Environmental Engineering Reference
In-Depth Information
Table 6.30
Statistical analysis, 15 NGT networks of 200 nodes (fixed D, variable Q)
R
-value
NCI1 NCI2 NCI3 NSI1 NSI2
NSI3
ADF
avg
NBI
I
n
AGD-dia
-0.18
-0.34
-0.35
-0.16
-0.32
-0.32
-0.31
-0.12
-0.02
GD
0.58
0.71
0.82
0.60
0.73
0.84
0.42
0.18
0.28
Deg
avg
0.58
0.71
0.82
0.60
0.73
0.84
0.42
0.18
0.28
BC
avg
-0.18
-0.34
-0.35
-0.16
-0.32
-0.32
-0.31
-0.12
-0.02
CC
avg
0.13
0.28
0.28
0.12
0.27
0.27
0.29
0.09
0.07
Cp
avg
0.18
0.50
0.37
0.21
0.53
0.40
0.34
0.13
-0.06
P
-value
NCI1 NCI2 NCI3 NSI1 NSI2
NSI3
ADF
avg
NBI
I
n
AGD-dia
0.522 0.217 0.207 0.574 0.250
0.239
0.261 0.664 0.947
GD
0.024 0.003 0.000 0.019 0.002
0.000
0.115 0.514 0.313
Deg
avg
0.024 0.003 0.000 0.019 0.002
0.000
0.115 0.514 0.313
BC
avg
0.522 0.217 0.207 0.574 0.250
0.239
0.261 0.664 0.947
CC
avg
0.641 0.307 0.307 0.669 0.326
0.326
0.288 0.754 0.793
Cp
avg
0.524 0.057 0.171 0.459 0.042
0.138
0.208 0.648 0.825
Correlation NCI1 NCI2 NCI3 NSI1 NSI2
NSI3
ADF
avg
NBI
I
n
AGD-dia
R
R
R
R
R
R
R
R
R
GD
95%
99% 99.9%
95%
99% 99.99%
R
R
R
Deg
avg
95% 99% 99.9% 95% 99% 99.99%
R
R
R
BC
avg
R
R
R
R
R
R
R
R
R
CC
avg
R
R
R
R
R
R
R
R
R
Cp
avg
R
R
R
R
95%
R
R
R
R
Finally, it is remarkable to see that the results in Tables 6.25 to 6.27 show better correlations
than those in Tables 6.28 to 6.30. Especially poor correlation is registered in relation to
ADF
avg
and
NBI
in case of the NGT networks, which could be a consequence of the impact of
hydraulic parameters, influenced by the selected pipe diameters i.e. increased resistance.
6.10
CONCLUSIONS
The research in this chapter aims to analyse impacts of node connectivity on network
reliability. Two approaches have been compared to evaluate this connectivity: one which
does it in relative terms and in simplified way, by taking into consideration the connectivity
of the two most dominant node categories (in three different ways), and the other which
considers connectivity measures based on graph theory. The first approach has been coded
within so called
network diagnostics tool
(NDT), while the NodeXL spreadsheet template in
public domain has been used for the second approach. The mutual correlation has been
analysed, and for each approach separately with the reliability measures discussed in Chapter
5, namely the
ADF
avg
,
NBI
and
I
n
. The test networks have included synthetic sets of various
degrees of complexity and fixed/variable nodal demands and pipe diameters. The conclusions
from the research can be formulated in the following bullets:
1.
The results show variable correlations, which are mostly weak but indicate potential for
further research. While it is obvious that the connectivity measures alone can hardly be
correlated to the reliability measures that are also based on pipe geometry and network
hydraulic performance, some of these could possibly be considered in combination with
additional network input parameters influencing network energy balance. Although drawn
Search WWH ::
Custom Search