Articol Deversor Fanta - v.10
Articol Deversor Fanta - v.10
Articol Deversor Fanta - v.10
Abstract— The current research presents a procedure to than the top width of the approach channel, B so that the ratio
calibrate rectangular contracted slit weirs for measuring flow b/B is in the range 0.01670.25. Experiments performed on
rates less than 5 L/s at various diversion facilities of water multiple slit weirs allowed authors to suggest limitations in the
intakes along small mountain rivers. The stage-discharge following ranges of the design parameters: weir wall, P 0.04
relationship obtained from CFD - 3D numerical simulations is m, h/b = (1.07 55.8) and h/P = (0.135 6.7). Slit weir was
validated with the corresponding curve obtained through
calibrated, so that in the stage - discharge relationship (or
experiments performed using a laboratory flume, on which the
rating curve), Q(h),
weir has been calibrated. Qualitative and quantitative
comparison of the flow nappe over the weir and of the rating (1)
curve obtained from simulations and experiments prove a very
good agreement. Increased sensitivity with head at low the discharge coefficient, Cd, was determined experimentally
discharges proves the slit weir being a good measuring device [4] with the following relationship
for these flow ranges. Results show a new methodology of using
CFD flow simulations over slit weirs to express small discharges . (2)
at river intakes. This methodology is expected to be used even
for more complex flow configurations.
In this equation, the Reynolds number was defined with the
Index Terms: CFD simulation, flow measurement, slit weir, VoF mean velocity in the weir flow nappe
method.
, (3)
I. INTRODUCTION and the weir width, b, as the characteristic length, as follows
Sharp-crested weirs are widely used as discharge , (4)
measuring hydraulic structures and devices in open channel
flows for laboratories, various industrial or environmental in which is the kinematic viscosity of water. Knowing the
water supply facilities, canals, diversions of river intake/outlet accurate discharge coefficient is important particularly in low
facilities, irrigations, green infrastructure monitoring in urban flow rates measurements, where the relative errors between
areas, etc. Different opening shapes of thin-plate weirs are measured and calculated discharges are the highest [4].
encountered in various applications, such as: rectangular (the Aydin et al. [7] increased the working range of slit weirs
most common), triangular, trapezoidal, parabolic, proportional for smaller values of h/b ratios (in the range of 0.28 ≤ h/b ≤
[1]-[3]. Of these, the triangular weir with a small notch angle 55.8). Aydin et al. [7] and Vatankhah [8] found later on other
has been considered for a long time to be the most appropriate empirical relationships of the discharge coefficient,
and easy-to-build device for measuring low flow rates, even if respectively as
for very low flow rates [4].
However, in 2002, Aydin et. al. [5] showed that for . (5)
accurate measurement of very low flow rates (below 5 L/s) in
open channels, a particular version of the contracted At river diversion structures from mountain areas is very
rectangular weir has a higher sensitivity than the triangular important to accurately measure the small water discharges
notch in terms of measurments of the head, h [6]. This slit withdrawn from the stream, especially during critical dry
weir, was designed to have a much smaller opening width, b, periods. However, since this type of weir can be easily
clogged, it is recommended to have an upstream screen to
retain floating materials [9]. At the same time, the small width II. METHODOLOGY
prevents larger fish to pass through this device.
A. Experimental set-up
The main objective of the current research is to compare The experimental measurements are carried out on a
rating curves obtained from experiments with 3D numerical closed-circuit, glass-walled, laboratory flume, with a total
simulations, for measuring and controlling discharges smaller length of 12 m and a rectangular cross-section of H = 500 mm
than 5 L/s at water intakes on mountain rivers. Visualisations in height and B = 400 mm in width (Fig. 1) [10].
of the flow nappe and a qualitative comparison of its shape are
to be performed. Flow is insured with the help of a centrifugal pump
(Făgăraș PCN 65-250 having a designed flow of Q = 50 m3/h,
A secondary objective of the study is to calibrate the built 10m water head and a rotational speed of 1450 rpm) installed
slit weir on a laboratory flume and compare the discharge on the recirculating pipe with a 0.1m diameter, D.
coefficients with values obtained by other researchers.
.
b = 3cm 2.
2
m
The weir was designed to comply with the requirements of In this equation was measured with the manometer.
previous researchers with the following dimensions (Fig. 2): The discharge coefficient of the weir was next computed by
opening b = 0.03 m (so that the ratio = 0.0075), height of equating the discharge values given by relations (1) and (6)
and the rating curve plotted for all measurements.
weir crest, P = 0.12 m, water level above the weir crest, h
(5.5 17.2 mm), so that depth would be smaller than H. Weir
wall, P, was designed so that the flow regime over the crest to
allways be free (not drowned) and the downstream level to
allways be below the crest. The slit weir was built of a 1.5mm
alluminium plate, which was placed vertically on a metal
frame sealed to the flume invert. The head, h, of the free flow
regime (not clinging) was measured at a distance of 0.5cm
upstream weir cross-section for all discharge values. Different
increasing and decreasing flow values were controlled by a
valve, and sufficient time was allowed between
measurements, so that the flow rate is considered constant and
regime steady for each value.
Figure 2. Geometric characteristics of the slit weir
The discharge was measured with a calibrated diaphragm
(orrifice plate meter) with a diameter contraction ratio of =
d/D = 0.42. A 800 mm-scale inverted U-tube differential weir
manometer was connected to the pressure tappings placed at D
and D/2 upstream and downstream distances respectively,
with respect to the diaphragm cross-section.
The diaphragm discharge coefficient, , was obtained (in
the range 0.60.64) from the SR EN ISO 5167-1:2003 and flume
5167-2:2003 through iteratations, as a function of the
Reynolds number in the pipe, ReD, (in the range 9,000-50,000)
by using the following equation:
. (6)
B. Numerical simulations Boundary type Conditions
Numerical simulations were performed with the ANSYS Inlet
Fluent code, which solves the flow governing equations based 0.015; 0.02; 0.025; 0.03 m/s
on the finite volume method, using the incompressible Outlet
unsteady solver, Volume of Fluid (VoF) method and
convergence criteria of 10-6 for the unknown functions.
Solid walls
The VoF model can simulate the flow of two or more stationary, smooth, impermeable walls
immiscible fluids by solving a single set of momentum Interface Free-surface interface
equations and tracking the volume fraction of each of the
fluids throughout the domain. Typical applications include III. RESULTS
the prediction of jet breakup, the evolution of large vapor
bubbles in a liquid, the motion of liquid downstream a dam or A. Experimental results
weir, and the steady or transient tracking of any liquid-gas Weir calibration consisted in fitting equation (1) for the
interface. whole range of discharge values with a power-law function
The 3D numerical domain (Fig. 3) is considered as a flume (7)
sector between two upstream and downstream cross-sections so that coefficients a and k could be determined. In Fig. 4
the slit weir mounting frame, with the dimensions L x H x B = where the measured stage-discharge curve of the slit weir is
2.5 x 0.5 x 0.4 m. The discretization network has the represented in logarithmic scale, one may see the power-law
following characteristics: coefficient, k is very close to 1.5 (1,47). R2 value close to 1
hexaedral, structured mesh cells along , unstructured shows very good agreement between the fitting equation and
along and axis, with variable dimension ranging with experimental curve.
a growth rate of 1.1, from 1 to 15 mm;
a size function was used in the vicinity of the weir, where 0.01000
the highest velocity gradients occur; 0.01 0.1 1
1.73 million cells with maximum shape skewness of 0.47.
y = 0.0504x1.4699
Q (m3/s)
0.60
Aydin, 2002
the gradient ?? normal derivative ?? ??, v is the wall local
Aydin 2006
velocity??, n is the normal direction. Pentru astea tb. Sa 0.55
Vantakhak 2019
vorbim, ca nu inteleg!
0.50
The turbulence model used for the simulations is 0 10000 20000 30000 40000 50000 60000
. The qualitative and quantitative hydraulic parameters Re (-)
resulting from the simulations are compared for validation
with the experiments and similar relevant results from the Figure 5. Discharge coefficients of the slit weir
literature.
Weir sensitivity [1], [10] was expressed by differentiating
equation (7) as
TABLE I. BOUNDARY CONDITIONS APPLIED TO THE NUMERICAL FLOW
DOMAIN . (8)
This means for higher values of dh/dQ, the weir sensitivity is B. Numerical results and validation
higher (very important for measuring low flows). Direct visualizations of the flow nappe through the slit
In Fig. 6 is shown the weir sensitivity as a function of weir opening are qualitatively compared with numerical
discharge in semi-logarithmic scale. The plot shows results for the hydraulic conditions. Lateral and vertical
contractions are analyzed and air/water interface colored by
increasing values up to 55 for decreasing values of the
velocity magnitude (Fig. 7). A good agreement is found
discharge below 1L/s.
between the two types of results, in the same conditions of
head and discharge, both in terms of span and width of the
100 flow nappe. Another aspect considered for the validation of
the numerical results is the position of the detachment point of
the vortex from the solid wall. Both these comparisons are
dh/dQ
10
0.000 0.001 0.002 0.003 0.004
Q (m3/s)
Figure 6. Weir sensitivity dh/dQ as the function of discharge, Q
a) b)
Figure 7. Comparison experiment (a) – numerical simulation (b) of the flow nappe over the slit weir
0.2
Figure 8 depicts the comparison of the stage-discharge
curve with values from experiments and numerical 0.16
simulations. A good agreement is found between the two: a
maximum relative error of 8% in terms of weir head, h, 0.12
h (m)
ACKNOWLEDGMENT