Nonlinear model predictive control for operational management of a canal system

A nonlinear model predictive control algorithm is applied for operational water management of a navigation canal system. The operational water management task is to maintain navigable water levels with minimum electrical energy (pump) costs. Optimal pump and discharge strategies are calculated using a process model based on the Saint Venant equations for the canal sections and predictions of the non-controllable inputs. A sequential quadratic programming (SQP) method with an interior-point solver for the linear-quadratic subproblems is used for the numerical solution of the resulting large-scale nonlinear programming problem. The model predictive control algorithm is applied for operational water management of the canal system Mittellandkanal/Elbe-Seitenkanal in Germany.