BY: D R Glynn, FLOWSOLVE, London. Tel: +44 181 944 0940

FOR: WRc plc, Swindon.

Quality of water supply is a prime requirement - can be affected by poor flow patterns in reservoirs, e.g. regions of stagnant flow.
Reservoirs are underground, hence flow cannot be observed directly.
WRc and Flowsolve carried out a research project to study flows in a variety of reservoirs, both by measurement and using PHOENICS.
Here a few examples are shown to illustrate the capabilities of PHOENICS in this area.

continued ...

Diameter 73 m.
Inlet and outlet at about 2/3 radius. Inlet submerged and vertically upwards. Outlet on floor.
Plot shows velocity vectors at surface.
Flow pattern is dominated by momentum of inlet stream. Reservoir is relatively shallow, so much of surface flow is dominated by water recirculating near floor.
Flow is "well mixed", no stagnant areas.

Rectangular reservoir 30m x 50m.
Baffles aimed at achieving "plug flow" and preventing recirculations.
Transient calculation, time step 20 mins.
Tracer injected into inlet stream for 3 hours.
Plot shows tracer concentration after 8 hours.
Plug flow is clearly not achieved - there are substantial recirculations behind each of the baffles.

Same case as before, steady flow.
Chlorine transport and decay modelled. Inlet concentration 0.5 mg/l, exponential decay with rate 0.02 per hour.
Low chlorine concentration behind each baffle points to weakness in the design.

Further details may be found in

"Simulation of Flow Patterns in Potable Water Reservoirs"

by D R Glynn, J Creasey and A Oakes,

presented at IPUC Seville May 1997.

wbs