Chakraei,Iman

Optimal Operation of Reservoir Using Management of Surface Water and Groundwater with with Fuzzy Techniqu

The managers and reservoir operators are often uncomfortable with the complicated optimization techniques used in the models, which are made much more complex by inclusion of stochastic of hydrologic variables. The fuzzy logic approach may provide a promising alternative for methods used for reservoir operation modeling. Frequently, the water releases from reservoir do not exactly follow the regular operating rule curves because of the uncertainty due to a variable hydrological condition, consideration of specific social and economic factors, and a regulating policy. Operators can then play a major role for guiding the decision on water releases in such situations. To deal with these uncertainties, a fuzzy set theory has been used. The water releases are decided based on operating policies which are developed by merging expert knowledge and observed data through the fuzzy rulebased model. The significant advantage of the fuzzy logic is that it is obvious and easy to understand due to its rule-based structure, which emulates the human thinking, even though specific release rules are not applied. New emerging soft computing techniques such as Artificial Neural Network (ANN), Fuzzy logic and Neuro-Fuzzy (ANFIS) can be gainfully employed to handle such problems when conditions of the systems are uncertain. A neuro-fuzzy system has the potential to capture the benefits of both neural network and fuzzy logic in a single framework. Neuro-fuzzy systems eliminate the basic problem in fuzzy systems design (obtaining a set of fuzzy if–then rules) by effectively using the learning capability of an ANN for automatic fuzzy if–then rule generation and parameter optimization. As a result, these systems can utilize linguistic information from the human expert as well as measured data during modeling. Main target of this study has been made of operation rules of Zayandeh Rood reservoir, using management of surface water and groundwater with inference of fuzzy system, Adaptive Neuro-Based Fuzzy Inference Systems (ANFIS) and fuzzy regression. According to water shortage compared with demand in Zayandeh Rood basin, and also dealing with groundwater intensive decrease in recent years, groundwater decrease during the simulation period is restricted. Then value of removable water from groundwater is calculated and deducted from sum of demands for domestic supply, irrigation and industry. Therefore, share of Zayandeh Rood dam in preparation of downstream demands is obtained. In this study, various simulated models with performance evaluation criteria are compared. In addition, disability of Zayandeh Rood dam in satisfaction of all this basin demands is understood. Results show that ANFIS model has made from optimal data, yielded more sustainability than other models.