ABSTRACT
Sedimentation occurring in reservoirs and rivers are always associated with floods being stored and conveyed. Continued sedimentation in the river over time physically diminishes the channel's capability to contain flows, posing a major threat to the economics of the system. Carriage and Mays developed nonlinear programming models to determine reservoir releases for sediment control. Uncertainties were also considered on some sediment transport parameters identified—like the sediment load and mean grain size—leading to the formulation of three chance-constrained models. The system model is described basically by the relations associated with the reservoir operation and the river hydraulics and sedimentation. The hyperbolic penalty function introduced by Lin is used for the constrained sedimentation control problem because of its analytical characteristics as described by Culver and Shoemaker in their groundwater remediation studies. In order to evaluate the capability of the optimal control model developed, three problem scenarios for the Agua Fria River are considered.
