The ecological quality of water depends largely on amount of oxygen that the water can hold. Oxygen enters water by entrainment of air bubbles. There is a significant oxygen transfer associated with most hydraulic structures because the air entrained into the flow is split into small bubbles, which greatly increase the surface area for transfer. To design efficient hydraulic structures they must be modeled and analyzed correctly before they are realized. Different methods based on mathematical, statistical and intelligent methods are used for modeling and analyzing. In this paper, comparison of intelligent methods for predicting aeration efficiency of high-head conduits was presented. The intelligent methods used were Neural Network (NN), Adaptive Network based Fuzzy Inference Systems (ANFIS) and Least Squares Support Vector Machines (LS-SVM). The 3-k cross validation test was applied to evaluate the performance of intelligent methods. The predicted values were compared with the experimental measured values and R-2 statistics were calculated and tabulated. All methods have good agreement with experimental results. According to calculated statistics, the best performance was obtained with the LS-SVM model at R-2 0.9815.