Beruflich Dokumente
Kultur Dokumente
The growing complexity of building projects and the need for q=NQ (1)
understanding social and economic consequences of its
conception, design, execution, and post construction usage Where,
call for good planning, utilization of resources, cost q = external inlet/outlet flow,
optimization and post construction performance appraisal [2]. Q = internal flow in element and
Water network simulation provides a fast and efficient way of N = incidence matrix defining topology of the
predicting the network behavior, calculating pipe flows, network (having only three values 0, -1, 1)
velocities, head-losses, pressures and heads, reservoir levels, connecting branches to nodes and showing the
reservoir inflows and outflows and operating costs [5, 6, 7]. direction of flow in the network. The -1 (inlet flow)
The problem of the management of water resources is more and 1 (outlet flow) entries are incident to the
and more important on a world scale. In particular, there is a direction of flow.
requirement for novel concepts helping to solve the water
management problem i. e. numerically efficient tools
supporting optimal design (redesign) process for water The relation between water head and head losses
networks. can be presented as follows:
20 www.ijeart.com
Simulation of Water Supply Distribution Networks Systems
(3)
1 1 2
Where, Qi = flow in that element
hi = energy loss in element i
Ri = constant depending on pipe diameter, length, 4
2 3
type.
The vector q* disregards the external inlet and outlet (the flow
1 1 2
is now provided by the imposition of virtual distortion), and it
accounts for the water flow distribution in the closed network
4 (cf. Eq. (6)). There is a set of j (j the number of branches)
2
2 5 equations (8) to be solved in order to create the full influence
matrix D. Each time the right hand-side changes as the unit
3 virtual distortion is applied to another branch. In practice this
3 4
can be realised by applying a pair of inlets-outlets Lik Rkj j
0
Fig. 1 Oriented graph modelling a 2-loop water So, the parameter modification in the system is accounted for
network by superposing the so-called linear response of the original
network and the so-called residual response due to imposition
of the virtual distortion. Therefore, the resultant water head
It has been assumed that the network is supplied distribution can be expressed as:
only through the node No.1 (inlet with intensity q1)
and the only outlet is through the node No.4.
Analogously to the Virtual Distortion Method (VDM) H i H iL H iR H iL D ij j0 (9)
j
applicable to the truss structures [4,5] let me postulate that
And the resultant water flow as:
local modification of a network parameter can be introduced
into the system through the virtual distortion 0 , incorporated
into the formula (4): Q j Q Lj Q Rj Q Lj R j LTij D ij ij j0 (10)
j
21 www.ijeart.com
International Journal of Engineering and Advanced Research Technology (IJEART)
ISSN: 2454-9290, Volume-2, Issue-2, February 2016
R1 R 2 R1 R2 0 H1@ 1 0 i iL Dij 0j
0
0
R 3 H@2 1 R 4 04
j
R1 R1 R 3 R 4 R4 (14)
(11)
Qi Qi Ri ( Dij ij ) 0j
L
R2 R4 R 2 R 4 R5 R 5 H@3 1 R 4 04
0
0
j
'
0 R3 R5 R 3 R 5 R 4 H@4 1 0
Therefore, the virtual distortion to be generated in branch
No.4 to simulate complete blocking of local flow can be
where 4 1 . Assuming the following data: K1=0.2 m3/s,
0
calculated from the following condition:
K2=K3=K4=K5=0.4 m3/s, l1=l2=l3=l5=10.000 m, l4=14.142 m,
q1=0.050 m3/s, H0=0.000 m, we get the following set of Q4 Q4 Ri ( D44 1) 04 0 , or making use of (4)
L
0 4L 1,34 m
0 0.016 0.016 1.032 H@4 1 0
0
(15)
D44 1
4
@ 0 1
The resulting distribution of water heads H = [0.151, Finally the pressure head as well as the flow in modified
-0.251, 0.251, 0.000]T constitutes the 4th column of the network is (after substitution value (15) to relations (6)) as the
influence matrix D. Continuing this procedure for virtual following:
distortions generated in other branches, the full influence
matrix can be determined as:
1 1L D14 04 3,04,396 *1,34 3,57m
0.243 0.757 0.393 0.151 0.607 2 2L D24 04 2,365,099 *1,34 2,23m
0.071 0.071 0.678 0.251 0.322 3 3L D34 04 1,225 0,247 *1,34 0,89m
D (12)
0.071 0.071 0.322 0.251 0.678 5 5L D54 04 1,9 0,248 *1,34 2,23m
0.000 0.000 0.000 0.000 0.000
and the flows:
Q1 Q1 R1 D14 04 0,01216 0,004 * 0,396 *1,34 0,0143 m3/s
L
Taking into account relation (3) and applying it consecutively
@ 0 1
Q2 Q2 R2 D24 04 0,03784 0,016(0,099)1,34 0,0357
L
to each influence vector H , another influence matrix D
can be created, collecting the response to unit virtual m3/s
@ 0 1
distortions in terms of the pressure head . Q3 Q3 R3 D34 04 0.0196 0.016(0.247)1.34 0,0143 m3/s
L
22 www.ijeart.com
Simulation of Water Supply Distribution Networks Systems
1 H1 H2 4,514 0,943 3,57m Linear constitutive relations have been assumed in the above
' '
considerations. In the case of nonlinear problem formulation a
2 H1 H3 4,514 2,282 2,23m
' '
superposition of two virtual distortion fields has to be taken
3 H2 H4 0,943 0,05 0,89m into consideration. The first one, 0 modeling system redesign
' '
VI CONCLUSION
23 www.ijeart.com