Addressing hydraulic parameter uncertainties for resilient irrigation canal modelling and control

Efficient water management in irrigation canal systems demands accurate modelling and control of hydraulic dynamics, especially under conditions of uncertainty. This study investigates the effects of variations in hydraulic parameters, particularly the roughness coefficient, on the performance and reliability of canal models. The roughness coefficient, a critical factor influencing flow resistance, often varies along a canal's length and over time, leading to deviations in water level and discharge predictions. Such uncertainties can significantly impact control strategies and overall water management efficiency.
To address these challenges, we utilized a comprehensive canal model structure (Pandey et al., 2024) to develop models for three distinct canal types and analyzed their behavior under extreme operating conditions. The study simulated scenarios with both uniform and non-uniform flow conditions, incorporating a 30% variation in the roughness coefficient to create tuned and untuned configurations. Through detailed simulations, we evaluated the sensitivity of model parameters, including upstream and downstream water areas and delays, and assessed water level deviations arising from parameter uncertainties. Disturbances were introduced at the downstream end to observe the model's robustness across varying operating conditions.
The findings highlight the substantial influence of roughness coefficient variations on model behavior, particularly in terms of discharge accuracy and water level control. Comparative analysis revealed the limitations of untuned models in handling parameter uncertainties, emphasizing the need for adaptive and robust control strategies. Additionally, the results demonstrate the varying impacts of uncertainties across different canal configurations and flow conditions, providing insights into model reliability and the design of resilient irrigation systems.
By addressing parameter uncertainties and evaluating model responses under diverse conditions, this research contributes to the development of adaptive and reliable water management strategies. The outcomes are crucial for advancing sustainable irrigation infrastructure capable of coping with real-world complexities and variabilities.