Research Topic: Accurate hydraulic modeling is essential for the optimal management of irrigation networks.
 
Objective: This research aimed to develop an automated method for calibrating the discharge coefficients of structures in the HEC-RAS model. To this end, an innovative computational framework based on the Firefly Algorithm and using the Python programming language was created, which automatically controls the HEC-RAS software through its Application Programming Interface (API).
 
Method: In this study, 9 discharge coefficients related to three regulatory structures and six offtake structures in the E1R1 canal of the Dez irrigation network were optimized. The objective function was defined as minimizing the Mean Absolute Error between the simulated water depth in HEC-RAS and the target depth (design depth). The algorithm was executed with a population size of 10 over 35 iterations.
 
Results: Quantitative results demonstrated that the proposed method achieved high accuracy, with the final mean absolute error reaching 0.016 meters across 10 independent runs. Sensitivity analysis performed on the optimized discharge coefficients revealed that the downstream offtake structures had minimum discharge coefficient values of 0.32 and 0.31 respectively, while a key regulatory structure had a maximum discharge coefficient of 0.70. The discharge coefficients of the other structures converged within an intermediate range (0.47 to 0.63), indicating the existence of multiple optimal combinations to achieve the desired accuracy.
Conclusions: This research shows that the integration of HEC-RAS with metaheuristic algorithms can serve as an efficient and precise tool for the automated calibration of hydraulic models.