|
One of the key tests a power station construction project is
subjected to prior to acceptance and hand-over to the owner is
validation of the power station's electrical power output and
thermal efficiency. To determine the thermal efficiency of a power
plant, it is necessary to accurately measure the fuel consumed.
A considerable number of recently constructed large power stations
are gas fired; thus, this article describes some of the factors
that test engineers need to be aware of when gas consumption is
measured by means of orifice flowmeters. It addresses what is
the ultimate aim of any test procedure-minimising as far as practical
the uncertainty inherent in test measurements.
Power Station Testing
The thermal efficiency of a power station is the net electrical
power divided by the heat energy input. The heat energy input
is the product of the gas calorific value and the volume flow
rate measured by the gas flowmeter. The gas calorific value is
obtained from gas samples taken during the power station tests
and analysed by an independent laboratory. The calorific value
is generally given at a standard reference volume, typically 15oC
(59oF) and 1.01325 bara (standard conditions) or 0oC
(32oF) and 1.01325 bara (normal conditions).
A gas fired 800 MW combined cycle gas turbine power plant will
consume in the order of 120,000 Nm3/hour of gas. A precise measurement
of such flows is essential to obtain an accurate measurement of
the power station thermal efficiency. It is also in the construction
contractor's interests to ensure plant performance testing is
carried out as accurately as practical—typical liquidated
damages payable to the owner for reduced efficiency of 0.01 percentage
points compared to the guaranteed value can be of the order of
£50,000.
Orifice Flowmeters
Whilst there are a number of devices used to measure gas flows,
the orifice meter is used extensively. It is relatively inexpensive
and can return high accuracy if constructed and installed correctly.
The orifice meter is a flow obstruction device—a thin plate
with a central circular hole—placed in the fluid flow to
create a pressure drop. Measurement of the pressure drop across
the orifice plate is used to derive the flow.
The volume flowrate at standard reference conditions
Where ÆP = differential pressure measured across the orifice
plate
K = orifice plate constant, given by
K is not a constant for any given size of orifice plate because
the discharge coefficient C is a function of Reynolds Number.
Reynolds Number is a dimensionless property of a fluid and varies
as the flowrate. e is a coefficient that takes into account the
compressibility of a gas contracting and expanding through the
orifice. It is a function of the differential pressure and line
pressure, and is also dependent upon the fluid properties. It
has a value of 1 for non-compressible fluids. Thus, K must be
computed for each flowrate measurement. b is the ratio of the
orifice diameter to the pipeline diameter.
The gas density is obtained from the laboratory analysis of samples
taken during testing.
Orifice plates are used extensively throughout many industries
and internationally recognised standards have been produced to
specify the geometry, flow calculation methodologies and installation
requirements. ISO 5167 "measurement of fluid flow by means
of pressure differential devices" is used widely in Europe.
The standards also specify installation requirements, such as
the required length of straight pipe work upstream and downstream
of the orifice plate. It is important to check that the installation
arrangements comply with the standards, otherwise the uncertainty
of the measurement increased.
Compressibility
Compressibility is a measure of how much a real gas varies from
the relationship between volume, pressure and temperature for
a perfect gas (PV = mRT). This characteristic is required because
the flow is invariably measured at line conditions (temperature
and pressure) that are far removed from standard reference conditions.
Whilst air conforms quite closely to a perfect gas, the behaviour
of natural gas can vary significantly from that ideal. The measure
of the variation is termed the compressibility factor (Z) and
for natural gas it is a function of the constituents of the gas
mixture, the temperature and the pressure. The higher the pressure
and temperature, the more the gas will vary from the perfect gas
relationship. A common industry standard for compressibility factors
for natural gas is according to AGA 8 (American Gas Association
Report No. 8).
Discharge Coefficient
The discharge coefficient of an orifice plate is determined from
empirical formulae based upon many measurements made on various
sizes, fluids and flow velocities.
ISO 5167 contains all of the formulae and data needed to determine
the discharge coefficient for any particular application.
Prior to 1997, ISO 5167 utilised the Stolz equation for deriving
the coefficient of discharge. Following a large international
research programme it was found that in practice there were significant
deviations from the values predicted by the Stolz equation for
C, particularly for large diameter pipes and at high Reynolds
Numbers. This is commonly the condition at which we test large
power plant fuel flowmeters.
Further research to expand the database of C at various pipe sizes
and Reynolds Numbers resulted in the re-issue of ISO 5167 in 1997
modified to calculate the coefficient of discharge according to
the Reader-Harris/Gallagher equation. The new formula, whilst
significantly different form the Stolz equation, is still a function
of the orifice dimensions and Reynolds Number.
Calibration
The instrumentation necessary to measure the flow rate is a pressure
differential transducer across the orifice plate, with temperature
and pressure transducers to measure upstream line conditions.
In common with industry standards and international test codes,
it is a requirement that all instrumentation used for the measurement
of power station thermal efficiency be calibrated at an accredited,
independent test laboratory that has test instrumentation calibrations
traceable to national standards.
It is important to ensure the differential pressure transducer
is calibrated at the expected line pressure. It has been found
that the transducer output for a given pressure difference can
be slightly dissimilar at different absolute pressures.
It is PB Power's philosophy that the complete orifice and pipeline
section should be also calibrated as a whole. Some contractors
query this viewpoint because the standards specify the installation
arrangements and calculation method. There are, however, a number
of reasons why deviations from the standards can occur, which
is why:
-
Care has to be taken that
orifice flange gaskets do not protrude into the flow, otherwise
uncertainty in measurement can be introduced.
-
The orifice should be central
within the pipeline since slight eccentricity can introduce
significant measurement errors.
-
The location of the upstream
and downstream pressure tappings have to be positioned within
close tolerances.
For example, in a recent case the contractor dispatched the
orifice plate section to a calibration facility and the calibration
indicated an unusual result for the measured coefficient of discharge,
significantly at variation to that calculated according to the
ISO standard. At PB's insistence, the calibration was repeated.
During the set-up for the calibration it was noticed that the
orifice flange clamping bolts were interfering with the plate
and causing it to distort, hence the abnormal characteristic for
the coefficient of discharge. This problem could not have been
identified unless the entire section had been sent away for calibration.
The calibration should ideally be performed using natural gas
at a flowrate similar to the expected power station fuel consumption,
i.e. at a similar Reynolds Number. Typical Reynolds Numbers will
be of the order of 7,000,000. Testing on water should not be accepted
as such high Reynolds Numbers cannot be obtained practically.
There are few independent test facilities worldwide able to perform
testing of large gas flowmeters at high pressures and flowrates.
Conclusions
To achieve high accuracy measurement of the gas fuel consumption
of a power station using orifice plate flowmeters it is important
that the installation arrangement complies with the appropriate
test standard in all respects (ISO 5167: 1997). Appropriate account
should be taken of the compressibility characteristics of the
gas; and all components of the orifice plate flow meter, including
the orifice pipeline section, should be calibrated at an approved
independent test facility on gas at the expected line pressure
and Reynolds Number just prior to power plant testing.
If these recommendations are followed, it should be possible to
determine the gas fuel flow rate within an uncertainty of 0.5%,
or better. Such accuracy will make a major contribution towards
keeping the overall test uncertainty within the limits specified
by the standards for power plant testing, such as ASME PTC46.
|
