STAR-CCM+ an invaluable tool for coal fired power plant design - - PowerPoint PPT Presentation

STAR-CCM+ an invaluable tool for coal fired power plant design optimization

Ignus le Roux, Aerotherm Computational Dynamics & Warwick Ham, Steinmüller Africa Bilfinger | 2014 STAR Global Conference

Boiler Process

18 – 19 March 2014

Presentation Structure

▪ Aerotherm Computational Dynamics & Bilfinger Steinmüller Africa History ▪ Problem Statement ▪ Secondary Air Flow Measurement Error ▪ Combustion Instability ▪ Case study – Komati Power Station Unit 3 ▪ Model creation ▪ CFD Revision of the burner flow measurement philosophy ▪ Flow only evaluation to improve burner stability ▪ Drop Tube Furnace Modelling - char burnout rate validation ▪ Full Furnace Combustion Model ▪ Conclusion ▪ Further development ▪ Questions

Aerotherm CD & Bilfinger Steinmüller Africa History

▪ Aerotherm CD have been supporting Steinmüller Africa with Computational Fluid Dynamic Simulation Services for the past 8 years (since 2006). ▪ Projects varying from burner optimization studies, detail coal combution modelling to boiler erosion analyses. ▪ Projects performed initially using STAR-CD and since 2008 solely in STAR-CCM+

Case Study – Komati Power Station

▪ Komati Power Station, is a coal-fired power plant operated by Eskom. ▪ The station is situated between the city of Middelburg and town of Bethal in South Africa’s Mpumalanga province. ▪ Technical details: ▪ Five 100MW units ▪ Four 125MW units ▪ Installed capacity: 1,000MW ▪ The first unit was commissioned in 1961 and the last in 1966. ▪ In 1988, three units at Komati were mothballed, one was kept in reserve and the other five were only

• perated during peak hours. In 1990 the complete station was mothballed.

▪ In 2008 unit 9 was the first to be recommissioned under Eskom's return to service project.

Case Study –Komati Power Station Unit 3

Implementation - Background ▪ There are 12 Pulverised Coal Burners in a common windbox firing through the rear furnace wall of the boiler. ▪ The original Pulverised Fuel burners were of 1950’s design ▪ During RTS (Return To Service) the old PF burners were found to be severely worn. ▪ The old burners used Light Fuel Oil for ignition and Komati was to be converted to Heavy Fuel Oil to save

• perating costs.

▪ It was noted that the original furnace burner openings were smaller on units 1-3 than on other units, but could not be replaced due to financial constraints. ▪ The old burners were to be removed and replaced with new PF burners. ▪ The new burners were designed, manufactured and installed in units 1-3 during 2011.

Case Study –Komati Power Station Unit 3 Background – Where this started

Unit 3 Firing Floor showing Common Windbox and 12 Original PF Burners

Case Study –Komati Power Station Unit 3 Background – Where this started

Photo of original Komati PF Burners – Top firing floor view, burner in situ with oil burner

Case Study –Komati Power Station Unit 3 Background – Where this started

Photo of original Komati PF Burners – View from inside windbox Note wear on Swirl vanes

Case Study –Komati Power Station Unit 3 Implementation

Photo of New Komati PF Burners – View from firing floor during installation

Case Study –Komati Power Station Unit 3 Implementation

Photo of New Komati PF Burners – View from firing floor during

• peration

Case Study – Komati Power Station Unit 3 Problem Statement

Secondary Air Flow Measurement Error. ▪ At low flow settings featuring small cylindrical damper openings, a lower pressure is measured at the back plate than in the burner throat producing complications with the flow rate calibration.

Case Study – Komati Power Station Unit 3 Model Creation

▪ The Komati PS Unit 3 features 12 swirl burners fired from the rear wall. ▪ Sufficient 2D drawing were available to create the model of the furnace, the detailed burners and the common windbox (incl. internal structures). ▪ However no geometric information was available of the ducting from the outlet of the regenerative air preheater up to the entrance to the common windbox. ▪ This section needed to be included in the analysis as it has a prominent influence on temperature and flow distribution in the windbox.

Case Study – Komati Power Station Unit 3 Model Creation

▪ The world class spatial laser scanning capabilities of Venter Consulting Engineers (Pty) Ltd. were used to capture the missing detail.

Raw point cloud scan data

Case Study – Komati Power Station Unit 3 Burner Flow Measurement Philosophy

▪ Flow through the swirl burners are governed by adjusting the offset of cylindrical dampers from the burner back plate.

Case Study – Komati Power Station Unit 3 Burner Flow Measurement Philosophy

▪ The burner flow rate is based on the pressure differential measured across a specific burner. ▪ The upstream pressure is measured at a single position against the burner back plate. ▪ The downstream pressure is based on the average of six pressure measurements measured in the burner throat.

Case Study – Komati Power Station Unit 3 Burner Flow Measurement Philosophy

▪ At low flow settings featuring small cylindrical damper openings, a lower pressure is measured at the back plate than in the burner throat producing complications with the flow rate calibration.

General damper position Small damper position

Case Study – Komati Power Station Unit 3 Burner Flow Measurement Philosophy

▪ The high amount of internal detail included in the burner windbox model allowed the cause of the negative burner dP measurement to be identified and allowed Aerotherm CD and Steinmüller Africa to determine an

• ptimal upstream pressure measurement location.

▪ The recommendation was made to replace the single pressure measurement against the burner back plate with a small diameter pipe located inside a local stagnation region with several openings distributed around the burner effectively obtaining an average pressure measurement.

Case Study – Komati Power Station Unit 3 Burner Flow Measurement Philosophy

▪ It is clear that measurements in the proposed location will yield a positive pressure drop for all damper positions evaluated.

General damper position Small damper position Pressure at proposed pressure probe location

Case Study – Komati Power Station Unit 3 Burner Flow Measurement Philosophy

▪ Certain spatial restrictions on site prevented installation of the measurement probe in the proposed location. ▪ The model was used to develop a conical fence which allowed the implementation of the probe in the proposed location as it increased the size of the stagnation regions sufficiently.

Conical flow restriction

Case Study – Komati Power Station Unit 3 Problem Statement

Combustion Instability. ▪ Unit 3 went into operation during 2012. ▪ During summer of 2012-2013 coal moisture content rose above design limits (10% max) and burner stability became problematic. ▪ Several oil burners had to be in service during high and low load operation to support PF combustion. ▪ ESKOM suggested cutting back PF and core air part to improve mixing between PF and Secondary Air. ▪ Windbox and burner model used to assess ESKOM proposed burner modifications. ▪ Several burner cutback modifications investigated (9 in total) and presented to ESKOM

• n a weekly basis.

▪ Operating requirement exceeds design specification.

Case Study –Komati Power Station Unit 3 Flow Evaluation To Improve Flame Stability

Feedback on implementation successes Komati Coal – Supply Limits Unit Limit Value Jan 2013 Apr 2013 Jul 2013 Mar 2014 Coal CV MJ/kg 20 min 19.7 19.45 18.62 20.33 Ash in coal (as received) % 32 max 28 28.2 29.3 27.3 Volatile matter in coal % 18.6 min 18.5 16.2 19.9 20.7 Total Moisture in coal % 10 max 14.7 20.4 13.6 12.6 PF fineness (passing 300 mm) % 98 min 96.8 95.6

• PF fineness (passing 75 mm)

% 65 min 62.2 66.7

• PF mass flow variance from

mean % 15 max 32.4

• 23.2
• Mill Outlet Temperature
• C

75 min 69.8 82.9 87.4 84.8 Oil Burners in operation n 4 2 2

Case Study – Komati Power Station Unit 3 Flow Evaluation To Improve Flame Stability

▪ Under certain operational conditions, the flame ignition front is located very far from the burner face on some

• f the burners in the Komati Power Station Unit 3 Boiler.

▪ To prevent losing the flame in such instances, the station will introduce oil burners to stabilize and maintain the flame. ▪ As the time scheduled was very tight to evaluate possible burner modifications, the highly detailed windbox model which extends up to the furnace exit was used to evaluate possible burner modifications which from a purely flow behaviour perspective could improve the flame stability.

Case Study – Komati Power Station Unit 3 Flow Evaluation To Improve Flame Stability

▪ 9 geometric burner modifications were considered to strengthen the reversed flow towards the burner ▪ This draws more hot flue gas closer to the burner ▪ The hot flue gas activates the newly introduced pulverised fuel at the point where maximum velocity and turbulence forces mixing of the pulverised fuel with the secondary air. ▪ This ensures high heating up rates of the coal particles and stabilizes the flame.

1 Burner return flow 3 2

Case Study – Komati Power Station Unit 3 Flow Evaluation To Improve Flame Stability

▪ In the existing burner design, the termination point of the secondary air, primary air and core air sections line up at the entrance to the cowls section.

Case Study – Komati Power Station Unit 3 Flow Evaluation To Improve Flame Stability

▪ A X-Y Plot visualizing the axial velocity component as a function of the radial position component (relative to the burner axis) on a section at the entrance to the burner cowl provided an excellent tool for comparison of the different burner modifications’ ability to increase the return flow behind the burner.

Case Study – Komati Power Station Unit 3 Flow Evaluation To Improve Flame Stability

▪ The second proposed modification dubbed Mod2 considered changes which would reduce the energy level in the primary air flow (increase flow area, add diffuser) and increase the energy in the swirled secondary air flow (reduce flow area). The core air and primary air termination points were trimmed back by 90mm to create earlier interaction between the fuel carrying primary air and the secondary air.

Case Study – Komati Power Station Unit 3 Flow Evaluation To Improve Flame Stability

▪ The Mod2 changes managed to reduce the energy in the primary air annulus by 19.9% whilst increasing the energy in the secondary air annulus by 11.5%. The increased swirl in the secondary air and the earlier interaction with the lower energy primary air flow results in an stronger and larger return flow region behind the burner. However practically it required major burner modification which was not ideal.

Case Study – Komati Power Station Unit 3 Flow Evaluation To Improve Flame Stability

▪ The fourth modification dubbed Mod4 considered changes which could be affected without removing the

• burner. The modification would however not address any energy imperfections in the different annuli but

would guarantee very early interaction between the fluid stream. The core air and primary air termination points were trimmed back by 200mm

Case Study – Komati Power Station Unit 3 Flow Evaluation To Improve Flame Stability

▪ The Mod4 changes managed through the earlier interaction with the primary air flow to create a much stronger and larger return flow region behind the burner. The simple implementation of this modification resulted in this modification being selected to replace the destroyed burner.

Case Study – Komati Power Station Unit 3 Flow Evaluation To Improve Flame Stability

▪ Mod4 was implemented, i.e. cut back burner PF pipe and core air tube by 200 mm. Boilers have been running with stable combustion since April 2013. ▪ When the coal total moisture exceeds the design range of 10% combustion stability is affected and 2 oil burners are required to be in service to support pulverised fuel stability. ▪ Boiler Pyro positions were moved towards the back wall to allow both pyros to give a stable flame signal. ▪ Unburnt carbon in ash is between 5 and 9% for units 1-3 depending on coal quality, pulverised fuel fineness and excess air. ▪ The decision was made to create a detail coal combustion model to evaluate the burner design ▪ PF and SA Cross Section ▪ Individual Burner Swirl Direction

Case Study – Komati Power Station Unit 3 Drop Tube Furnace Model

▪ A coal sample is passed through the furnace in pure Nitrogen more than 1000°C, devolatilizing the sample leaving only the combustible solid matter. ▪ The sample is then dropped through the furnace, burning in a 3% Oxygen, 97% Nitrogen composition gas at three different temperatures: 1000°C, 1200°C & 1400°C. ▪ The sample is introduced and collected with water cooled probes, which ensures the reaction only starts in the furnace and stops when collected. ▪ Two size ranges are tested: 0-38µm and 38-75µm. The residence time is calculated based on the distance from the particle release point in the furnace that the sample is collected. The composition is analyzed and the percentage burnout established. The results are mass-weighted and presented in the form shown below.

Case Study – Komati Power Station Unit 3 Drop Tube Furnace Model

▪ A CFD model of the drop tube furnace was used to evaluate the coal char burnout rates (shown below) were used to calibrate the combustion model for application in the full furnace model. ▪ Three temperatures were considered: 1000°C, 1200°C & 1400°C. ▪ Two size samples were evaluated: 0-38µm and 38-75µm. Distribution described using a Rosin-Rammler distribution with 19µm & 49µm mean sizes respectively. (Exponent 2) ▪ The inlet and collection probes were not modelled explicitly, but the effect of the cold carrier fluid trans- porting the sample to the furnace have been accounted for in the analyses.

Case Study – Komati Power Station Unit 3 Drop Tube Furnace Model

▪ First order char oxidation reactions. (Smoot, D.J., and Smith, P.J. 1985. “Coal Combustion and Gasification”, The Plenum Chemical Engineering series, New York). ▪ First order reactions for char oxidation in O2 as well as Arrhenius coefficients Smoot and Smith calculated describing the gasification. C+0.5O2→CO C+CO2→2CO (Gasification, Boudouard reaction) C+H2O→CO+H2 (Gasification)

Char oxidizes in the 3%

• xygen background gas as

it falls through the Drop Tube Furnace.

Case Study – Komati Power Station Unit 3 Drop Tube Furnace Model

▪ Mass weighted burnout curves from CFD DTF overlaid onto the experimental DTF results.

10 20 30 40 50 60 70 80 90 100 0.5 1 1.5 2 2.5 3 3.5 1000°C Mass Weighted Burnout 1200°C Mass Weighted Burnout 1400°C Mass Weighted Burnout

Case Study – Komati Power Station Unit 3 Furnace Combustion Model

▪ Half of the boiler was modelled as with the detailed windbox model. ▪ The velocity vector field, temperature distribution and turbulence levels were mapped from the highly detailed windbox model.

Furnace cutting plane modelled as a zero shear wall or symmetry plane. Burner throats modelled as velocity boundaries with velocity profiles, temperature and turbulence levels mapped from the detailed windbox analysis. Outlet from radiation section of the boiler modelled as pressure boundary. Heat extracted through furnace walls.

Case Study – Komati Power Station Unit 3 Furnace Combustion Model

Case Study – Komati Power Station Unit 3 Furnace Combustion Model

▪ As raw coal is introduced, the moisture is driven off using a quasi-steady evaporation model. ▪ A single rate devolatilization model defined in Arrhenius form (shown below) developed by Badzioch and Kawsley, 1970 was built into the model. k=Ae-(E/RT) ▪ The gaseous portion of the volatile matter produces the following products where the coal composition determines the stoichiometry. CHON (coal volatile) → CH4 + CO + H2 + N2 CH4 + O2 → CO + H2 H2 + O2 → H2O CO + O2 → CO2

Case Study – Komati Power Station Unit 3 Furnace Combustion Model

▪ Full furnace combustion model provided insight into the complex interaction between the different burners. ▪ Iso-surfaces proved to be a handy diagnostic tool to understand flow interaction (see example below).

Iso-contours of flow towards the rear wall.

Case Study – Komati Power Station Unit 3 Furnace Combustion Model

▪ Animating the particle tracks also provides clear insight into the complex interaction in the furnace.

Case Study – Komati Power Station Unit 3 Furnace Combustion Model

Case Study – Komati Power Station Unit 3 Furnace Combustion Model

Case Study – Komati Power Station Unit 3 Furnace Combustion Model

▪ The first burner modification considered is based on the knowledge gained during the first phase of flame stability improvement simulations where purely flow were considered. ▪ The modification features a reduction in the secondary air annular cross section and an increase in the primary air annular cross section to respectively increase and reduce the amount of energy in these regions.

Increase in PF cross section

Case Study – Komati Power Station Unit 3 Furnace Combustion Model

▪ Evaluation of the particle tracks reveals that the modification drastically improved flow uniformity. The increase in energy in the secondary air swirl annuli and the reduction in energy in the primary air annuli also managed to establish properly defined swirl coned in front of each burner.

First Proposed Modified Design Base Case Burner Design

Case Study – Komati Power Station Unit 3 Furnace Combustion Model

First Proposed Modified Design Base Case Burner Design

Case Study – Komati Power Station Unit 3 Furnace Combustion Model

First Proposed Modified Design Base Case Burner Design

Case Study – Komati Power Station Unit 3 Furnace Combustion Model

▪ With the improvement in burner flow uniformity, the global interaction changed to such an extent that fewer of the burners managed to pull back hot gas from the furnace. ▪ It became apparent that the interaction resulting from the base case burner flow distribution problems assisted the central burner columns to pull back hot gas from the furnace.

First Proposed Modified Design

Case Study – Komati Power Station Unit 3 Furnace Combustion Model

Base Case

Case Study – Komati Power Station Unit 3 Furnace Combustion Model

▪ Three additional geometric burner modifications were considered, but none of these improved the overall flame stability. ▪ The final investigation considered reversing the swirl direction of the middle row of burners as this result in a situation where the flow from all burners sustain each others swirl action.

Modified Swirl Direction Original Swirl Direction

Case Study – Komati Power Station Unit 3 Furnace Combustion Model

▪ Evaluation of the particle tracks reveals that the modification drastically improved flow uniformity. The increase in energy in the secondary air swirl annuli and the reduction in energy in the primary air annuli also managed to establish properly defined swirl coned in front of each burner.

First Proposed Modified Design First Proposed Modified Design Incl. Swirl Mod.

Case Study – Komati Power Station Unit 3 Furnace Combustion Model

First Proposed Modified Design First Proposed Modified Design Incl. Swirl Mod.

Case Study – Komati Power Station Unit 3 Furnace Combustion Model

First Proposed Modified Design First Proposed Modified Design Incl. Swirl Mod.

Case Study – Komati Power Station Unit 3 Furnace Combustion Model

First Proposed Modified Design Incl. Swirl Mod

Case Study – Komati Power Station Unit 3 Furnace Combustion Model

First Proposed Modified Design First Proposed Modified Design Incl. Swirl Mod.

▪ Based on the summary of ash fusion temperatures for several samples taken from site, the likelihood for fouling to occur could be investigated. The algorithm identifies all particles with a temperature exceeding the ash fusion temperature located within 10mm from a wall with a velocity component towards that wall.

Case Study –Komati Power Station Unit 3 Flow Evaluation To Improve Flame Stability

Feedback on implementation successes Operating data from 2013 - 2014 showing differences: * these coal CV values calculated by unit controller. Unit Jul 2013 Mar 2014 Load MW el 95.6 97.8 SA flow kg/s 29.3 29.1 SA Temp

• C

219 227.7 PA flow kg/s 8.33 9.1 Mill out T

• C

90.2 88.1 Coal flow kg/s 4.7 5.9 Coal CV MJ/kg 20.5 17.7 No Oil Burners i/s n 2

Case Study –Komati Power Station Unit 3 Implementation success

Feedback on implementation successes Operating data from Jul 2013:

Case Study –Komati Power Station Unit 3 Implementation success

Feedback on implementation successes Operating data from Mar 2014:

Case Study –Komati Power Station Unit 3 Flow Evaluation To Improve Flame Stability

Komati Furnace Pyrometers – During operation

Case Study –Komati Power Station Unit 3 Flow Evaluation To Improve Flame Stability

Komati Furnace Pyrometers – During operation

Conclusion

Video of healthy PF flame

Case Study – Komati Power Station Unit 3 Flow Evaluation To Improve Flame Stability

Video of oil flame

Case Study – Komati Power Station Unit 3 Flow Evaluation To Improve Flame Stability

Video of pf flame

Conclusion

▪ Original burner design installed during 2011. Unit 3 went into operation during 2012. ▪ During Jan 2013 burner stability became problematic. Several oil burners had to be in service during low load operation to support PF combustion. ▪ The highly detailed windbox model provided clear insight into the complex flow behaviour in the windbox and the resultant individual burner flow distribution. ▪ The explicit modelling of all internal structures and leakage flow paths allowed for the modification of the measurement device which will produce a positive pressure drop across the burner for the entire range of cylindrical damper positions. ▪ The model allowed the evaluation of 9 burner modifications to from a flow perspective maximised the amount

• f return flow behind the swirl burners.

▪ Mod4 Implemented during April 2013. ▪ Burners operated during the rest of 2013 with no stability problems. ▪ Mar 2014 PF combustion stability is maintained without oil burner firing in spite of excessive coal moisture content. ▪ The full combustion model provided valuable insight into the complex interaction between the 12 swirl burners firing from the rear wall. The model also revealed the strong influence of the individual burner flow distribution emanating from the common windbox on the combustion behaviour. ▪ Model results used to predict furnace wall fouling locations

Further Development

▪ Steinmüller is purchasing a suction pyrometer and gas sampling probe to measure Furnace Exit Gas Temperature (FEGT). ▪ This will allow traverses to be made measuring flue gas temperature as well as flue gas species such as CO, and O2 plus sampling particulates for unburnt carbon content ▪ This will allow verification of the results of the CFD model by comparison with the control room operating results. ▪ Suction pyrometer has multiple ceramic shielded tip with high induced flue gas flow rate for high convective heat transfer with the measuring tip. ▪ Obtain data on coal combustion to verify devolatilisation rate model.

Questions?

Thank you for your support, do you have any questions?