Abstract: In view of the current situation of lack of scientific and technical guidance in farmland irrigation and serious waste of water resources, in order to improve the utilization efficiency of irrigation water, a smart agricultural irrigation prediction model based on LSTM algorithm was proposed based on the architecture of smart agricultural irrigation system.Accurate irrigation can be achieved based on data such as crop growth needs, growth environment and planting soil, which can save water resources to the maximum. Through experiments, the predicted values of LSTM irrigation prediction model and traditional irrigation prediction model are compared and analyzed.The results show that the prediction result of the LSTM model is closer to the actual value and the performance is excellent, which can provide a reliable basis for the realization of intelligent water-saving irrigation.