close

Вход

Забыли?

вход по аккаунту

?

NOx Reduction Technology in PF Boilers.

код для вставкиСкачать
Dev. Chem. Eng. Mineral Process., 7(1/2),pp. 1IS-130,1999.
N0,Reduction Technology in PF Boilers
S. McCahey*,J.T. McMullan and B.C. Williams
NICERI: University of Ulster, Coleraine, BT52 ISA, Northern Ireland
Technical, environmental and economic assessments were performed, using the ECLIPSE
process simulator,for a supercritical coalfired PF power station to investigate the impact
of NOx reduction technologies. The study was modelled upon the Amer 9 Power Station
at Geertruidenberg, The Netherlandr, and both combustion and post-combustion NO,
reduction technologies were evaluated. SCR gave the largest reduction in NO, emissions
to below 5Omg/Nm3, but at an electricity cost of 2.44p/kWh, compared with 2.23p/kWhfor
the reference caxe and 2.26p/kWh for coal-over-coal reburning. The latter produced a
50% reduction in NOx emissions to below 200mg/Nm3. Use of a microniser, to reduce
coal particle size, in coal-over-coal reburning is not justified, unless its absence produces
excessive unburnt carbon losses or requires a reburn coal 50% more expensive than the
main coal. With natural gas-over-coal reburning, an unrealisable natural gas price of
fO.98/GJ is required to make it competitive with coal-over-coal reburning and natural gas
prices between fl.76/GJ and €1.93/GJ are requiredfor it to compete with SCR.
Introduction
Nitrogen oxides, generally termed NO,, include NO, NO, and N,O. Combustion of fossil
fuels contributes significantly to the annual emissions of NO,. In pulverjsed coal-fired (PF)
* Authorfor correspondence.
115
S.McCahey, J.T. McMullan and B.C. Williams
boilers the N20and NO, emissions are usually of lesser significance than the emissions of
NO. NO, emissions, in general, are considered a major pollutant of the atmosphere. N,O is
said to participate in global warming. NO and NO, are believed to play a major role in the
formation of ground level ozone, photochemical smog and acid rain [I, 21. Increasingly
stringent regulatory requirements provide impetus for the development and use of advanced
technologies to reduce air pollutant emissions, includingNO,.
Conventional coal based power generation technology relies heavily on PF boilers and
over 1000 of these plants currently exist worldwide [3]. It is unlikely that use of low NO,
burner technology on its own will be sufficient to meet projected European NO, emission
standards. Therefore, additional NO, reduction technologies, such as SCR,coal-over-coal
reburning and natural gas-over-coal reburning, are required.
NO, Reduction
NO, emissions can be controlled either during or after combustion. Post-combustion
techniques such as SCR, address emissions after they form and are generally more expensive.
Combustion methods prevent NO, fiom forming in the first place.
Selective catalytic reduction (SCR). In the SCR process, NO, is reduced with NH3(or
other nitrogen source like urea) in the presence of a catalyst to nitrogen and water. This
technique can be used at temperatures between 150'C and 450°C [2], depending on the
catalyst employed. Zeolites and activated coals as well as titanium oxide, iron oxide and
vanadium based catalysts have been used [4].Positioning the catalyst after the flue gas
desulphurisation (FGD) system is known as tail gas arrangement. After desulphurisation the
flue gas is cleaner but at a reduced temperature. Therefore, reheating is necessary upstream of
the SCR system. Use of SCR provides a high conversion of NO to N2,of around 80-90 ~01%.
Reburning technology. The term reburning was fist introduced in 1973 to describe the
process of secondary fuel injection [S] and is shown schematically in Figure 1 [6]. In the
main combustion zone, which generally operates under low excess air conditions, coal is
burned releasing between 70-90% of the total heat input to the boiler. Downstream, in the
reburning zone, additional hydrocarbon fuel is introduced without combustion air, creating a
fuel-rich, oxygen deficient environment wherein the secondary fuel breaks down to produce
116
NO, reduction technology in PF boilers
I
reburn znnc
I
Figure 1. Coal Reburning Boiler.
n
Figure 2. PF Coal-Fired Power Station.
117
S. McCahey, J.T. McMullan and B.C. Williams
hydrocarbon radicals. These react with NO, produced by combustion of the main fuel,
reducing it to atmospheric nitrogen. Overfire air completes the process in the burnout zone,
ensuring complete combustion of any remaining hydrocarbons and also carbon monoxide.
This work is based on computer simulations of alternative NO, reduction technologies
when applied conceptually to the Amer 9 power station, which is a 600 MWe supercritical
PF power station with low-NO, burners [7]. The technical, environmental and economic
evaluations have been performed using the ECLIPSE process simulator [8]. The fust study
looked at the effect of installing a SCR NO, reduction system after the FGD system. The
second and third studies examined the performance of the Amer 9 power station if either
natural gas-over-coal reburning or coal-over-coal reburning was installed. The effect of
using micronisers, instead of mills, in the coal-over-coal reburning system was also
assessed and a number of sensitivity analyses were performed. These studies incorporate
the results of reburning test work performed by Mitsui Babcock at their pilot plant facility
in Renfrew, Scotland.
Process Descriptions
The evaluated systems have all been based around the Amer 9 power station at
Geemidenberg in the Netherlands [7], a description of the process follows and is shown
schematically in Figure 2.
Coal for the power station is shipped by barges from the seaports. Normal coal storage
facilities are provided from where the coal is pulverised in mills and then transferred
pneumatically using preheated air to a two pass once through boiler, with a spirally wound
single furnace and tangentially fued low NO, burners. Most of the unburned coal and ash
is removed at the base of the furnace, with the rest carried forward with the hot gases and
removed in cold-side electrostatic precipitators.
Before reaching the electrostatic precipitators the hot gases are cooled first by
transferring heat to steam in the superheater tubes and the reheater tubes, then by
transferring heat to condensate in the economiser section. Finally heat is transferred to
combustion air in the air preheater section. The steam cycle is a supercritical single reheat
system. The steam leaving the superheater is sent to the turbine stop valve where it is
118
NO, reduction technology in PF boilers
I
T I
-
Figure 3. Wet Limestone FGD Process.
Figure 4. SCR NOx Reduction Technology.
119
S.McCahey, J. T. McMullan and B. C.Williams
expanded in the high-pressure turbine. The steam turbines have facilities for steam
extraction and allow for steam to be tapped off to the regenerative feedwater heaters.
Drains from the three high-pressure feedwater heaters are fed to the deaerator. The steam
from the high-pressure turbine is then reheated before passing through one double flow
intermediate pressure and three double flow low-pressure turbines. At the crossover from
the intermediate to the low-pressure turbines steam is extracted for the deaerator. The
steam from the low-pressure turbine is condensed and the condensate is pumped through
the four low-pressure feedwater heaters to the deaerator. From the deaerator tank the boiler
feed pump forces the condensate through the three high-pressure surface-type feedwater
heaters and the economiser before entering the boiler and completing the steam cycle.
The cooled gases are exhausted via the induced draught fan to a wet limestone gypsum
flue gas desulphurisation (FGD)system where most of the SO, is removed. This process is
based on the Deutsche Babcock design [9],which is shown schematically in Figure 3.
Process design conditions are available [lo]. The flue gas from the electrostatic
precipitators is first cooled against the clean gas and then fed to the base of the spray
tower. Limestone solution is circulated through the sprays in the tower and the SO, in the
flue gas reacts to form calcium sulphite. In the base of the spray tower the calcium sulphite
is oxidised to gypsum which then settles out. The gypsum solution is pumped through a
hydrocyclone and then fed onto a filter table where most of the water and impurities are
removed.
The gypsum is then ready for sale for use in plasterboard manufacture and the
wastewater is treated to separate the impurities. The clean gas is then reheated before
being vented up the stack to the atmosphere.
Modifications to this basic model were investigated to evaluate alternative NO,
reduction technologies. Four variations involving SCR and reburning technology were
explored. The SCR process that was evaluated is based on the Hitachi tail-end DeNO,
design [ 1 13 and is shown in Figure 4. The flue gas from the FGD system is preheated with
clean hot flue gas from the DeNO, reactor and natural gas is burned to maintain a
temperature of 300°C. Ammonia solution is evaporated and mixed with air before being
injected into the flue gas upstream of the DeNO, reactor. The DeNO, reactor contains a
120
NO, reduction technology in PF boilers
honeycomb of TiO, which catalyses the reaction of the ammonia with NO, to form
nitrogen and water. The SCR process reduces the NO, emissions by about 90%.
With the coal-over-coal and natural gas-over-coal reburn technologies a NO, emissions
reduction of about 50% can be achieved by staging the combustion. In the Amer 9 power
station the coal is burned in six levels of burners all with the same stoichiometry of about
1.2. In the rebum systems modelled, the bottom 5 sets of burners bum the main coal fuel
but with a stoichiometry of 1.12. The top set of burners bum the reburn fuel, which is
either natural gas or coal, at a stoichiometry of 0.9. Above this top set of burners, overfire
air is added to complete the combustion and maintain the overall stoichiometry at 1.2.
This staged combustion reduces the formation of NO, in the combustor. One difficulty
that can be found with the reburning process is to achieve a high carbon burnout in this
staged combustion mode. With coal-over-coal reburning, using a more volatile reburn coal
helps, as does grinding the rebum coal finer. The option of using a microniser to provide a
fine rebum coal is assessed in the fourth study. A microniser is a jet or fluid energy mill,
which uses high velocity gas jets to entrain the coal particles and promote self-attrition.
The gas jets use high-pressure steam taken from the boiler steam system and the steam
travels along with the coal into the furnace. The design conditions for the microniser were
supplied by James Howden Group Technology Limited [12].
Results and Discussion
The technical and environmental results from the four cases investigated are given in
Table 1. Use of SCR involves burning natural gas to raise the flue gas temperature to that
required by the DeNO, catalyst, thus increasing the thermal input. The auxiliary power
requirement also rises due to the additional load on the induced draft fan. This reduces the
net power production by 3.8MWe to 597.7MWe and the efficiency from 42.2%to 41.3%,
in terms of its higher heating value (HHV). The specific CO, emissions rise from
759gkWh to 772gkWh due to this reduced efficiency, but the main environmental change
is the reduction in specific NO, emissions from 0.96gkWh to O.lg/kWh. The capital cost
121
S.McCahey, J. T.McMullan and B. C. Williams
Table 1. Technical and Environmental Results
System
Thermal Input 0
HHV
122
With SCR
Coallccal
Rebum
With Micmniscr
1444
1426
1420
Nat WCoal
Rebum
1451
NO, reduction technology in PF boilers
breakdown for all the technologies considered, and their economic implications have been
discussed elsewhere [ 101. Case 1, with a coal price of €25/tonne and natural gas price of
f2.0/GJ gives a break-even electricity selling price (BESP) of 2.44pkWh, which compares
with the 2.23pkWh for the reference case. Case 3, the equivalent case only with coal
costing E40/tonne gives a BESP of 2.91pkWh compared with the reference case of
2.70pkWh. The effect of natural gas price, Cases 2 and 4, is small due to the relatively
small amount used. The additional capital cost required for the SCR was calculated at
f37M. There are several ways that SCR can be applied, such as high dust, low dust or tail
gas configuration. These affect the capital cost estimate and the catalyst life for the SCR
system. Cases 5 and 6 look at the sensitivity of the BESP to estimated capital costs for
SCR, in the range f25M to f60M. For a coal price of f40/tonne and natural gas price of
€3.O/GJ, the BESP varies from 2.88pkWh to 3.02pkWh over this range of capital costs.
The main change in the technical and environmental results for coal-over-coal
reburning is the increased power required by the fans. This results from the need to recycle
flue gases to the reburn coal mill and to provide good mixing of the overfire air with the
furnace gases to achieve good burnout. This reduces the net power production by 4.1MWe
to 596.4MWe and the HHV efficiency from 42.2% to 41.8%. The specific CO, emissions
rise from 759gkWh to 767gkWh due to this reduced efficiency, but the main
environmental change is the near 50% reduction in NO, emissions from 0.96gkWh to
0.49gkWl-1. Case 7, with a coal price of f25/tonne a BESP of 2.26pkWh is achieved,
which compares with the 2.23pkWh for the reference case. Case 8, the equivalent case but
with the coal costing f4Oltonne gives a BESP of 2.74pkWh compared with the reference
case of 2.70pkWh. Therefore, there is an additional charge on the BESP of
- 0.04pkWh
to compensate for the loss of efficiency and the increased capital cost. With Case 9 a
difference in cost between the main and rebum coal is introduced. The main coal costs
f2Ytonne and the reburn coal costs MO/tonne, giving a BESP of 2.36pkWh, an increase
of 0.13pkWh over the reference case. This price difference is used to illustrate the
sensitivity of the BESP to the relative price that is paid for the main and reburn coal. Case
10 shows that a price of f54honne could be paid for the rebum fuel and still give a BESP
similar to the SCR Case 1. Cases 11 and 12 look at the sensitivity of the BESP to
123
S.McCahey, J.T. McMullan and B.C. Williams
estimated capital costs. The capital cost for building a new plant to the coal-over-coal
reburn technology specification was calculated as E365M compared to building a new
reference plant at f363M. The maximum additional capital cost required for retrofitting
coal-over-coal reburn technology to an existing plant was estimated as €16M, depending
on the cost of air heater modifications. This gives a BESP of 2.32pkWh. In order to
achieve the same BESP as SCR,the allowable additional capital investment is E46M.
The third NO, reduction technology studied was coal-over-coal rebuming using a
microniser to give a finer coal particle in the reburn zone. Compared with the previous
coal-over-coal reburn technology, the thermal input is similar, but there is a reduction in
power output from the low pressure steam turbine of 11.3 MWe due to the steam required
by the microniser. The total auxiliary power requirement is down 0.4 MWe due mainly to
taking one of the coal mills out of service. This gives an overall HHV efficiency of 4 1.2%
compared with 41.8% for the previous coal-over-coal reburn case. The specific CO,
emissions rise from 767g/kWh to 780gkWh due to this reduced efficiency, and there is a
slight increase in SO, and NO, emissions. The increased capital cost to E376M for new
plant is associated with the purchase of two 20tonnesh micronisers to replace one coal
mill, and additional surface area required in the economiser section and the air heater. Case
13, with a main coal price of €25/tonne and a reburn coal price also of €25/tonne gives a
BESP of 2.34p/kWh, which compares with the 2.26pkWh for the basic coal-over-coal
reburn case. Case 14, the equivalent case only with both the main and reburn coals costing
f40honne gives a BESP of 2.82pkWh compared with the basic coal-over-coal reburn case
of 2.74pkWh. Therefore, there is an additional charge on the BESP of 0.08pkWh to
compensate for the loss of efficiency and the increased capital cost associated with the
microniser. The capital cost calculated for the coal-over-coal rebum technology with
microniser was €376M, which is the cost of building a new plant. However, if an existing
plant was retrofitted then the total capital cost would be higher due to modification and
equipment replacement costs. Alternatively, some of the existing equipment (heat
exchangers) could be retained but with a loss in efficiency due to higher economiser and
air heater exit temperatures. These two scenarios are examined in Cases 15 and 16. An
estimate of the maximum additional capital cost that is required, whilst maintaining the
124
NO, reduction technology in PF boilers
same efficiency as quoted was E23M (Case 15). This gives a BESP of 2.39pkWh. By not
making changes to the economiser and air heater there is a potential reduction in efficiency
of 0.25% points to 40.95% and a saving in capital cost o f f 18M (Case 16). This gives a
BESP of 2.3 l p k w h .
The fmal NO, reduction technology studied was natural gas-over-coal reburning.
Compared with coal-over-coal reburn technology the thermal input is higher due to an
expected increase in unburnt carbon levels. Test results from the Mitsui Babcock pilot
plant indicate that the unburnt carbon percentage increases by a factor of two when natural
gas is used as the reburn fuel instead of coal[lO]. The total auxiliary power requirement is
down 0.4 MWe due mainly to taking one of the coal mills out of service. This gives an
overall HHV efficiency of 41.1% compared with 41.8% for the basic coal-over-coal
reburn case. The CO, emissions fall from 767gkWh to 704g/kWh as 20% of the thermal
input comes from natural gas. This also accounts for the reduction in SO, emissions from
0.94gkWh to 0.75gkWh, with the NO, emissions remaining the same. The capital cost of
the system is slightly lower than for coal-over-coal reburning due to the reduced cost of
coal storage and one less coal mill. With the Amer 9 power plant a natural gas supply is
already available to the plant and so only a small natural gas connection cost is involved.
Case 17, with a coal price of f25/tonne and natural gas price of EZ.O/GJ gives a BESP
of 2.45p/kWh, which compares with the 2.26pkWh for the basic coal-over-coal reburn
case. Case 19, the equivalent case only with coal costing €40/tonne gives a BESP of
2.84pkWh compared with the basic coal-over-coal reburn case of 2.74pkwh. The effect
of natural gas price, Cases 18 and 20, is now significant due to the larger quantity used. An
increase in natural gas cost from E2.0/GJ to f3.0/GJ increases the BESP by -0.19pkWh.
Cases 21 and 22 look at a retrofitting situation and an estimated additional capital cost in
the range f5M to f 16M, depending on the extent of modifications to the air heater section.
For a coal price of €25/tonne and a natural gas price of €3.O/GJ the additional capital
investment increases the BESP by 0.03-0.07pkWh. Cases 23 and 24 examined the natural
gas price that would be required in order to give the same BESP as the equivalent coalover-coal reburn technology. For a coal price of E2Ytonne the equivalent natural gas price
is €0.98/GJ and for a coal price of €40/tonne the equivalent natural gas price is f1.53/GJ.
125
S.McCahey, J.T. McMullan and B.C. Williams
Table 2. Economic Results
126
NO, reduction technology in PF boilers
The next two cases, 25 and 26, look at the maximum natural gas price that could be paid
and still give the same BESP as the basic SCR case, for both a new natural gas-over-coal
reburn plant and a minimum cost retrofit plant. For Case 25, the new natural gas-over-coal
reburn plant, the maximum allowable natural gas price is €1.93/GJ, and for Case 26, the
minimum cost retrofit plant, the maximum allowable natural gas price is €1.76/GJ. The
final two cases examine the situation where a natural gas connection does not already
exist, which was not the case for Amer 9 power station. In Case 27 it is assumed that a
5km pipeline connection is required, at a cost of €6SM and in Case 28 it is assumed that a
20km pipeline connection is required at a cost of €2 1.5M [ 131. The effect is to increase the
BESP from 2.45pkWh for Case 17 to 2.48pkWh for Case 27 and to 2.54pkWh for Case
28.
Conclusions
The assessment of alternative NO, reduction technologies, involving modifications to the
basic Amer 9 model, was successfully completed using the ECLIPSE process simulator.
These studies provided a detailed technical, environmental and economic analysis from
which the following conclusions can be drawn.
All of the technologies investigated provide considerable reductions in NO, emissions.
The coal-over-coal and natural gas-over-coal reburn systems both produced approximately
50% less NO, than the reference case, however the SCR post-combustion method showed
the most substantial change, producing a 90% reduction. Of the technologies considered,
SCR is the only method of achieving NOx emission levels below 50mg/Nm3.
Use of these NO, reduction technologies generally incurred a small increase in
auxiliary power and consequently a slight increase in CO, emissions as net electrical
efficiencies dropped. This increase in auxiliary power was mainly due to higher fan duties.
Variation between the systems’ auxiliary power requirements was slight and mainly
resulted from the fan duty requirements of the reburn systems and the requirements of the
optional sixth coal mill. Operating a microniser in a coal-over-coal reburning system
I27
S. McCahey, J.T. McMullan and B.C. Williams
reduced the power output as steam was taken from the turbine cycle and fed to the
microniser. In consequence, whist this system had the lowest thermal input, it also had the
highest specific emissions of CO,, as well as SO, and NO,, due to its low electrical
efficiency.
Use of natural gas reduced overall system efficiency. In the natural gas-over-coal
reburn system this was due to the difficulty in achieving complete burnout, and with SCR
this was due to the flue gas heating requirements. However, the natural gas-over-coal
system, with 20% less thermal input from coal, showed a marked reduction in C 0 2 and
SO, emissions.
At a coal price of €25/tonne and a new build capital cost of f363M, the reference case
had a break-even selling price (BESP) of 2.23 pikwh. The SCR system was estimated to
add f37M to the capital cost resulting in a BESP of 2.44 p/kWh. At the same coal price,
the BESP for the coal-over-coal reburn system was calculated at between 2.26 p k w h and
2.32 pikwh for capital costs in the range E365M to f379M, depending on new build or
retrofit. At the lower capital cost, a BESP of 2.44 pikwh was achieved when the reburn
coal price rose to f54/tonne. It can be seen, therefore, that the SCR system did not
compete economically with the coal-over-coal reburn system, but rather is a necessary
feature of a very low NO, system.
Use of a microniser was estimated to add an additional f 11M to the capital cost of a
new coal-over-coal rebum plant and, at a coal price of €25/tonne, a BESP of 2.34 pkwh
was determined. Hence this route was regarded as unfavourable from an economic as well
as an emissions point of view and not to be followed unless the normal coal-over-coal
reburn configuration was found to produce excessive unburnt carbon losses.
Assuming an on-site supply of natural gas, the natural gas-over-coal system was
estimated to have the lowest capital cost of the systems considered. For a natural gas price
of E X J , the BESP was 2.45 p k W , much greater than either the basic coal-over-coal
reburn case or that which employed a microniser. To compete with these systems, an
unrealisable natural gas price of €0.98/GJ was necessary. The natural gas-over-coal reburn
plant was calculated to be competitive with SCR at natural gas prices between fl.76/GJ
and El.93/GJ, but it was, of course unable to match the Iatter in reducing NO, emission
I28
NO, reduction technology in PF boilers
levels. Without a pre-existing gas connection, the economic viability of this natural gasover-coal reburning system was considered to be further disadvantaged by distance from
supply.
Acknowledgements
This work has been partly funded through the European Commission JOULE-THERMIE
Programme - “Clean Coal Technologies for Solid Fuels R&D (1996-1998)”. I would like
to acknowledge the assistance provided by Ing A.J.C Korthout, Plant Manager of the
h e r 9 Power Station at Geertruidenberg, by Jim Cooper of James Howden Group
Technology Limited, as well as the other partners in the Group, particularly KEMA,
Mitsui Babcock Engineering Limited and ENEL SPA.
References
1. Moore, M.J. 1997. NO, Control in Gas Turbines. Seminar papers: Emission Control
Technologyfor Power and Process Plant, IMechE S487, London.
2. Muzio, L.J. Quartucy, G.C. 1997. Impl. NO, control: research to application. Prog.
Energy Combust. Sci., 23(3), 233-266.
3. McMullan, J.T. Williams, B.C. Sloan, E.P. 1997. Clean coal technologies. Seminar
papers: Emission Control Technology for Power and Process Plant, IMechE S487,
London.
4. van der Lans, RP. Glarborg, P. Dam-Johansen, K. 1997. Influence of process parameters
on nitrogen oxide formation in pulverised coal burners. Prog. Energv Combust. Sci.,
23(4), 349-377.
5 . Wendt, J.O.L, Stemling, C.V. Matovich, M.A. 1973. Reduction of sulphur trioxide and
nitrogen oxides by secondary fuel injection. 14th Symp. (Int.) on Combustion. 88 1.
6 . Hjalmarsson, A.K. Soud, H.N. 1990. Systemsfor controlling NOxJi.om coal combustion.
IEACW30, ISBN 92-9029- 182-6, London.
7. EPZ Reports. c1988. The 600 MW supercritical monotube steam generator Amer 9 and
Amercentrale eenheid 9, technische gegevens.
129
S. McCuhey, J.T. McMullun and B.C. Williams
8. Williams, B.C. 1994. The development of the ECLIPSE process simulator and its
application to the techno-economic assessment of fossil fuel based power generation
technologies. DPhil Thesis.
9. Gramelt, S . 1994. FGD system for 600 MWe coal fired power plant - process
description, PFD, mass and energy balance, equipment specifications. Deutsche
Babcock Anlagen GMBH. Private Communication.
lO.McCahey, S. Campbell, P.E. McIlveeWright, D.R. Williams, B.C.McMullan, J.T.
1997. Performance Prediction in Advanced Coal Fired Boilers Contract No JOF3 CT95 - 0005.
11. Breihofer, D. Mielenz, A. Rentz, 0. 199 1. Emission control of SO, ,NO, and VOC at
stationary sources in the Federal Republic of Germany. University ojkhrlsruhe.
12.Cooper, J. 1997. James Howden Group Technology. Coal Microniser - the Howden
Micromill. Private Communication.
13. Feines, G. 1997. British Gas. Private Communication to MBEL.
130
Документ
Категория
Без категории
Просмотров
0
Размер файла
691 Кб
Теги
nox, technology, reduction, boilers
1/--страниц
Пожаловаться на содержимое документа