Univrsity Of Tehran Press Water and Irrigation Management 2251-6298 14 4 2025 02 19 Supervised learning to manage irrigation canals’ operation Supervised learning to manage irrigation canals’ operation 1005 1018 99940 10.22059/jwim.2024.384476.1187 FA Kazem Shahverdi Department of Water Sciences and Engineering, Faculty of Agriculture, Bu-Ali Sina University, Hamedan, Iran. Journal Article 2024 10 27 Due to the ever-increasing need for water due to rapid population growth, increased need for food, urbanization, and industry, the pressure on water resources is high. Optimal management of water in the canal networks can play an effective role in reducing this pressure. The research literature review shows that the support vector machine method, as one of the artificial intelligence methods, has received less attention for optimal water management in the canal network. In this research, the support vector machine method was used to manage the operation of the eastern Aghili canal, by considering the discharge of the canal as the input and associated gate opening as the output so that the water depth remains at the target depth, the training of the support vector machine was done. In the next step, the prediction was made for different inputs, and canal simulation was done using a hydrodynamic model, and the criteria for evaluation of equity, dependability, efficiency, and adequacy were calculated, which were respectively smaller than 0.1, smaller than 0.1, larger than 0.85, and greater than 0.9. The results of the vector machine were compared with the results of the artificial neural network, which indicated the superiority of the support vector machine. Due to the ever-increasing need for water due to rapid population growth, increased need for food, urbanization, and industry, the pressure on water resources is high. Optimal management of water in the canal networks can play an effective role in reducing this pressure. The research literature review shows that the support vector machine method, as one of the artificial intelligence methods, has received less attention for optimal water management in the canal network. In this research, the support vector machine method was used to manage the operation of the eastern Aghili canal, by considering the discharge of the canal as the input and associated gate opening as the output so that the water depth remains at the target depth, the training of the support vector machine was done. In the next step, the prediction was made for different inputs, and canal simulation was done using a hydrodynamic model, and the criteria for evaluation of equity, dependability, efficiency, and adequacy were calculated, which were respectively smaller than 0.1, smaller than 0.1, larger than 0.85, and greater than 0.9. The results of the vector machine were compared with the results of the artificial neural network, which indicated the superiority of the support vector machine. https://jwim.ut.ac.ir/article_99940_154e0944fad77aa4ef8c5aeac8f79eb5.pdf