Model predictive control based on an integrator delay coupled model applied to open canal irrigation systems

Model-based control of open canal irrigation systems (OCIS) has become a crucial strategy to guarantee reliable water resources and enhance water-use efficiency in irrigation districts. The single-pool canal modeling strategy, exemplified by the integrator delay (ID) model, neglects the coupling effects among canal pools. This study proposes the integrator delay coupled (IDC) model by incorporating the coupling effect from the adjacent downstream pool. The model parameters are identified using the least squares method for comparison of model prediction accuracy. A centralized model predictive control (CMPC) is designed for control performance validation. Results show that the IDC model accurately predicts water level fluctuations, with the maximum absolute error reduced by 16–86%, and the root mean square error (RMSE) improved by 16–89%, compared with the ID model. Furthermore, the application of the ID model-based CMPC (ID-CMPC) and IDC model-based CMPC (IDC-CMPC) to the Mawan Canal demonstrates that IDC-CMPC outperforms ID-CMPC, with a 14–42% reduction in overshoot and a 30–35% decrease in transition time. In a word, the IDC model provides a more accurate prediction for cascaded canals, contributing to more precise and faster control of OCIS.