety • Preparation • Onsite Data Collection • TWT Survey • Observation/Troubleshooting • Modelling and Analysis • Results/Outputs • Case Studies • Conclusions
- 28. Reformer Surveys Introduction • Primary is the most complicated and expensive piece of equipment on the plant •Heat transfer - Provides sensible heat and heat of reaction •High pressure and very high temperature •Data collection can highlight trends •Reformer survey required to allow full diagnosis
- 29. Main additional risks are burns and overheating, ◦ Burns from exposed hot surfaces ◦ Radiation burns via open peepholes ◦ Burns due to hot gas or flames ◦ Heat stroke/Dehydration
- 30. In addition to standard PPE the following should be considered, ◦ Heat resistant gloves ◦ Flame retardant overalls ◦ Furnace eye protection
- 31. Reformer Surveys Typical Work Remit Typically a reformer survey consists of a number of actions: •Preparation •On-site data collection •Tube wall temperature measurement •Observations and trouble shooting •Modelling and analysis •Report writing
- 32. Reformer Surveys Preparation Usually carried out prior to site visit and would normally include: • A wish list of requirements from the plant • Mechanical design of the reformer • Piping and instrument drawings • Process flow diagrams • Any known process problems
- 33. Reformer Surveys On-site Data Collected •Feed, Steam, Fuel, Combustion air data including, •Flows •Pressures •Temperatures •Gas analysis from on line analyzers & laboratories •Reformer dimensions •Tube temperatures using an optical pyrometer (or gold cup)
- 34. Reformer Surveys Tube Wall Temperature Survey •Tube skin temperature used to fit temperature profile •Generates an activity figure •No one ideal method of measurement •Two methods currently used •Optical Pyrometer •Gold Cup •Both have advantages and disadvantages
- 35. Introduction Background Radiation and Temperature Measurement Reformer Survey Inputs Case Study 1 Case Study 2 Case Study 3 Conclusions
- 36. • Large scale ammonia plant • Tube temperatures split in box • No apparent process reason Hot Zones Cold Zones
- 37. • Eliminated other possibilities • Maldistribution due to • Process gas • Fuel gas • Firing • Only left with combustion air • Subsequent shut down • Found one of the two air dampers stuck • Repaired
- 38. • After shut down temperatures were
- 39. Survey highlighted an problem on the furnace By working closely with plant personnel, determined root cause Subsequent work proved root cause Problem worth US$750,000 per year
- 40. Customer complained of high ATE Survey found ◦ High box pressure (-2 or -3 mm H2O) ◦ Afterburning in centre of furnace but O2 levels exit box in excess of 2.5 % ◦ Cool outer rows ◦ Hot centre rows
- 42. Design of combustion air duct was symmetrical Combustion air and flue gas fans at limit Insufficient driving force to get air to centre of furnace Cause after burning
- 43. • Survey on plant found odd temperature distribution • Not explained by burner pressure • Not explained by combustion air mal- distribution 10 18 26 34 42 50 58 66 2 3 4 5 860 880 900 920 940 860 880 900 920 940 Temperature 940+ 932 to 940 924 to 932 916 to 924 908 to 916 900 to 908 892 to 900 884 to 892 876 to 884 868 to 876 860 to 868 Row Number Tube Number
- 44. Checks on furnace geometry highlighted an issue ◦ Outer lanes were the same size as the inner lanes ◦ Outer row of burners were rated at 70% of the inner burners Injected dry powder from fire extinguisher into furnace ◦ Unusual flow patterns
- 47. Computational Fluid Dynamics was used to model reformer in detail Burners Tunnel Ports Velocity Vectors
- 48. A B C D E F Tube Number
- 49. CFD simulations matched the observations from the plant ◦ Dry powder tests and TWT measurements Three proposed solutions to eliminate the effect ◦ Increase burner size to match tunnel size ◦ Decrease furnace width to match burner size ◦ Increase velocity through the burners
- 50. 70% 100% burner burner 100% 100% burner burner 70% 100% burner burner Recirculating Case Solution 1 Solution 2 100% 2.1 m 100% 2.1 m 100% 2.1 m 100% 2.1 m 70% 1.5 m 100% 2.1 m
- 51. Solution 1 - Requires 100% burner in outside rows ◦ Difficult to achieve ◦ Requires either Modification of burners Replace with 100% burners ◦ But too much heat flux ◦ Must increase process gas flow ◦ Install orifice plates inlet all tubes ◦ Outer rows are larger than inner
- 52. Solution 2 is to reduce furnace width so outer lane width matches the 70% burners ◦ Requires modification to refractory ◦ Increase in number of ports on the outer rows of tunnels Solution 3 - Increase velocity through outer row of burners ◦ 154% of existing velocity
- 53. Highlighted a mal-distribution Costing plant approximately US$350,000 in lost production Reduce peak tube temperatures Methodology proved initial theory Allowed for a set of solutions to be proposed
- 54. Visual Inspection ◦ Look at tubes, refractory and burners ◦ Inspect external casing ◦ Check for deviations from expectation Design Philosophy ◦ Check for deviations from expectation Fuel gas pressures ◦ Check for fuel mal-distribution ◦ Use standard pressure gauge
- 55. Combustion air pressure ◦ Use standard manometer ◦ Check by row and then by burner Fire Extinguisher ◦ Inject via side peepholes or burner ignition port ◦ Check for flue gas maldistribution ◦ See case study 3 ◦ Can use K2CO3
- 56. Check wind box pressure ◦ Ensure even firing through out furnace Check oxygen levels ◦ Ensure complete combustion ◦ Ensure even combustion air flow Thermal Imaging ◦ Check for refractory damage
- 57. Reformer Surveys Modelling and Analysis Computer packages used: • VULCAN REFSIM • Heat and Mass Transfer in radiant box •Aspen HYSYS • Flowsheeting package • VULCAN TP3 or VULCAN CERES •Match data between models
- 58. Reformer Surveys Modelling and Analysis - VULCAN REFSIM •Developed using research and plant data •Accurate analysis of Radiant box •Results are: •Kinetic model •Equilibrium model •Tube wall temperatures & margins •Pressure drops •Carbon laydown prediction
- 59. Reformer Surveys Modelling and Analysis – Aspen HYSYS V8 •Flowsheeting package •Contains VULCAN REFSIM Reformer and Reactor models •Used for detail modelling of the plant •Both front end and loop •Steam system •Heat recovery •Results include: •Flow sheet of the plant •Heat loads of coils and exchangers
- 60. Reformer Survey Results - Statistical Temp. Analysis •Look at various splits of box •Depending on design and size •Look at •Average •Maximum •Minimum •Standard deviation •Spreads •Three dimensional plots •Frequency plots •Compare to others
- 62. Reformer Survey Frequency and Cumulative Plot Frequency Plot for the Bottom Corrected TWT's 0 5 10 15 20 25 0 -840 840 -850 850 -860 860 -870 870 -880 880 -890 890 -900 900 -910 910 -920 920 -930 930 -940 940 -950 950 -960 960 -970 970 + Temperature Range, C Percent 0 10 20 30 40 50 60 70 80 90 100 Btm Corrected (%) Btm Corrected Cumulative (%)
- 63. • Detailed heat and mass balance of Primary reformer • Using kinetics and equilibrium • Pressure drop prediction • Process and tube temperature profiles • Flowsheet of plant • Ideas for plant improvements • Efficiency or Rate increases
- 64. Reformer Survey Tube Wall Temperature Results •Max tube wall temperature •Predicted by VULCAN REFSIM •Tube wall temperature margin is •Predicted by VULCAN REFSIM •Worst case analysis •Based on GBHE Codes •Based on 100,000 hours operation
- 65. Introduction Background Radiation and Temperature Measurement Reformer Survey Inputs Case Study 1 Case Study 2 Case Study 3 Conclusions
- 66. Reformer Surveys GBHE Tube wall Temperature Margins •Based on •inlet pressure •hoop stress calculation •GBHE Tube wall temperature margins do not include •transient stresses (Start Ups/Shut Downs) •longitudinal stresses •bending stresses •weld region stresses
- 67. Reformer Surveys General Conclusions Indications of: •Tube appearance •Hot spots or bands •The operation of reformer •Optimization •Current catalyst performance •Benchmarking •Instrument Calibration •Oxygen levels
- 68. • Air damper stuck • Air preheater leaks • Correct exit temperatures • Flue gas recirculation • Flue gas maldistribution • Explanation of early tube failures
- 69. Accurate assessment of reformer requires ◦ Tube wall temperature survey ◦ Extensive data collection ◦ Data reconciliation by H&M balance ◦ Fully predictive reformer model All of the above used together
- 70. Proven and robust methodology ◦ Used on over 30 plants Allows identification of problems ◦ Identified NEW issues with designs Has saved customers money ◦ Short Term - Efficiency/Production improvements ◦ Long Term - Extended tube life