Least-Cost Design of Urban-Drainage Networks

Network layouts were assumed to be known from previous planning activies; inlet hydrographs were arbitrarily chosen. The optimization problem was posed as an N-stage sequential process; network nodes or junctions were treated as the stage variable, conduit invert elevations at the junctions as the state variable, and maximum flow rates and conduit sizes (restricted to commercial sizes) as decision variables. Unsteady flow routing between stages is incorporated into the optimization model. Compared with conventional design methods, use of the approach could achieve overall drainage network cost reductions amounting to 30%. Of this saving, 40% is attributed to use of storage elements as a flow control option, and the remaining 60% achieved by optimal combinations of conduit sizes and use of advanced hydraulic modeling.