Irrigation in agriculture is a major source of electricity demand flexibility that goes largely unexploited. In this paper we provide a model and a solution to the problem of scheduling irrigation time to minimize electricity costs while satisfying crop water requirements. We propose to apply rebates (aimed to consume renewable energy surplus) that were traditionally offered to the industrial sector, in the agricultural one. Furthermore an architecture is proposed to overcome some of the limitations that can hinder the adoption of such rebates. The architecture integrates scheduling techniques best studied in the networking literature. Numerical analysis is performed to validate our model and evaluate the proposed scheduling mechanisms, based on real data from a soybean producer and from the corresponding electricity operator. Results indicate that significant cost reductions can be obtained, specially if the rebates are considered.