Note: Descriptions are shown in the official language in which they were submitted.
CA 02217614 1997-10-08
Method and device for determining the flow velocity and/or
throughput of a flowing fluid
The invention relates to a method for determining the
flow velocity and/or throughput of a flowing fluid in a
channel, by transmitting and receiving ultrasonic sound
waves along paths between transducers which are disposed
along the periphery of the channel and can act as trans-
mitter and receiver, measuring the transit times of the
sound waves transmitted in opposite directions along a path,
and determining the difference between the transit times
thereof and calculating the flow velocity and/or throughput
using multiplication factors, at least one path being
situated between two transducers positioned at a distance
from each other in the direction of flow of the fluid.
Such a method is known from, for example, Applicant's
European Patent Application 0,639,776. This known method is
based on the recognition of certain characteristics of the
flow profile - symmetry, swirl and/or pulsating flow - and
on the basis thereof automatically adjusting the
multiplication factors, in order to calculate the throughput
and/or flow velocity from the measured transit times. In a
preferred embodiment of this known method pairs of
transducers are used, said transducers being disposed in
such a way that the paths thereof comprise three paths
through the centre of the channel with single reflection
against the wall and two paths according to an inscribed
triangle.
For the carrying out of measurements, such as those in
the case of gas producers, the requirements as regards
accuracy and reliability of the measured results are
becoming increasingly strict. One of the causes of the
"inaccuracy" and "unreliability" - insofar as such terms can
be used - of the results obtained by known methods is the
residual sensitivity to different flow profiles, which is
caused, inter alia, by the position of the acoustic paths,
CA 02217614 1997-10-08
- 2 -
and thus of the assigned multiplication factors.
One object of the invention is to provide a measuring
method in which the basic sensitivity to axially symmetrical
f low patterns is reduced and the f low prof i 1e need not be
recognized.
A further object of the invention is to provide such
a measuring method on the basis of a suitable choice of the
arrangement of the transducers, and thus of the paths.
Yet another object of the invention is to improve the
accuracy and reliability of the measuring method, and
consequently of the measured results obtained by it.
This object is achieved according to the invention by
a method of the abovementioned type, in which the paths
comprise at least one first path through the centre of the
channel, and at least one second path in the form of an
inscribed triangle, and at least one third path with three
or more reflections against the wall of the channel.
A path, also called an acoustic path below, is the
route between the transducers concerned, irrespective of the
direction in which the transmitted sound waves pass along
said path.
In the case of the method according to the invention,
sound waves are therefore transmitted along at least three
different acoustic paths, a path through the centre of the
channel, a path in the form of an inscribed triangle, and a
path with more than two reflections against the wall of the
channel. Along each path the sound waves are transmitted in
both directions, so that the flow velocity and/or the
throughput can be calculated by means of the difference in
transit time of the sound waves transmitted in opposite
directions.
By means of this combination of acoustic paths, the
mathematical multiplication factors to be assigned can be
made equal for all kinds of axially symmetrical flow
profiles, irrespective of the prevailing flow profile.
In practice, if it is considered necessary, a
combination of the method according to this invention with
CA 02217614 1997-10-08
- 3 -
the method described above according to EP-A-0,639,776 can
be used.
As will emerge below, the results obtained by the
method according to the invention are better, and thus more
reliable, than the results achieved until now.
Preferred embodiments of the method according to the
invention are defined in the subclaims. The first path
through the centre of the channel preferably comprises one
reflection against the channel wall, because such a path is
insensitive to swirl. Moreover, the third path with the most
reflections is preferably an inscribed square, so that the
total path length has little effect on the accuracy of the
measurement, which can be reduced as a result of absorption
of the sound waves by the medium.
The invention also relates to a device for determining
the flow velocity and/or throughput of a flowing fluid in a
channel, which device comprises pairs of transducers which
are disposed along the periphery of the channel and can act
as transmitter and receiver, of which at least one pair of
transducers are situated at a distance from each other in
the direction of flow of the fluid, and also comprises means
connected to the transducers for determining the flow
velocity of the fluid in the channel from the corresponding
transit times of the acoustic waves between the pairs of
transducers. Such a device is also known in the art, for
example from the abovementioned European Patent Publication
0,639,776.
The device according to the invention is characterized
in that the pairs of transducers are disposed in such a way
that they comprise at least one first path through the
centre of the channel, and at least one second path in the
form of an inscribed triangle, and at least one third path
with three or more reflections against the wall of the
channel.
Preferred embodiments of the device according to the
invention are described in subclaims 6 and 7.
The invention will be illustrated in greater detail
CA 02217614 1997-10-08
- 4 -
below with reference to the appended drawing, in which:
Figure 1 is a graph of the weighting factors for
various acoustic paths;
Figure 2 is a graph of the weighting factors for an
ideal acoustic path compared with that for a combination of
acoustic paths according to the prior art;
Figure 3 is a graph of the residual error of the
combination according to Figure 2;
Figure 4 is a diagrammatic illustration of a preferred
embodiment of configurations of transducers and acoustic
paths in a channel which is used in the case of the present
invention;
Figure 5 is a graph of the weighting factors and
residual error for the combination of acoustic paths
according to Figure 4.
It is pointed out that the term weighting factor is
intended to convey the geometrical weighting factor, in
order to distinguish it from the arithmetical multiplication
factors which are used as correction factors when
calculating the velocities and/or the throughput of the
medium.
The invention is based on the following principles.
For determining the flow through a cylindrical channel the
surface area of a cross-section through the channel is
divided into a series of concentric rings. Each ring has a
surface area of 2~rr.8r, 8r being the width of the concentric
ring. The surface area of a ring therefore increases in
direct proportion to the value of r. See curve 1 in Figure
1, which shows the relative contribution of a ring as a
function of the radius.
For an axially symmetrical flow the volume Q flowing
through the pipe can be found by adding or integrating the
velocity V(r), multiplied by the surface area of the
corresponding concentric ring, or in formula form: Q - f
V(r).2~rr.8r. The magnitude of the velocity V(r) is
determined by the ultrasonic measuring device. The
multiplication factor to be assigned depends on the position
CA 02217614 1997-10-08
- 5 -
of the acoustic path. In addition to the ideal weighting
curve 1 to be achieved, Fig. 1 shows the weighting factors
for three different paths. In the case of an acoustic path
through the centre of the pipe, each segment with a width
frequency has the same weighting factor, as indicated by
curve 2 in this figure. Curve 3 shows the weighting factor
of an acoustic path in the form of an inscribed triangle
(mid-radius path with double reflection). Curve 4
corresponds to the weighting factors of a path in the form
of an inscribed square.
Fig. 2 shows the ideal weighting curve (1) and the
composite weighting curve (+) of a combination of a single
ref lection path with a double ref lection path, as known from
the prior art. Fig. 3 shows the relative measuring error (*)
thereof (weighting curve minus ideal weighting curve) as a
function of the position. It can be seen from this that
weighting errors occur, varying from approximately -0.02 to
+0.054. Such weighting errors in turn lead to measuring
errors if the actual flow profile differs from the flow
profile which was present during calibration or from what
was expected.
Figure 4 shows a preferred embodiment of the
arrangement of transducers and thus acoustic paths which is
used in the case of the method and device according to the
invention. Pairs of transducers are disposed along the
periphery of a channel 1. A first pair of transducers 2a, 2b
is situated at a distance from each other in the lengthwise
direction of the channel . These transducers define a first
acoustic path A with a single reflection against the wall of
the channel 1. In a comparable manner, a second pair of
transducers 3a, 3b is disposed in such a way that the second
acoustic path B is in the form of an inscribed triangle. A
third pair of transducers 4a, 4b is disposed in such a way
that the third acoustic path C thereof is in the form of an
inscribed square.
With the combination of acoustic paths according to
the invention, a curve of composite weighting factors coming
CA 02217614 1997-10-08
- 6 -
as close as possible to the ideal weighting curve is now
obtained.
Figure 5 shows the weighting curve (x) of this
combination, and the residual error thereof, indicated by ~,
relative to the ideal weighting curve (not shown). As can be
seen from this figure, the error in this case remains within
the -0.02 and +0.02 range, which means a considerable
improvement compared with the known combination. Another
advantage is that the peaks in the residual error are
distributed better over the space (cf. the single peak in
curve 2 of Fig. 3 with the double peak in the residual error
according to Fig. 5). The sensitivity to a local profile
disturbance is reduced as a result.
With the measuring technology existing until now, an
inaccuracy of approximately 0.5% can be achieved, caused by
(metastable) variations in the flow profile. In the case of
the measuring method according to the present invention,
this can at least be reduced to about 0.15%.
