ABSTRACT
Dynamic programming (DP) transforms a sequential or multistage decision problem that may contain many interrelated decision variables into a series of single-stage problems, each containing only one or a few variables. Many hydrosystems optimal control problems can be formulated to minimize the sum of the squared deviations of a state variable from a specified target of the state variable, subject to the state equation. The feedback method to solve linear-quadratic optimal control problems is a dynamic programming approach consisting of a stage-by-stage optimization of the objective function subject to the system state equation. The control solves the optimal control problem by deriving a set of feedback rules from a set of recursive equations. Dynamic programming requires that the objective function be separable in order to perform the stage-by-stage optimization. Many groundwater management problems can be formulated, so that the feedback method of control can be used for the solution procedure.
