close

Вход

Забыли?

вход по аккаунту

?

Fundamental Principles of Incinerator Design.

код для вставкиСкачать
Dev. Chem. Eng. Mineral Process., 7(5/6),pp.623-640, 1999.
Fundamental Principles of Incinerator Design
J. Swithenbank, V. Nassezadeh, R. Goh and R.G. Siddall
Shefield University Waste Incineration Centre, Dept of Chemical and
Process Engineering, University of Shefield, Shefield SI 3JD, UK
Waste management has become a major concern world-wide and incineration is now
being increasingly used to treat waste which cannot be economically recycled. The
combustion of conventional well specifiedfossil fuels is a very complex process since it
involves two-phase turbulent reucting flow including radiant heat transfer.
Incineration is even more complex since the waste is poorly specified and its
composition variesfiom moment to moment. In the past, the design of incinerators has
not been based on fundamental understanding and modelling of the process, and
empirical rules have had to be used.
Over the lastfew years, computationaljluid &namics (CFD) has provided a means
to model thefieeboard region in a conventional municipal solid waste incinerator but
the open literature contains no rigorousfundamentally based model of the bed region.
The prediction of the jlow composition emerging fiom this region is particularly
important since it provides the "upstream" boundary condition for the jlow
calculations in the fieeboard. For example, the calculation of the subsequent history
of heavy metals requires knowledge of their emission ratefiom the burning bed.
The processes in the bed include dying, pyrolysis, oxidative burning, and
gasifcation of the char. Furthermore, the movement of the grate is designed to mix
the waste as it burns. Indeed, the existence of a rigorous bed model would also permit
the grate design to be optimised, and, if immediate data on the feed were available, a
rational combustion control strategy could be devised. A preliminary model of
combustion in the bed is proposed herein based on governing equations for the
burning of individual "regions" of waste in the upward gas flow, their motion, and
radiant heat transfer within the bed. The emissions of gases fiom the surface of the
bed is very non-uniform with oxygen emitted fiom either end of the bed, organic
compounds fiom the one-third region, and carbon monoxide fiom the centre. The
surface layer of this centralpart of the bed consists of char with gases coming up @om
the oxidising layer below containing NOx derived from the fuel nitrogen. The char
region therefore acts as an important reburn zone and it is suggested that this reduces
some of the NOx to nitrogen. This is important since the minimisation of NOx by
optimising the basic combustion process is likely to be environmentallypreferable to
subsequent control of NOx by the injection of reactants such as ammonia.
The calculation ofjlow and combustion in the fieeboard using CFD can quantifi
the consequences of design concepts, however guidelines are neeakd to devise specifc
design concepts which are worthy of investigation. Thus the computer cannot "invent"
a design but it can quantifi the results of ideas.
Incinerator design thus requires a judicious combination of fundamental
combustion science, ingenious engineering guided by an understanding of the miring
process, and last but not least, practical experience ofpreviousfailures andsuccesses.
623
J. Swithenbank et al.
Introduction
With up to one tonne per year of municipal waste being generated for every man,
woman and child each year across the globe, its safe disposal has become an urgent
environmental problem. Once re-use and recycling have been maximised, there still
remains a large fraction of municipal, clinical and hazardous wastes which must be
disposed of in the best manner possible. Landfill leads to problems of methane
leakage, groundwater contamination, and land usage, and is now perceived as
environmentallyunfriendly. In recognition of these problems, the UK government have
recently imposed a landfill tax to discourage burying waste and encourage energy
recovery fiom waste. Incineration is now accepted as the most environmentally
friendly means of disposing of these forms of waste, and the design of incinerators
must be optimised. It is important to note that the recovery of energy fiom waste
reduces the net release of carbon dioxide to the atmosphere since it displaces the fossil
fuel that would have been used and eliminates the production of methane fiom landfill.
Furthermore, the total energy available fiom waste in Europe alone amounts to a
staggering 100 million tomes per year of coal equivalent.
On the other hand, the emission to the atmosphere of heavy metals and acid gases and
perhaps more important in the longer term, the dioxins/furans (PCDDdPCDFs), is one
of the key issues facing not only the incineration industry but also society as a whole
over the next decade and more. Thus the design of the incinerator plant must include
reliable means to control any emissions. Furthermore, there is a growing requirement
to eliminate all residues fiom industrial processes and thus incineration is now being
used increasingly to convert into inert building material, waste which cannot be
economicallyrecycled. The combustion of conventional well specified fossil fuels is a
very complex process since it involves two-phase turbulent reacting flow including
radiant heat transfer. Incineration is even more complex since the waste is poorly
specified and its composition varies h m moment to moment. In the past, the design of
incinerators has not been based on fimdamental understanding and modelling of the
process, and empirical rules have had to be used. The topic is now receiving increased
attention by the engineering research community and the gap between fundamental
scientific principles, and plant conshuction and operation is now being bridged. Thus
incinerators are evolving rapidly fiom simple covered bonfires, to very sophisticated
process plants. However, reliable data and comprehensive, fundamentally based
design procedures are not yet available to optimise the design, manufacture,operation
and control of incinerators. As a result of our present ignorance, they all fall short of
attainable throughput, quality of ash and plant availability. The price of our ignorance
can be measured as hundreds of millions of pounds to over the next few years. It is an
objective of SUWIC's research programmeto help minimise these costs.
The Problem to be Addressed
Combustionof waste in an incinerator is arguably one of the most complex combustion
problems known The design and control of such a combustion device poses many
problems since there is no satisfactory model of the system as a whole. In particular, a
process as basic to their operation as the combustion on the grate has not yet been
fblly elucidated. Thus at present, a typical design "equation" for the combustion on the
grate is merely that the buming rate is about 400 kg/m2hr. Clearly this is a gross oversimplification of the many complex processes taking place simultaneously. This poor
624
Fundamental principles of incinerator design
state of knowledge can be contrasted With the dramatic progress which has been made
recently in the modelling of reacting; pulverised coal, gas and liquid fuel systems, which
has resulted in impressive reductions in the emission of pollutants with little sacrifice in
efficiency.
The fimdamental analytical approach typically involves identhe governing
differential equations for all of the relevant processes, then solving these equations
simultaneously,often using a numerical code in which the equations are converted fiom
differential into algebraic form. This generic approach has resulted in fundamentally
based commercial codes for fluid dynamics which have developed remarkably rapidly.
They have proved to be very versatile in their applicability to a wide range of design
problems.
The primary objective of our present research programme at S W C is to identify the
appropriate set of governing equations for combustion on the grate of an incinerator
and to develop and validate a procedure for solving these equations. A key aspect of
incinerator bed combustion is the mixing of the solid material on the grate.
Experiments are required to characterise and quantify this mixing process so that it can
be modelled accurately during the numerical solution of the governing equations.
The efficiency of incinerator combustion, and particularly the residual carbon in ash
which determines the suitability of the ash for reuse, depends critically on this mixing
process. There are several significantlyMerent types of grate employingdistinct waste
mixing strategies that are used in current MSW incinerator design. The three most
common are:1. The stepped grate, in which the burning waste is mixed as it falls from one level of
the grate to the next, thus giving oxygen access to the unburned material.
2. The reciprocating grate (and its variants), in which the grate bars reciprocate
resulting in local mixing of the waste.
3. The rotating cylinder grate which consists of a series of slotted rotating drums. The
waste mixes as it tumbles from one drum to the next.
These experiments are fimdamental to practically all future research on incineration,
from understanding of the process to the design and operation of these units.
Experimental Investigations
As pointed out above, the design of incinerators with a burning bed of waste material is
still being hampered by the lack of an accurate mathematical model of the process. The
prediction of the incinerazor gas phase flow is in a more advanced stage of development
using computational fluid dynamics (0)
analysis, although further experimental data
is still required
Unforhmately, it is not possible to scale down many aspects of waSte incineration and
tests on full scale incinerators are essential. Thanks to a close relationship between
SUWIC and Sheffield Heat & Power Ltd, an extended research programme has been
carried out at the Bernard Road Incinerator plant in Sheffield. This plant consists of
two Municipal Incinerators (35 MW) and two Clinical Waste Incinerators (5 MW).
These provide district heating for a large part of the city d e r e the heat is distributed as
hot water to commercial, domestic and industrial buildings. To improve the
economics, a 6 MW generator is now being added to the system. During the last
decade, many investigations have been carried out (please see References. 1 to 10) and
a SUWIC laboratory is located at the plant.
625
J. Swithenbank et al.
Mathematical Modelling of the Gas and Particle Flow
Over the last few years, computational5uid dynamics (CFD) has provided a means to
model the freeboard region in conventional municipal solid waste, clinical waste,
sewage sludge, and special waste incinerators. The prediction of the flow composition
merging from the bed is particularly important since it provides the "upstream"
boundary condition for the flow calculations in the freeboard. For example, the
calculation of the subsequent history of heavy metals demands precise knowledge of
their emission rate fiom the buxning bed.
Computational fluid dynamics is arguably the most important development in modem
engineering practice since it is having such a great in5uence on research, design,
development and production. Almost all CFD codes were originated by combustion
technology research groups as a result of our efforts to umkstand the complexities of
the process by solving the set of interacting governing differential equations. Thus
combustiontechnology can claim to be the father of CFD.
Developments in CFD have always gone hand-in-hand with developments in digital
computer technology and even a basic PC is now a very powerful machine. What of
the fuhae of CFD? In a few short years, the range of applications of the codes has
increased dramatically. In response to demand, their capabilities are still expanding.
For example, the need for the user to be involved in internal details such as grid
generation is gradually receding. Nevertheless, engineers must still uuderstand the
hdamentals of 5uid dynamics and exercise their ingenuity since CFD can only be
used to quantify ideas.
Since many industrial processes involve the behaviour of flow in packed beds, with or
without mixing of the bed material, an important long-tenn spin-off from this
investigation will be a mathematical model which has remarkably wide applicability.
The fundamental analytical approach to model a moving grate type incinerator involves
identifying the relevant mathematical equations governing all the relevant combustion
processes in the burning bed of waste. Due to the heterogeneous nature of the burning
waste and the type of waste burned, these processes are often so complex that attempts
to model all the process are quite difficult.
This section describes the approaches taken in attempts to mathematically link the
reduction of bed height and volume with the complex set of inteawing processes
occurring in the combustion of solid waste in a systematic modelling procedure. In
order to keep this problem manageable whilst producing simulations that are useful for
design pu~poses,careful assumptions and model simplifications which retain the main
essential characteristics of the combustion process are needed. The main processes of
municipal solid waste combustion are drying, pyrolysis or devolatilisation, char
gasiscation and gas phase combustion. These four main processes fonn the basis of the
present theoretical study of the incinerator bed modelling.
626
Fundamental principles of incinerator design
Figure 1.
In the present study, the solid waste is considered to consist of four components:
moisture, volatiles, bound ash and fixed carbon as illustrated in Fig. 1. As the solid
waste is heated, the moisture is removed by vaporisation at 373 K. The heat transferred
to the solid supplies the heat required for the change of phase from liquid fonn into the
vapour form. The regressing drying surface is assumed to remain at 373 K until all
moisture is evaporated.
The volatiles consist of carbon, hydrogen and oxygen and are released during pyrolysis
at 533 K. In the absence of 0,, the gaseous volatiles released are assumed to consist of
C,H,, CO and water. The moving pyrolysis front is assumed to remain at the pyrolysis
temperature until all volatiles are driven out. The pyrolysis products are assumed to be
non-reacting until the volatiles are driven off the bed.
After the moisture and volatiles have been driven off, the solid surface which consists
only of carbon and bound ash begins to char. Any carbon in the solid surface is
assumed to be oxidised by any available O2 to form CO and CO,. During the char
oxidation process the bound ash crumbles and is removed from the solid surface as free
ash. The bound ash and carbon in the solid are assumed to remain in the same
proportion relative to each other. Therefore, the rate of ash removal from the solid
surface will be in the same proportion as the rate of carbon reacted at the solid surface
to CO and CO,. The gas phase consists of 2 volatile species, i.e., CO, and GH,.which
oxidises to CO and C02.
Since the process of incineration reduces the bulk volume of the municipal waste by
approximately 90%, it is apparent that a suitable bed model must also pennit
simulations of reduction in the bed volume. Due to the heterogeneous nature of the
waste incineration process, it may be impractical to adapt existing models describing
the progress of a single particle during gasification of coal, such as the Ash Segregation
model and the Shell Progressive model, to the bed. Hence, the major problem to
overcome in this area was to establish a physical representation of the volume of the
627
J. Swithenbank et al.
solid bed. As moisture and volatiles are liberated during drying and pyrolysis, the solids
become more porous. The signScauce of the intemal pore space to the incineration
process as a whole is yet unhown. Although the pores fonned can be considered as
b e i i occupied by gaseous fluids, the local conditions within the pores in the solid are
very merent h m those in the gaseous fluids occupying the voids between the solid.
Therefore, it is sensible that the void fiaction and the internal pore space are treated as
different elements.
An added advantage of this strategy is, if the shape of the solid is allowed to remain
relatively unchanged in the incineration process, then the void hction of the bed
occupied by the gas phase will be unchanged despite the reduction in the bulk volume of
the bed. To represent the void space in the solids left by the drying and pyrolysis
process, a seventh component hown as the internal pore space is introduced into the
model. This additional component differentiates the solids internal pore space fiom the
gas space occupied by the gas phase at the outer surface of the waste. Component 7
therefore, has neither mass nor enthalpy but has a volume to allow changes in the bed
volume to be represented. This internal pore space is removed during the char
gasificationprocess, hence reducing the bulk volume of the bed. No pore space remains
at the end of gasification.
If the three processes of drying, pyrolysis and gasification of the solid waste do not
occur simultaneously, the change in the bulk volume of the bed with time is as
illustrated as in Fig. 2. The first vertical column in Fig. 2 show the initial volume
distribution of the six main components in the bed. During the drying and pyrolysis
stages, the overall bed volume does not change but the solid becomes less dense due to
removal of moisture and volatiles. The reduction in the volume of moisture and
volatiles in the solid is replaced by an increase in volumes represented by the internal
pore space. In the gasification zone, no additional pore space is being formed but the
volume of the pore space is simultaneouslyreduced as the bound ash becomes fkee ash
and is subsequently removed fkom the solid. Thus, as the char gasifies, the bulk volume
of the bed reduces whilst maintaining a constant volume ratio of the bound ash to the
pore space witbin the solid.
Figare 2.
Fig. 3 illustrates the change
in the bed bulk volume
with time due to the
simultaneous OcCllITence of
drying, pyrolysis and
Intnnal pore space
gasification If the process
of drying and pyrolysis do
not produce any reduction
in the bed volume, the
change in the bed volume
as shown in Fig. 2 is solely
due to the gasification
process. However, the net
change in the internal pore
space in this case is given
by the internal pore space fonned during drying and pyrolysis, less the pore space
removed during gasification. If the rate of formation of the internal pore space is
628
Fundamental principles of incinerator design
greater than the rate of removal of the pore space during gasification, then the volume of
internal pore space will continue to increase until drying and pyrolysis ceases.
Figure 3.
A simplified model for the
simulation of a steady state
moving bed model is an
unsteady static bed model. Any
variation in the static bed
condition with time can be
related to changes in the bed at
positions at the same time h m e
for the moving bed. Variations
in the flow rate of the under
grate air along the bed can be
considered as variation of the
under grate air supply with time.
0assp pace
Fixedcarbon
Volatiles
Drying, pyolysis and gasification
Figure 4.
The static bed can be divided into 4
layers as shown in Fig. 4. Layer A is
the layer of material in its initial state,
layer B consists of the dried material,
layer C is the dried and pyrolysed
material and layer D is the layer of free
ash (i.e., the dried, pyrolysed and
gasified material). The top surface of
the bed is exposed to hot radiating
gases and the air is supplied from the
bottom of the bed. Heat is transferred
through the bed
by
radiation,
/
Internal pore space
I
Time
Radiation
Hot gas
/
/
I
LayerD
t and volatiles
Gasification front
Pyrolysis front
Layerc
Layer B
Drying front
layer^
t
Airin
629
J. Swithenbank et al.
Figure 5.
If the process fkont above layer L
moves a distance AYLFin time At as
shown in Fig. 5, the volume of the layer
L per unit surface area of bed is
decreased by AVu (where, AVU = -AYu). The volume of solid removed
due to movement of the process fkont is .given by:
AVLF,S
wpmessfimts
= (1 - &,)AVLF
The vol&e of the component in the process can be expressed as:
=
(3)
Over the same time step At, the volume of the layer L increases due to factors
immediatelybeneath the layer. The addition of solids volume due to movement of the
process fkont can be given by the volume of the solids added plus the volume of internal
pore space generated by the process:
"W.N
LF.NAvLF,S
AVLB,S= (1- 0 LF.N ) A ~ L F , +
S FNAVLF,N
(4)
where FN is the volume hction of component N which leaves internal pores spaces as N
is being removed fiom the solid. FN has a value which lies between 0 and 1. Therefore
the total volume added due to movement of the process fiont becomes:
1
The net change in volume in any layer L, is given by the sum of AVm - AVu.
Based on the above volume change, the enthalpy balance for the solid for the drying,
pyrolysis and gasification processes can be written in the general form:
where % is the mass rate of removal of component N &om the solid and 4is the net
heat change associated with the process under consideration. 4 and 4 are
parameters which depend on the type of kinetic model or scheme chosen as the process
or reaction rate controlling mechanism. (%)N is related AVws by the equation:
A pot burner experiment is currently underway to provide both validation and constants
for the model. A closely related study is outlined below-
Proposed Optimisation of NO, Reduction by the Bed
The major source of NO, in municipal waste incinerators is the nitrogen in the fuel
(0.4-0.8%). This contrasts with conventional combustion systems where NO, is
produced principally by thermal mechanisms. The so-called prompt NO, production
630
Fundamental principles of incinerator design
only accounts for about 10% of the NO, in a conventional combustion system, and this
mechanism can be neglected as a significant contributor in incinerators.
In the UK, the permissible NO level emitted by incinerators is being gradually
reduced &om about 400 mgMm to a value of about 200 mg/Nm3. The acceptable
level is determined by the principle of BATNEEC. This principle gives scope for
innovation if a superior pollution control technique emerges. At present, it is generally
accepted that the reduction of NO, in incinerators should be achieved by either
catalytic or non-catalytic d o n with ammonia or urea. The selective non-catalytic
reaction (SNCR) takes place in the temperature range 870-950°C. The ammonia
injection is therefore in the radiation shaft above the incinerator finnace section. A
typical reaction is:4NO + 4 w 3 + 0 2 + 4N2 + 6H20
The quantity of ammonia required is 2 to 4 times stoichiometric, however problems
arise due to ammonia slip. Typically, 75 to 90% of the ammonia slip is absorbed in a
semi-dry scrubber cake, whilst 100% is absorbed in a wet scrubber. The surplus
ammonia thus appears as a compound in the filtrate, filter cake or fly ash and this can
lead to an increase in filter M and waste disposal problems. Thus removing NO, in
this manner gives rise to other pollution problems. The use of urea (NH92CO instead
of ammonia in a SNCR system is effective fiom 540 to 1000°C but leads to the
formation of N20. The proportion of N20 formed increases fiom zero at 780°C to
about 18% at 1000°C.
As an alternativeto the SNCR system, selective catalytic reduction (SCR) of NO, takes
place at 400 to 450°C using ammonia and a catalyst such as zeolite. The SCR system is
often located after the normal scrubber system and the temperature requirement means
that the flue gases must be re-heated then cooled to about 140°C before discharge to
atmosphere. Inevitably, this is an expensive operation requiring a heat exchanger,
supplementary burner and a catalytic reactor, however the SCR system is much more
effective than the SNCR system since it can remove 95% of the NO,, whereas the
SNCR system only removes about 50% of the NO,. However, the SNCR system tends
to be installed if it will meet the required NO, level.
Here it is proposed that a degree of NO, reduction can be achieved by opthising the
combustion process in the bed. From the forgoing it will be appreciated that this gives
a major advantage since it does not incur the inherent problems of ammonia slip,
ammonia storage and ammonia supply. The concept involves the complex mechanisms
taking place within the burning bed of waste. The following diagram illustrates the
different combustion zones.
r
63I
J. Swithenbank et al.
Figure 6.
In Figure 6 can be seen that
to C02 & H 2 0
the waste is progressively
dried, pyrolysed, oxidised,
then enters into the important
reducing region. Notice that
there is a thin oxidising layer
undemeath the reducing
region. This thin region
consumes all the oxygen
whilst releasing carbon
dioxide. In the reducing
region, the charred waste bums by converting the carbon dioxide to carbon monoxide
and the H20to CO and H2.Notice that a region can exist where all the C02 has been
consumed. As the bed thins fbther along, the reducing layer graduaIly vanishes and
eventually only ash remains. This situation is further complicatedby movements of the
grate which are designed to mix the bed in the vertical direction whilst slowly
transporting the waste towards the ash pit. The total transit time of the waste along the
grate is typically about 40 minutes.
Figure 7.
It is apparent that the gases emitted fiom the bed will vary significantly depending on
the axial position along the
bed. These gases vary &om
excess oxygen, to fuel gases
consisting of hydrocarbon
pyrolysis products carbon
CO & H2 burn above the bed
monoxide and hydrogen.
with secondary air
Considering the reducing
region of the bed, it is
proposed that this can be
used to reduce the N4(
formed fiom nitxogen in the
fuel to nitrogen gas.
Equations
for
these
reactions are given in
Reference 11. These processes are illustrated in the following diagram, Figure 7.
Very little information could be located in the literature on the reduction of NOx in a
bed of char. However Reference 12 describes some relevant experiments in which air
was passed though a buming bed of simulated waste. The results indicated that the
NO, was least when the air was passed through the bed rather than added above the
bed, and injecting the air well above the bed also minimisedthe NO, This data can be
interpreted as confirming the concept of NO, reduction within the bed, and NO,
formation when overfire air reacts with waste at the top of the bed The same reference
also reports measurements on a municipal incinerator located at Issy-les-Moulineaux
where a 10% reduction in overfire air injection resulted in a 10% reduction in NO,
632
Fundamental principles of incinerator design
production. Again, this result can be interpreted as reduction of NOx in the reducing
region of the bed.
Freeboard Combustion
Turning next to the processes above the bed. In present practice in incinerators,
secondary air jets are located above the bed to bum out the hydrocarbons and carbon
monoxide. The oxygen level in the flue is typically 8- 10% and this represents a great
waste of energy as sensible heat of the excess air. Here it is proposed that this level of
excess air should be dramatically reduced by the use of recycled flue gases to provide
the mixing energy rather than air jets. First we must consider the needs of the "furnace"
region immediately above the bed. Here the main requirement is to mix the various
gases emitted fiom Merent regions of the bed as explained above. In this case, flue
gases can be used since the gases fiom the bed include both fuel and oxygen, However,
it is important that these "inert" jets do not impinge on the surface since they would
then lead to the removal of many particles fiom the surface, thus increasing both
erosiodcorrosion of the boiler and the load on the particle scrubbing system.
Following the furnace regioq excess air should be injected at the base of the "radiation
shaft" to ensure that the hydrocarbons are completely bumed out. These regions and
concepts are illustrated on Figure 8.
Figure 8.
The injection of air to mix these
gases emerging fiom the bed has
two disadvantages. Firstly, the
supply of excess air will reduce the
efficiency of power generation
since there will be an increase in
the energy carried away in the flue
gas flow. Secondly, if the air jet
impinges on the bed surface, this
could result in burning of the
surface layers of the waste fiom
above with a consequent increase
in the emission of fuel NOx One solution to this problem is to replace some or all of
the air in the jet with recycled flue gases. Furthennore, if the jets impinge on the
surface, there will be an undesirable increase in the entrainment of particles into the
flow.
The criteria goveming the features of the required flow field in the radiation shaft are:0
Little or no recirculation zones since they provide dead space which reduces the
residence time and bum-out of the hot gases.
0
Near uniform velocities minimise the cany-over of particulates by high velocities.
0
Flow nearly parallel to the walls to minimise particle impingement and consequent
erosiodcorrosion.
633
J. Swithenbank et al.
Figure 9.
Turbulence production and dewy behind a
There is a b & m e n M
between the
flame itabilber
completeness of reaction in a jet mixing
'
system and the mixing power available
h m the jets. The analysis is summarised
below and isillustrated in the diagram:The key factor to note is that the jets
produced either by a bafne or by jets
located in a wall produce a turbulent
"stit~ed"reactor followed by a plug flow
Beffle
Plug flow
reactor. This is an optimum combination
reactor
to achieve complete reaction since the
stirred reactor initiates the reaction but -
is dissipated in approximately 15 jet
diameters.
Jet Mixing Theory
Figure 10.
In the case of the jets in an incinerator, it is
interesting to consider the history of the jet
mixing power. Figure 10 shows how this
power flows &om the fan to the turbulence
decay. This energy balance can be expressed
mathematicallyby:-
=t
-
Fan power (Watts) -Pressure and flow
I
1
4
-
-
k t kinetic energy (power) k t velocity
2
kt turbulence kinetic energy (power)
Hence:
Turning now to partially stirred reactor theory; in a partially stirred reactor the oxygen
consumption rate is equal the dif€erence between the rate without diffusion and the
diffusion rate:
k'
k'
= Kinetic time
8 = RTI E = dimensionlesstemp.
C, = Oxygenconc at reactor entry
Tk
634
d'
=Mixing/diffusiontime
c k = Average O2 conc in reacto
C = O2 conc after mixing
T,,
Fundamental principles of incinerator design
Where:
The dif€erence between the entering oxygen concentration and the exit concentration
must be bumed in the stay (residence) time:
2,
k’
T~ =Kinetic
time
8 = RT/E= dimensionless temp.
C, = Oxygen conc. at reactor entry
T~ = Stay (Re sidence) time
C, = Average 0, c0nc.h reactor
C = 0, conc. after mixing
Where:
The rate of mixing is proportional to the concentrationdifference, hence:
dC - C,-C
dz
=D
t, = Diffusion(Mixin& time
Where:
From these three equations,the completenessof
combustion in a partially stirred reactor is:
Ck=AverageO, conc
C = 0, conc after mixing
1
(3 =Fraction of O2 untreated
q, = Combustion efficiency
If the kinetic time is very much less than the mixing time, which is true for combustion
reactions at high temperature, then unmixedness limits the combustion efficiency in a
partially stirred reactor to :1
T~
=
t,/-rd
76
= Ratio of residence time to mixing time
q, = Combustion efficiency
To determine the parameters governing the ratio of residence time to mixing lime, we
conclude that Tsd = .t / ,
’u based on dimensional anaiysis, whilst the value of ‘Fd is
given by X/U where X is the length of the stirred reactor.
635
J. Swithenbank et al.
Now
d , z 0 0 . 2 h and
Thus inserting relations fiom
above, the maximum co~bustiOn Hence rd z 5
efficiency limited by mixing is
approximately:
and is independent of h
1
Thus :
'IIC
-
1
1
1+
5 0 d m
-
tl, =
1
1+
XzlOh
1+
1
1
50 (u'/U)-
1
50,/m
Where the dynamic head loss of
themkingsystem is
E,=APIq
Figure 11.
In an incinerator, the dynamic head of the flow (q) should be determined at the crosssection of the duct into which the jets are being injected, whilst A€' is the pressure loss
across the jet. Due to the approximationsof the analysis, it is clear that these relations
must not be interpreted quantitatively, nevertheless, they do illustrate the parameters
governing the maximum efficiency which can be achieved in the secondary combustion
r,y,-;
,.
01'
size and location of the secondary mhg efficiency factor m = exp[-h/
jets of air and recycled flue gases.
Once a design has been selected, Size of energy containing eddy = CT
the flow
is then calculated by Initial separation of material to be mixed =
the CFD procedures described
previously and illustrated in
References 1 to 10. The location
of the jets must also take into
0.:
account the macro-mixing in order
to ensure that the adjacent micromixing eddies contain the two
0
materials to be mixed rather than
0
OA
0.8
1.2
1.6
2 &
the same material. This is given by
0
0
636
Entrainment theory; - An i s o t h d jet entrains its own original mass flow every
3.2 jet diameters.
Fundamental principles of incinerator design
Partially stirred reactor theory; - The completeness of combustion by a partially
stirred reactor, followed by a plug flow reactor, is limitedby the degree of mixing.
0
Mixing power concept;
The power supplied to the air fan is transfmed
progressively to the jet, then to the turbulence kinetic energy, and finally to the
turbulence dissipation of the Kolmogorov scale eddies which slightly heats the gas.
0
Molecular mixing; The dissipation of kinetic energy between eddies takes place
by the movement of individual molecules, which simultaneouslyresults in mixing
at the molecular level.
0
Macro-mixing; In order that the species in the adjacent Kolmogorov eddies are
merent, thus ensuring good mixing efficiency, the large scale of the energy
containing eddies must be greater than the initial separation of the materials to be
mixed. The dimensions of the energy containing eddies are comparable to the
dimensions of the shear layer at the edge of the jets.
0
Mixing is the product of both micro-scale mixing power and macro-mixing
efficiency.
Quantifying the above parameters provides a means to estimate suitable locations and
sizes for the various jets installed in an incinerator, includingthe secondary air jets.
0
-
-
-
Energy From Waste
Many industrial wastes such as oil sludges have a significant energy content and in
some cases, fuels can be derived from them. One such example, derived from waste
organic chemicals, is Cemfuel which is used by the cement industry. Power is also
generated in some countries from waste sludges such as those produced by the paper
industry.
Municipal solid waste typically has an energy content of about 10,000 Wkg. Thus an
incinerator burning 10 t/hr is releasing 10 x 10,000/3600 M/s = 27.8 MW of heat. If
this heat is used in a steam cycle to generate electricity with 20% efficiency, then
5.56 MW of electricity would be produced. Much of the balance is available as low
grade heat at a temperature of say 120°C for district heating. Fortunately, both the heat
and the power are produced near the consumer hence the transport of both the heat and
the power is convenient.
Since electricity is about 8 times more valuable than the same amount of heat as hot
water, there is interest in maximising the efficiency of power generation fiom waste.
Thermodynamic analysis show that the efficiency of the Rankine steam cycle,
consisting of a boiler, superheater, turbine, condenser and pump, depends on the steam
temperature and pressure, therefore to increase the efficiency, we must increase the
steam temperature.
637
J. Swithenbank et al.
Figure 12.
Unfortunately, due to the corrosive
nature of flue gases, incinerator boilers
are prone to corrosion. Figure 12
shows the ranges of flue gas
temperature
and
boiler
tube
temperature at which corrosion can be
expected.
The steam operating
temperatures and pressures for
Euro ean incinerators are typically
400 C at 40 bars. On the other hand
Japanese incinerators operate at about
300 OC at less than 30 bars. The
reason for this difference stems fiom
the European requirement to generate
energy fkom the waste, although the
extent to which this applies varies
from country to country.
The objective towards which we are now working is to integrate the incinerator and
power production system into the needs of a sustainable city. The incinerator alone can
only provide about 20% of the power needs of the city and the balance must be
provided fiom a power unit which is acceptable in the city environment. Although
this need, our research centres on
options such as fuel cells are being developed to 1the intercooled, regenerated, reheated, gas turbine. By using the high temperature heat
exchanger (up to 1200 OC at high pressure) technology which we have developed
(Reference 13), gas turbines with an efficiency of about 60% are feasible. The
convenient utilisation .of their 'waste ' heat, to heat or cool the surrounding city
buildings raises their overall efficiency to more than 80%. The integration of this CHP
concept with the incinerator provides a practical strategy for the introduction of
combined heat, power and waste management (CHPWM), leading to the sustainable
city.
t
Conclusions
0
0
0
0
0
0
Energy fkom waste is a more environmentallyfriendly technology than landfill.
Incineration technology is maturing rapidly at the present time.
Modeling of the burning bed is being developed.
CFD can be used to model the gas flows but interpretationrequires care.
Mixing theory can be used to guide the design
The practical experience of plant operators must be combined with fimdamental
studies to guide the f b r e of incinerator design and operation.
Nomenclature
Ap
4
4
T,
638
convective heat transfer area
heat transfer coe5cient
conduction within the solid
temperature of the solid phase
Fundamental principles of incinerator design
temperature of the gas phase
volume
volume hction of the gas space
El
P
density
0
volume hction of component N in the solid (VNNs)
Subscripts
1
gas Phase
2
moisture
3
volatiles
4
bound ash
5
fixed carbon
6
fiee ash
7
internal pore space
layer immediately behind a process
LB
layer in front of a process front
LF
solids (includingthe internal pore space)
S
T,
V
References
Three Dimensional Mathematical Modelling Of Sheffield Municipal Solid Waste
Incinerator (35 MW) Using Experimental Data and Computational Fluid
Dynamics, V. Nasserzadeh, J. Swithenbanlc, B. Jones, Jomal of Institute Of
Energy, Vol. 64, September 1991, pp 166-175.
Design Optimisation Of a Large Municipal Solid Waste Incinerator, V.
Nasserzadeh, J. Swithenbank, D. Scott, B. Jones, Journal of Waste Management,
Vol. 11, pp 249-261, 1991.
Three Dimensional Mathematical Modelling of Coventry Municipal Solid Waste
Incinerator (65 Mw) Using Computational Fluid Dynamics and Experimental
Data', V. Nasserzadeh, J. Swithenbank, C. Schofield, D. Scott and A. Loader,
Journal of Process Safety and Environmental Protection, Tramactions of the
Institution of Chemical Engineers, Vol. 71, Part B, November 1993, pp 269 - 279.
Effect of High Speed Secondary Air Jets on the Overall Performance Of a Large
Municipal Incinerator with a Vertical Shaft, V. Nasserzadeh, J. Swithenbank and
B. Jones, Journal of Combustion Science and Technology, 92,4 6, pp 389 - 422,
-
1993.
Emission Testing and Design Optimisation of Sheffield Clinical Incinerator, V.
N a s s d e h , J. Swithenbank, D. Lawrence and N. Garrod, Journal of Process
Safety and Environmental Protection, Part B., Institute of Chemical Engineering,
August 1995, pp 57 71.
Measurement of Gas Residence Times in Large Municipal Incinerators Using the
PRBS Tracer Technique, V. Nasserzadeh, J. Swithenbank., Journal of Institute of
Energy, Vol. 11, September 1995, pp 43 61.
Environmental Advantages and Disadvantages of Modern Landfill versus Waste
Incineration, V. N a s s d e h , J Swithenbank, Paper presented at
INACAP/SOFOFA Conference Chile, April 1995
Chief Inspector's Process Guidance Note: Techniques for Integrated Pollution
Control, Issue Series 2 (S2); Process Sector, Waste Disposal & Recycling, Section
-
-
639
.
I
Swithenbank
.
et al.
-
9
10
11
12
13
640
on Computational Fluid Dynamic Simulation of Incinerators, pp 31 35 V.
N a s s d e h and J. Swithenbank, London: HMSO,December 1996
Control of Waste Incinerators, K. Young, M. Vara and J. Swithenbadc, British
Flame Days, Paper No. 2.4, L e d , UK,September 1994.
A Review of Factors Influencing Boiler Corrosion in Incinerators, J. Swithenbank,
B.C.R Ewan, Conference on Waste Incineration, MTG 95, March 1995,
Copenhagen, Denmark.
An Emissions Model for a Bubbling FBC using Detailed Chemical Kinetics:
Significance of Destruction Reactions, Goel S K, Beer J M and S a r o h A F ,
Journal of the Institute of Energy December 1996 69 pp 201-213.
Nox Emissions of Municipal Solid Waste Incineration; Experiments in a
Counerflow Fixed Bed Reactor and in a Rotary Kiln Incinerator, Joabouille F,
Zhou X, Kerdsuwan S, Bregeon By Goudeau J C, The Third Asian Pacific
I@ntemational Symposium on Combustion and Energy Utilisation, Hong Kong,
(1995)
New Developments in Power Generation; Technologies for a Cleaner
Environment Seminar, Swithenbank J, Smyth R, Lmgston P, Beck S,Cuernavaca,
Mexico, July 1995.(available fiom SUWIC)
Документ
Категория
Без категории
Просмотров
0
Размер файла
1 382 Кб
Теги
incinerator, design, fundamentals, principles
1/--страниц
Пожаловаться на содержимое документа