Figure 5: Sequential source allocation with a developed methodology for
multiple contaminants
However, the optimal source allocation strategy in a multi-contaminant
water network is not straightforward. The prioritisation of the cleaner
source is no longer applicable for the multi-contaminant case as the
ranking of the sources is not obvious. This is demonstrated in Figure 6.
Figure 6a shows the demonstration plot of the sink and source CC for a
single sink and two sources. Source 1 and Source 2 are conflicting. The
sources are arranged based on the ascending order of contaminant ’A’
concentration. In this example, the Pinch point occurs at the
contaminant ’B’ if Source 1 is prioritised over Source 2, while for
contaminant ’A’ Pinch is not reached. However, there is room for
freshwater reduction by reducing Source 1 flowrate allocated to sink 1.
This is because Source 1 has a higher concentration in Contaminant B
than Source 2. Reducing the use of Source 1 helps to reduce the load for
Contaminant B. As presented in Section 2.3 for the case of conflicting
sources, the optimal freshwater requirement occurs when all contaminant
limits are reached. To achieve this, Source 1 can be reduced until the
distances from the endpoint of both sink CCs to both of the source CCs
are identical (see Figure 6b), i.e. Pinch Points occur for all the
contaminants. The source CCs can be shifted to the left until both Pinch
points are reached (Figure 6c). This is the optimal source allocation
strategy for the specific sink. Similar steps can be repeated for the
subsequent sinks.
The determination of the sources flowrates by manually adjusting the
curves can be time-consuming, especially when there are plenty of
sources available. This issue can be solved by using Eqs(4-5). To
determine the reduced flowrates of the sources, one can set the
following condition from Eq(4):