Economic Model Predictive Control for Irrigation Systems

In this paper, we present a watering system based on Economic Model Predictive Control (EMPC) that anticipates the water demand of the crop, and maintains an optimal soil moisture while reducing water consumption. The algorithm relies on a model of the soil moisture incorporating variables such soil evapotranspiration and deep percolation. This model, that was developed based on climatological data of the Kingdom of Bahrain, is formulated as a linear stochastic state space model. Simulation studies demonstrate the potential of the proposed model and algorithm. Compared to traditional operation of on-off automated irrigation systems based on sensors, the optimized EMPC operating strategy leads to significant saving in water consumption approximately around 10-15%, while maintaining an optimal soil moisture level.