Identification and hybrid control of astation of irrigation by sprinkling

The effects of climate change and the exponential increase in needs, the issue of saving water wastage becomes more important. The objective of this work is to improve the performance of a station of irrigation by sprinkling which we tried to incorporate a new control law guaranteeing a certain performance. Development is done on a real pumping station. The appearance of a temporal delay and the existence of unknown variables make the use of mathematical model founded very difficult. Identification of an approximate model was provided by the application of the algorithm identification N4SID subspaces. To get the best performance, we analyzed the control law PI planted in an educational station sprinkler, and then applied a fuzzy control law and we ended up with encouraging results. An idea of combining the two control laws through a fuzzy supervisor, has improved the results significantly.