Research and design of precision irrigation system based on artificial neural network

In China, the shortage of water resources and low efficiency of irrigation coexist. To improve the utilization of water resources, this paper presents a set of precision irrigation system solutions. In this study, back-propagation (BP) neural network algorithm was applied on the crop evapotranspiration (ET) estimation. The prediction performances of model with different architectures were compared, it indicates the BP model with two hidden layers, using adaptive learning rate, and air temperature and soil moisture as inputs, has acceptable accuracy. Thereafter, using wireless sensor network (WSN) to collect real-time temperature and soil moisture as inputs of ET estimation model, the precision irrigation control system was presented, including its hardware and software system architecture, its intelligent decision-making process and its main flowcharts.