Author(s): Fajing Xiong; Guanghua Guan
Linked Author(s):
Keywords: Canal-tunnel systems; Emergency operation; Mixed-flow transition; Model predictive control; Operational safety
Abstract: Canal-tunnel systems, integral components of large-scale water transfer projects, are highly susceptible to severe hydraulic transients induced by emergency events like abrupt gate closures. Such events can trigger rapid transitions from free-surface to pressurized flow called mixed-flow, posing significant safety risks, including canal overtopping and extreme pressure surges in tunnels. Conventional control strategies, typically reliant on reactive logic or predefined operational rules, are inadequate for managing the complex, nonlinear dynamics inherent in mixed-flow phenomena. To address this deficiency, this study proposes a novel emergency regulation framework based on Model Predictive Control (MPC). The framework utilizes a computationally efficient implicit finite-volume model to predict the evolution of mixed-flow transients over a defined horizon. This predictive capability is integrated into a receding-horizon optimization scheme to compute optimal, real-time gate operations. The control objective is explicitly formulated to prioritize hydraulic safety by minimizing peak water levels and piezometric heads, while concurrently promoting a balanced post-event water storage distribution. A comprehensive case study demonstrates the superior performance of the proposed MPC framework over a conventional manual control strategy. The MPC controller successfully maintains all hydraulic states within predefined safety limits, effectively mitigating the risks of overtopping and over-pressurization. These findings underscore the potential of MPC to transform emergency response from a reactive procedure into a proactive and robust risk management strategy, thereby enhancing the operational safety and resilience of modern water transfer systems.
Year: 2026