Integrated multi-objective model predictive control framework for cascaded open-channel systems with multi-source lateral inflows

Terminal ten cascaded pools of the Middle Route of the South-to-North Water Diversion Project (MR-SNWD), China, with lateral inflows from along-route reservoirs. Reliable water delivery in terminal canal reaches is challenged by demand fluctuations and lateral inflows acting under wave-propagation delays. To address timing and allocation mismatches, we propose a control-oriented Multi-source Lateral Inflow Integrator–Delay (MLI-ID) model that treats the main inflow and each lateral inflow as independent delayed inputs. Building on MLI-ID, a Lateral Inflow Feedforward Guided MPC (LIFG-MPC) framework embeds multi-objective lateral inflow allocation—operation performance, demand satisfaction, and source allocation—into rolling-horizon control, supported by a radar chart–based integration index for real-time compromise selection. Coordinated regulation of multi-source lateral inflows markedly improves water-level regulation and increases supply reliability across varying demand scenarios in the terminal reach. The results reveal a capacity–delay controlled mechanism: along-route reservoirs function as fast-response buffers that can absorb incremental demand within their combined supply capacity, whereas exceedance of this capacity shifts the system’s dependence toward upstream releases whose longer propagation delays restrict timely compensation and define a practical reliability boundary for regional water delivery. These findings underscore that explicitly accounting for multi-source inflow delays and allocation–control coupling is essential for stable operation of cascaded open-channel systems in the MR-SNWD and similar regulated transfer canals. • A Multi-source Lateral Inflow Integrator–Delay (MLIID) model is developed. • A feedforward guided MPC framework embeds optimized allocation as references. • A radar chart-based method integrates multi-objective trade-offs in real time. • Case study shows improved prediction accuracy and expanded Pareto frontier. • Framework supports resilient and sustainable operation of inter-basin transfers.