 |
| Titel |
Adaptive method for quantifying uncertainty in discharge measurements using velocity-area method. |
| VerfasserIn |
Aurélien Despax, Anne-Catherine Favre, Arnaud Belleville |
| Konferenz |
EGU General Assembly 2015
|
| Medientyp |
Artikel
|
| Sprache |
Englisch
|
| Digitales Dokument |
PDF |
| Erschienen |
In: GRA - Volume 17 (2015) |
| Datensatznummer |
250102571
|
| Publikation (Nr.) |
 EGU/EGU2015-1900.pdf |
|
|
|
|
|
| Zusammenfassung |
| Streamflow information provided by hydrometric services such as EDF-DTG allow real time
monitoring of rivers, streamflow forecasting, paramount hydrological studies and engineering
design.
In open channels, the traditional approach to measure flow uses a rating curve, which is
an indirect method to estimate the discharge in rivers based on water level and punctual
discharge measurements. A large proportion of these discharge measurements are performed
using the velocity-area method; it consists in integrating flow velocities and depths through
the cross-section [1]. The velocity field is estimated by choosing a number m of verticals,
distributed across the river, where vertical velocity profile is sampled by a current-meter at ni
different depths. Uncertainties coming from several sources are related to the measurement
process.
To date, the framework for assessing uncertainty in velocity-area discharge
measurements is the method presented in the ISO 748 standard [2] which follows the GUM
[3] approach. The equation for the combined uncertainty in measured discharge
u(Q), at 68% level of confidence, proposed by the ISO 748 standard is expressed
as:
/
2 2 2 –q2i[u2(Bi)+-u2(Di)+-u2p(Vi)+-(1ni) x-[u2c(Vi)+-u2exp(Vi)]]
u (Q ) = us +u s + (/ qi)2
Limitations of this method are well described by Le Coz [4] who proposed an alternative
method for computing uncertainty. The major disadvantage of ISO 748 formula comes from
the estimation of the uncertainty component (noted um) related to the limited number m of
verticals. This component is determined by a table and depends only on the number
m of verticals without taking into account their spatial distribution, complexity
of the riverbed shape and flow distribution. These empirical values are based on
non-traceable experiments while most of the computed uncertainty stems from
this component. Thus, this method is not applicable given the diversity of river
cross-sections.
In this study, we propose a new computation of um depending on the riverbed shape and
the flow distribution complexity. We used a set of 20 gaugings (each is based on a number of
verticals between 33 to 80) at different flow conditions. In order to assess the um term, we
degraded by subsampling the number of verticals by simulating the behavior of stream
gaugers. This method of degradation shows different trends depending on a sampling quality
criteria and flow distribution complexity. Streamgaugings with perfectly smooth riverbed lead
to a small value of um whereas the one with rough shape riverbed lead to a greater value of
um.
The new method has been applied to a set of 3000 streamgaugings and produces more
diversified results compared to the ISO 748 method.
References:
[1] Herschy, R. W. “ The velocity-area method ”. Flow measurement and instrumentation
4, n1 (1993): 710.
[2] ISO. “ÂISO 748:2007 - Hydrometry - Measurement of Liquid Flow in Open Channels
Using Current-Meters or Floats”, 2007.
[3] JCGM. “ Evaluation of measurement data - Guide to the expression of uncertainty in
measurement ”. Guide. BIPM, 2008.
[4] Le Coz, J., B. Camenen, X. Peyrard, et G. Dramais. “ Uncertainty in open-channel
discharges measured with the velocity–area method ”. Flow Measurement and
Instrumentation 26 (août 2012): 1829. doi:10.1016/j.flowmeasinst.2012.05.001. |
|
|
|
|
|
|