Real-time control of floods along the Demer River, Belgium, by means of MPC in combination with GA and a fast conceptual river model

Today, retention reservoirs along the Demer River in Belgium are controlled by means of adjustable gated weirs based on fixed rules. To improve the regulation of these weirs, this paper investigates a method for real-time flood control. Because of the excessive calculation time of detailed hydrodynamic models, a conceptual model was used. This model combines the Model Predictive Control (MPC) technique with a Genetic Algorithm (GA). For several scenarios, MPC controls the future state of the river network by changing the positions of the adjustable weirs. The GA generates these positions. Damage functions depending on water levels, were introduced to evaluate the efficiency of each scenario, based on flood damage minimization. The influence of the most important parameters of the MPC-GA-algorithm is investigated by means of a sensitivity analysis. Simulation results show that the algorithm manages to reduce the total flood volume during the historical event of September 1998 by 46% in comparison with the current regulation based on fixed rules.