Fuzzy control of artificial reservoirs

In this paper a fuzzy control strategy for an artificial reservoir is tackled. Water resources management is characterized by very specific problems, due to non-linearity of systems, complexity of the management rules, water saving needs, etc., therefore it’s necessary to develop specialized modern control strategies to face all the aspects together. For these reasons fuzzy logic was chosen in order to formalize empirical rules management and integrate them in a control system to take in to account different considerations, often inhomogeneous and very hard to express in mathematical terms, in a unique rules framework. The proposed approach is based on a PID (Proportional, Integral and Derivative) fuzzy controller and a closed loop strategy. The goal is to determine a water release management policy, and is accomplished by solving a classic tracking problem, where the variable to track is a given reference water flow profile (water demand). The use of a controller based on fuzzy logic concepts offers the opportunity to distinguish the case of water deficit and water waste and, at the same time, to face the non-linearity of the problem. The effectiveness of the proposed approach is validated by using a MATLAB/SIMULINK dynamic simulator on a 16-years time series data of a real case (Pozzillo reservoir in Sicily, Italy).