Hybrid modeling and receding horizon control of sewer networks

In this work, a control‐oriented sewer network model is presented based on a hybrid linear modeling framework. The model equations are described independently for each network element, thus allowing the model to be applied to a broad class of networks. A parameter calibration procedure using data obtained from simulation software that solves the physically based model equations is described and validation results are given for a case study. Using the control model equations, an optimal control problem to minimize flooding and pollution is formulated to be solved by means of mixed‐integer linear or quadratic programming. A receding horizon control strategy based on this optimal control problem is applied to the case study using the simulation software as a virtual reality. Results of this closed‐loop simulation tests show the effectiveness of the proposed approach in fulfilling the control objectives while complying with physical and operational constraints. Key Points Sewer Network Modeling: common sewer network elements are modeled individually Optimal Control Problem formulation for hybrid linear delayed systems Case Study: a receding horizon control strategy is applied to a real network