Aeration of flow downstream of outlet gates is an Effective way to eliminate the risk of cavitation. Many works have been done and various relationships have been developed to predict the quantity of entrained air. Owing the complexity of flow in the aeration zone arising from the two-phase flow, these relationships cannot however be used in general. On the other hand, in recent years, applications of Artificial Intelligence, such as Neural Network, Fuzzy Logic, and Generic Algorithm have attracted the attention of many investigators. These are known as powerful tools to solve engineering problems with uncertainties. In this paper, based on experimental data obtained from field measurements and physical model studies, an Artificial Neural Network (ANN) with a general back propagation error, is suggested to estimate the air demand downstream of bottom outlet gates. The results with a regression parameter of 0.992 showed that the model is very well capable of predicting air demand.