 |
Description  |
|
|
TECHNICAL FIELD
This invention relates to the measurement of the unburnt carbon content of
fly ash produced by a coal fired boiler.
BACKGROUND ART
In the combustion of pulverised coal for steam generation in coal-fired
power stations there are certain fixed losses determined for example, by
plant design, and certain controllable losses caused by operating under
non-ideal conditions. The controllable losses comprise:
(a) losses due to incomplete combustion of both solids and combustible
gases;
(b) losses due to the need for excess air.
In practice the controllable losses show a minimum as a function of oxygen
in the flue gas and it is preferable to operate near this minimum. One way
this can be achieved is by basing control of the boiler on the measurement
of oxygen and carbon monoxide in flue gas. Most large boilers today are
equipped with oxygen analysers which measure O.sub.2 at one point in a
duct. A problem with these analysers is that the reading is drastically
distorted by air infiltration into the furnace and in the convection
passages downstream of the burners. Also, as measurements are made at one
point, sampling errors are large.
Carbon monoxide in flue gas stays at very low levels at high excess air and
rises as excess air is reduced. Infrared CO analysers are available which
direct the IR beam across the stack, thus minimising sampling errors.
However, optimising excess air using CO monitors generally produces a
large amount of unburnt carbon in the ash, because CO levels are very low
at optimum excess air.
An alternative technique is to base control of the boiler on the
determination of unburnt carbon in the fly ash. A 500 MW power station
burning black coal of 20% ash will produce about 2500 tonnes/hr flue gas,
and 37 tonnes/hr fly ash. The carbon content of this fly ash will be
normally in the range 2-5 wt % although it may contain up to 15 wt %
carbon. Typically the fly ash concentration in flue gas is about 20
g/m.sup.3. Present instruments for the determination of the carbon content
of the fly ash rely on extracting a sample, typically less than 1 gram,
from the duct and analysing this on a batch basis typically at 10-20
minute intervals.
One prior art carbon concentration monitor [Rupprecht and Patashnick Co.,
Inc, NYSERDA Report 86-2, January 1986] is based on a microbalance and
small furnace. The instrument collects a 10-50 mg sample of fly ash from
the outlet duct of a boiler and determines the unburnt carbon in this
sample from the mass loss after heating at 750.degree. C., this
measurement cycle being repeated at approximately 15 minute intervals. One
disadvantage of this analysis technique is that it is very difficult to
collect a representative sample of such small size, and therefore sampling
uncertainty significantly limits the accuracy of the unburnt carbon
determination. The analysis accuracy for replicate samples in laboratory
tests was approximately .+-.0.5 wt % at 2.3 wt % carbon.
Another commercially available device [Energy and Environmental Research
Corporation, 18 Mason, Irvine, Calif., USA; December 1987] for the
determination of unburnt carbon in fly ash collects an approximately 1
gram sample from the duct using an isokinetic sampler and analyses this
for unburnt carbon content from the measured surface reflectance of the
sample. The sample collection and measurement cycle is repeated at
approximately 5 minute intervals. In a plant test of the instrument at the
Nefo power plant, Denmark, the analysis accuracy was approximately .+-.1
wt % at less than 3 wt % carbon and .+-.0.5 wt % at greater than 3 wt %
carbon. The analysis accuracy is limited by sampling uncertainty, due to
the sample size and measuring principle (i.e. surface reflectance) used,
and the sensitivity of the reflectance measurement to coal type.
A device based on a measurement of the capacitance of a fly ash filled
capacitor has been proposed for the determination of carbon in fly ash in
Australian Patent 562440. In this arrangement ash is taken from an ash
hopper using a screw conveyor, fed into a measuring chamber into the
electric field established by the electrodes of a capacitor and the change
in capacitance of the capacitor measured, and finally returned to the ash
hopper using a second screw conveyor. The bulk density of the ash in the
measuring chamber is assumed to be approximately constant, although
compensation for variation in the bulk density is possible using a
weighing device.
A microwave technique has been proposed for simultaneously reducing and
measuring the carbon content in fly ash in U.S. Pat. No. 4,705,409. In
this technique ash is taken from an ash hopper and passed through a
metallic waveguide. Microwave radiation directed through the guide is
preferentially absorbed by the carbon in the fly ash, and the
concentration of carbon is determined from measuring the temperature rise
of a water wall surrounding the guide. Sufficient microwave power is
injected into the guide to burn the excess carbon in the ash and generate
a reduced carbon product. One disadvantage of this technique is that the
heat conduction out of the guide, and the associated temperature rise in
the water wall, is a function of not only the carbon content of the ash
but also the chemical characteristics, temperature and heat conduction
properties of the ash. These factors need to be taken into account in the
calibration and operation of the device.
Nuclear measurement of carbon in fly ash has also been investigated
[Steward, R. F., ISA Transactions, (3), 1967, 200-207]. In this technique
carbon concentration is correlated with counts of 4.43 MeV gamma rays
produced from carbon atoms by the inelastic scatter of neutrons. Using
this technique in laboratory measurements on 10 kg fly ash samples the
analysis accuracy is repeated as .+-.0.5 wt % over the range 2-16 wt %
carbon.
DISCLOSURE OF THE INVENTION
It is an object of this invention to provide a method and apparatus to
measure the unburnt carbon content in fly ash.
Accordingly, in one aspect this invention consists in an apparatus to
measure the unburnt carbon content of fly ash comprising means to generate
a microwave signal, transmitter means to launch said microwave signal for
transmission through a fly ash sample, receiver means to receive a signal
passed through the sample and processing means to determine the
attenuation or phase shift of the signal passed through the sample with
respect to the launched signal and to produce a measure of unburnt carbon
content.
In a second aspect this invention consists in an apparatus to measure the
unburnt carbon content of fly ash comprising means to generate a microwave
signal, antennae means to launch a microwave signal into a fly ash sample
and to receive a reflected signal and processing means to determine the
attenuation or phase shift of the reflected signal with respect to the
launched signal and to produce a measure of unburnt carbon content.
In a third aspect this invention consists in a method of measuring the
unburnt carbon content of fly ash comprising the steps of launching a
microwave signal into a fly ash sample, receiving the transmitted signal,
determining the attenuation or phase shift of the received signal with
respect to the launched signal and producing a measure of unburnt carbon
content from said attenuation or phase shift.
In a fourth aspect this invention consists in a method of measuring the
unburnt carbon content of fly ash comprising the steps of launching a
microwave signal into a fly ash sample, receiving a component of the
signal reflected from the sample, determining the attenuation or phase
shift of the reflected signal with respect to the launched signal and
producing a measure of unburnt carbon content from said attenuation or
phase shift.
In one preferred form of the invention separate microwave transmitters and
receivers are used. These are provided with suitable antennae, for
example, horns or microstrip radiators in an open system, and capacitative
or inductive probes in waveguides.
In another preferred form of the invention a single transceiver is used for
transmitting and receiving. This arrangement is particularly advantageous
where a reflected signal is measured but can also be used where a signal
transmitted through the sample is measured by utilising a suitable
microwave reflector and effecting a double pass of the sample.
The microwave signal can be generated using any suitable microwave
oscillator. Preferably the frequency of the microwave signal is in the
range of from 1 to 20 GHz.
Although the attenuation of the transmitted or reflected microwave signals
has been found to provide a useful measurement of unburnt carbon in fly
ash it is presently preferred to use the change in the characteristics of
a microwave resonant cavity induced by the presence of a fly ash sample to
produce a measure of unburnt carbon.
Accordingly, it is preferred that a measurement chamber in the form of a
microwave resonant cavity receives a fly ash sample and the processing
means determines from the attenuation or phase shift of the received
signal with respect to the launched signal the change in the resonant
cavity characteristics induced by the fly ash sample and produces
therefrom a measure of unburnt carbon content.
The resonant cavity characteristics determined from the attenuation or
phase shift are preferably resonant frequency, transmitted or reflected
power at the resonant frequency, and Q-factor. These are preferably
determined from a swept frequency measurement. The presently preferred
technique utilises a swept frequency measurement of attenuation.
In a preferred technique two microwave resonant cavities are utilised. The
fly ash sample is placed in or passed through one microwave resonant
cavity and the other provides a reference measurement. In a further
preferred technique a single microwave resonant cavity is used to provide
both a reference measurement (made when the cavity does not contain the
sample) and subsequent measurements when the cavity contains the sample.
The methods and apparatus of this invention can be used to measure unburnt
carbon content of collected fly ash samples or of a fly ash sample
entrained in the flue gas from a coal fired boiler.
It will be apparent that the method and apparatus of this invention have
several advantages over the prior art. The measurements according to this
invention are non-destructive and require no special sample preparation.
The microwave measurement can be completed almost instantaneously and
therefore a continuous measurement of unburnt carbon content can be
provided. Further, the method and apparatus of this invention are not
limited by sample size and can be used with samples varying from a few
grams to tens of kilograms. The ability to analyse large samples allows
sampling uncertainty to be reduced and enables improved measurement
accuracy. The method and apparatus are also applicable to both collected
samples and in situ measurement.
BRIEF DESCRIPTION OF THE DRAWINGS
This invention will now be described, by way of example only, with
reference to the accompanying drawings in which:
FIG. 1 is a schematic block diagram of an apparatus to measure unburnt
carbon in fly ash according to a first embodiment of this invention;
FIG. 2 is a schematic diagram of the antennae and sample measurement
chamber in FIG. 1 for measurement in free space;
FIG. 3 is a schematic diagram of the antennae and sample measurement
chamber in FIG. 1 for measurement in a waveguide;
FIG. 4 is a schematic diagram of the antennae and sample measurement
chamber in FIG. 1 for measurement in a microwave resonant cavity;
FIG. 5 is a schematic diagram of a further apparatus to measure unburnt
carbon content of fly ash according to this invention.
FIG. 6 is a graph showing correlation of (A/W) with wt % carbon for
measurement in a waveguide;
FIG. 7 is a graph showing correlation of (.phi./W) with wt % carbon for
measurement in a waveguide;
FIG. 8 is a graph showing correlation of (A/W) with wt % carbon for
measurement in free space;
FIG. 9 is a graph showing correlation of (.phi./W) with wt % carbon for
measurement in free space; and
FIG. 10 is a graph showing correlation of change in resonant frequency with
wt % carbon for measurements in a resonant cavity; and
FIG. 11 is a graph showing correlation of (change in 1/Q) x f.sub.r with wt
% carbon for measurements in a resonant cavity.
MODES FOR CARRYING OUR THE INVENTION
The propagation of an electromagnetic wave (EM) in a dielectric medium is
described by Maxwell's equations,and the complex amplitude given by
E(1)=E.sub.o exp (-.gamma.1) (1)
where 1 is the distance travelled by the EM wave in the dielectric medium
from some reference point where its amplitude was E.sub.o, and .gamma. is
the propagation constant of the wave given by
.gamma.=.alpha.+j.beta. (2)
where .alpha. and .beta. are the attenuation and phase constants
respectively. For a non-magnetic dielectric medium .alpha. and .beta. are
given by
##EQU1##
where .epsilon..sub.o is permittivity of free space, .lambda..sub.o the
wavelength is free space, .epsilon.' the dielectric constant of the medium
and .epsilon." the loss factor of the medium. The attenuation constant
.alpha. represents the attenuation of the EM wave (in nepers per meter)
and the phase constant .beta. represents the phase shift of the EM wave
(in radians per meter).
From equations (3) and (4), it can be seen that the attenuation and phase
shift of an EM wave in a dielectric is a function of the complex
permittivity of the medium,
.epsilon.*=.epsilon.'-j.epsilon." (5)
For a multicomponent dielectric medium the complex permittivity may be
approximated by
##EQU2##
where v.sub.i and .epsilon.*.sub.i are the volume fraction and complex
permittivity of the i.sup.th component respectively.
When a plane EM wave is incident upon a dielectric interface, part of it is
reflected and part transmitted. For a non-magnetic dielectric in air the
reflection coefficient, R, and transmission coefficient, T, are given by
##EQU3##
where E.sub.o, E.sub.R and E.sub.T are the incident, reflected and
transmitted electric field vectors. From equations (3) and (4) it can be
seen that the phase shift and attenuation of a transmitted microwave
signal are functions of the effective complex permittivity of the sample
given by equation (6). For fly ash the complex permittivity of the unburnt
carbon is significantly different from the remaining matrix which
principally comprises oxides of silicon, aluminium and iron. Therefore the
measured attenuation and phase shift for fly ash are strong functions of
the unburnt carbon cshift of a reflected signal are therefore also
functions of the unburnt carbon content of the samples.
For a cylindrical microwave resonant cavity the resonant frequency of the
microwave cavity, f, can be calculated from,
##EQU4##
where `n,m,l` refer to the particular resonant mode (and correspond to the
number of electric field maxima in the standing wave pattern .phi., r and
z directions), `a` and `d` are the cavity radius and length respectively
and .rho. is a constant determined for each resonant mode. For a
TM.sub.010 resonant cavity, equation (1) reduces to
##EQU5##
When a sample with permittivity .epsilon.*=.epsilon.'-j.epsilon." is
placed about the axis of a TM.sub.010 cavity, and the sample radius, r<<a,
it is found that, the change in the resonant frequency, .DELTA.f, and
Q-factor, .DELTA. 1/Q , are related to the dielectric properties of the
sample by,
##EQU6##
where V.sub.S is the volume fraction of the cavity filled by the sample.
Therefore for a constant volume sample, .DELTA.f/f is proportional to
.epsilon.' and .DELTA. 1/Q is proportional to .epsilon.". It follows that
for measurements on fly ash in such a cavity, .DELTA.f/f and .DELTA. 1/Q
are both strong functions of the weight percent unburnt carbon in the fly
ash.
In the method for determining unburnt carbon content of fly ash according
to one aspect of this invention a microwave signal is directed through a
fly ash sample using suitable transmitting and receiving antennae and the
attenuation and phase shift of the signal due to the fly ash sample are
measured. These are normally calculated as the difference between the
attenuation and phase shift determined with the sample and air. To
compensate for variation in the density and thickness of the fly ash
sample the phase shift and attenuation can be normalised to a unit sample
mass per unit area. This is not necessary where the variation in sample
density and thickness can be maintained within acceptable limits by a
suitable sample presentation system.
To obtain a measure of unburnt carbon content in terms of weight percent
(wt %) the attenuation or phase shift data are correlated with wt %
unburnt carbon, determined by standard laboratory analysis, using least
squares regression and equations of the form:
wt % unburnt carbon=a.sub.0 +a.sub.1 (.phi..sub.c) (13)
wt % unburnt carbon=b.sub.0 +b.sub.1 (A.sub.c) (14)
where .phi..sub.c and A.sub.c are the corrected (compensated for variation
in sample density and thickness) phase shift and attenuation respectively,
and a.sub.0, . . , b.sub.1 are fitting constants. The unburnt carbon
content may also be determined from a combined measurement of attenuation
and phase shift, independent of variation in sample density and thickness,
using an equation of the form
wt % unburnt carbon=C.sub.0 +C.sub.1 (.phi..sub.m)+C.sub.2 (A.sub.m)(15)
where .phi..sub.m and A.sub.m are the measured phase shift and attenuation
respectively, and C.sub.0, . . , C.sub.2 are fitting constants.
In the method for determining unburnt carbon content of fly ash according
to another aspect of the invention a microwave signal is directed at a fly
ash sample and the reflected signal detected. Either a transceiver or
separate transmitting and receiving antennae can be used for transmitting
and receiving the microwave signal. As with the transmission method the
attenuation and phase shift of the reflected signal are measured and
preferably are correlated with wt % unburnt carbon using least squares
regression and equations of the same form as (13), (14) and (15).
FIG. 1 schematically shows the arrangement of the apparatus to measure
unburnt carbon content of fly ash according to this invention. As shown
the apparatus comprises a microwave source which takes the form of a
Yttrium-Iron-Garnet oscillator 1 tuneable over the range 2 to 4 GHz and
controlled by a data logging computer 2. The output of oscillator 1 is
modulated by a PIN diode modulator 3 and directed through a low pass
filter 4 to a power divider 5. Power divider 5 diverts a small amount of
the microwave signal to an 8-port junction 6 as a reference signal. The
remainder of the microwave signal is directed via a circulator 7 to a
transmitter antenna 8. Circulator 7 is provided to direct any reflected
signal to an appropriate instrumentation amplifier 9 to provide a
measurement signal for computer 2. Transmitter antenna 8 directs the
microwave signal through a sample measurement chamber 10 to a receiver
antenna 11 from which the received signal is directed to 8-port junction 6
and instrumentation amplifiers 9 to provide a measure of the attenuation
and phaseshift of the received signal in the known manner. This data is
transmitted for processing in the manner described herein.
The microwave antennae can be of any type suitable to the selected sample
presentation technique. FIGS. 2 to 4 show three preferred arrangements of
the antennae and sample measurement chamber.
Referring to FIG. 2, an arrangement for measuring an ash sample in free
space is provided. The antennae are horn antennae 12, 13 and the ash
sample 14 is contained in a container 15 formed of a material such as wood
or plastic which allows the transmission of microwaves. In this
arrangement the ash sample 14 is packed in container 15 and suitably
positioned between horns 12, 13. The phase shift and attenuation are
determined as described above and used to calculate the wt % of unburnt
carbon as described above.
FIG. 3 shows an arrangement for measurement on sample in a waveguide. In
this arrangement the antennae are capacitive posts or inductive loops 16,
17. The sample 14 to be measured is packed into a section of waveguide 18
of circular or rectangular cross section suited to the frequency range of
the microwave signal. For measurements in the 2.6 to 3.95 GHz frequency
range an RG-48 rectangular waveguide can be used. The sample is confined
to the selected region of the waveguide by plastic sheets 19 which allow
transmission of the microwave signal. The phase shift and attenuation are
determined as described above and used to calculate the wt % of unburnt
carbon as described above.
FIG. 4 shows an arrangement for measurement on a sample in a microwave
resonant cavity. In this arrangement the ash sample is contained in a
non-conducting, for example, ceramic or plastic tube 20 located along the
axis of a TE or TM mode resonant cavity 21. The microwave signal is
coupled in and out of the resonant cavity using H-field (inductive loop)
probes 22, 23.
FIG. 5 shows another arrangement for the measurement of the unburnt carbon
content of a fly ash sample. A variable frequency microwave oscillator 30
provides a microwave signal to a microwave power divider 31. Power divider
31 produces two output signals which are respectively directed to a
reference microwave resonant cavity 32 and a measurement microwave
resonant cavity 33. The fly ash sample (not shown) is placed in or
appropriately passed through the measurement cavity 33. Detectors 34 and
35 respectively measure the attenuation of the microwave signal
respectively propogated in the reference cavity and measurement cavity.
Detectors 34 and 35 can be of any suitable known type such as diode
detectors. The outputs of detectors 34, 35 are fed to a processor 36 which
is used to determine a measure of the resonant frequency, transmitted
power at the resonant frequency, and Q-factor of both cavities from the
swept frequency response (i.e. attenuation) of the received signal. A
measure of weight percent unburnt carbon can then be provided by the
processor as explained by the following.
If the resonant frequency and Q-factor of the reference cavity are f.sub.r
and Q.sub.r respectively, and the resonant frequency and Q-factor of the
measurement cavity are f.sub.m and Q.sub.m respectively, then the weight
percent unburnt carbon in the fly ash is determined of the sample bulk
density from a function of the form
##EQU7##
where,.DELTA.f=f.sub.r -f.sub.m (17)
##EQU8##
Typically
##EQU9##
is a correlation function of the form
##EQU10##
where a.sub.o, a.sub.1,a.sub.2 . . . . are fitting constants or,
##EQU11##
where b.sub.o, b.sub.1,b.sub.2 . . . . are fitting constants
The significant advantages of this arrangement compared to that using a
single measurement in a microwave resonator are that the measured .DELTA.f
and .DELTA.(1/Q) are effectively independent of drifts in the microwave
oscillator output frequency due to ambient temperature variations or drift
in the oscillator control voltage. This is a consequence of the
measurement period of the frequency sweep being much less than the period
over which such drifts normally occur. Therefore using this technique high
measurement accuracy can be achieved without the need for a highly
stabilized microwave source or electronics. This enables .DELTA.f to be
determined from measurement of .DELTA.V, the difference in the control
voltage of the microwave oscillator at f.sub.m and f.sub.r, rather than
from the more difficult and expensive technique of using a microwave
frequency counter.
The measured .DELTA.f and .DELTA.(1/Q) are independent of temperature drift
in the microwave detectors, as such drift only effects the amplitude of
the detected microwave signal. If the reference and measurement cavities
are substantially similar in design and dimension the measured .DELTA.f
and .DELTA.(1/Q) are also independent of drifts in the resonant frequency
of the cavities due to metal expansion with ambient temperature
variations. In this arrangement it is desirable to place a standard
absorber in the reference cavity such that f.sub.r is just greater than
the maximum f.sub.m that occurs in the particular measurement application.
In this case the swept frequency range, .DELTA.f, is minimised.
When the method described above is performed using a single microwave
cavity a reference measurement is made when the cavity does not contain
the sample. Preferably the period between reference measurements is
substantially shorter than oscillator, electronic and temperature drifts.
The apparatus described with reference to FIGS. 1 and 2, FIGS. 1 and 3, and
FIGS. 1 and 4 respectively were used to perform measurements on a range of
fly ash samples from New South Wales and Queensland power stations. The
unburnt carbon content of these samples was determined by standard
chemical analysis using LECO analyser and was in the range 0.5 to 13 wt %.
For measurement, in free space and in waveguides the samples were packed
in an open container to a depth of approximately 100 mm, and in a 200 mm
length of RG-48 waveguide section respectively, and the phase shift and
attenuation of a 3.3 GHz microwave signal determined. The data were
correlated with wt % carbon using the equations,
wt % carbon=a.sub.o +a.sub.1 (.phi..sub.fly ash /w) (21)
wt % carbon=b.sub.o +b.sub.1 (A.sub.fly ash /w) (22)
where a.sub.o , . . , b.sub.1 are fitting constant, w is sample mass per
unit area (in g cm.sup.-2) and .phi..sub.fly ash and A.sub.fly ash are the
phase shift (in degrees) and attenuation (in dB) of the fly ash sample
respectively.
The apparatus described with reference to FIG. 5 was also used to perform
measurements on one of the fly ash samples. In this case the data were
correlated with wt % carbon using equation 19.
R.m.s. errors from correlations on the data using equations (12) and (13)
are given below in Table 1.
TABLE 1
______________________________________
R.m.s. (wt %
Unburnt Measure- Error Carbon)
Equa-
Power Carbon ment Equation
Equation
tion
Station (Wt %) Geometry (21) (22) (19)
______________________________________
Waller- 3-13 Free space
0.41 1.41 --
awang Waveguide 0.28 1.22 --
Swanbank
0.5-5 Free space
0.17 0.83 --
Waveguide 0.22 0.70 --
Resonator -- -- 0.34
Eraring 0.5-2.5 Waveguide 0.19 0.29 --
______________________________________
Plots of the data for Swanbank fly ash samples are presented in FIGS. 6 and
7 for measurements in waveguide and FIGS. 8 and 9 for measurements in free
space and FIGS. 10 and 11 for measurements in a resonator. The r.m.s.
errors in Table 1 represent the total analysis error due to gauge
inaccuracy, sampling and chemical analysis. These results indicate that a
measurement of phase shift or the resonator characteristics is the most
accurate for the determination of carbon content, and the accuracy of
analysis is comparable to or better than that obtained with previous
methods.
The apparatus described above is particularly suitable for on-line analysis
of the unburnt carbon content of fly ash sampled from a boiler outlet
duct. Fly ash is removed from the boiler outlet duct by conventional
sampling means (not shown), for example using a Cegrit sample and cyclone,
and passed through the sample measurement chamber of the apparatus. The
fly ash can be fed continuously or in batches, and carried to and from the
measurement chamber by any suitable means, for example by a screw
conveyor.
The foregoing describes the invention with reference to some specific
examples and it will be apparent to those skilled in the art that
modifications can be made without departing from the scope of the
invention.
* * * * *
|
|
 |
|
 |
Description |
|
