Model Predictive Controller Applied to Optimize Energy and Water Resources in Strawberry Crops

Abstract

This paper presents an automated irrigation system for strawberry crops, which the main goal is to keep soil moisture in a suitable range by making efficient use of energy and water resources. A Constrained Matricial Predictive Controller- CMPC was used because it allows projecting the behavior of moisture through a prediction horizon, estimating the perturbations and responding appropriately to them. To feed the model, evapotranspiration data from the climatological station of RAMPA research group were systematized for three years. The solution was evaluated on four scenarios with different evapotranspiration curves. The key advantage of the proposed controller is the possibility to get the minimum amount of energy and water resources necessary to achieve the control objective using an optimization scheme. The results show that the proposed controller obtains a better performance with respect to reference PID controller, widely used in this kind of applications.