Optimization of Fuzzified Hedging Rules for Multipurpose and Multireservoir Systems

In the conventional zone-based hedging rule, as a common approach, the rationing factors are changed from one zone to another at once, which is not proper for actual reservoir operation. Hence a fuzzy-rule-based approach is employed to create a transition zone to assign rationing factors. This remedy causes gradual variation in the rationing factors and mitigates severe water shortages during drought periods. In this paper, a monthly simulation model linked to an evolutionary algorithm is developed. The proposed model was applied to the Zohre multireservoir system in southern Iran. There are two objective functions to supply minimum flow and agriculture demands over a long-term simulation period. Therefore, a multiobjective particle swarm optimization (MOPSO) algorithm was applied. The results showed that annual and long-term modified shortage index (MSI) values have been improved compared to the current conventional hedging rule. Implementation of the proposed hedging rule for the case study results in improving at least 34 and 21% in the maximum MSI, respectively, for the annual minimum flow and agriculture shortages.