Real-time optimization of urban channel gate control based on a segmentation hydraulic model

With the urban water resources becoming increasingly scarce, the optimal control engineering has emerged as a promising approach to improve the efficiency of water use in the environment. A hydraulic model is capable of accurately modeling and predicting the complex hydrodynamic processes occurring within a channel. However, its optimization and simulation time are often prolonged by the complexity of the channel system, resulting in poor real-time performance. This study presents a segmented hydraulic real-time optimization approach that combines rule-based simulation (RS) with real-time optimization (RTO). The aim of the proposed method is to reduce hydraulic model complexity and improve optimization time by dividing the full hydraulic model (FHM) into optimized segmented hydraulic model (SHMO) and non-optimized segmented hydraulic model (SHMN). The approach presents two main improvements: (1) a segmentation point recognition method based on RS is used to obtain SHMO from the FHM; and (2) a segmented optimization framework is employed to enable RTO based on SHMO. We demonstrate the effectiveness of the approach using a case study of China's Qing River. The results indicate that FHM can be successfully divided into SHMO and SHMN with similar simulation effect (R > 0.88 and RMSE < 0.1) by using the segmentation point recognition method, and the segmented hydraulic real-time optimization approach can reduce optimization time (average 68%) of hydraulics model. The case study indicated that the proposed method is a computationally efficient and feasible approach for real-time regulation of urban channel gate control based on hydraulic model.