Reliability-based operation of reservoirs: a hybrid genetic algorithm and cellular automata method

Abstract

A novel hybrid genetic algorithm and cellular automata method is presented here for solving reliability-based reservoir operation problems. The method uses a decomposition approach in which the reliability part of the problem is handled with a GA, while the operational side of the problem is solved by a CA method. In the proposed method, a binary genetic algorithm is employed to determine success/failure periods of the operation. A period is considered as a success period if the corresponding gene value is one, and failure period if otherwise. The success/failure pattern suggested by the genetic algorithm is used to explicitly rewrite the reliability constraints of the operation problem in terms of the reservoir releases. A CA method is then used to solve the resulting problem in which the two types of the constraints, namely operational constraints and reliability constraints, are dealt with differently. Operational constraints are applied to all periods, while reliability constraints are only applied to the success periods determined by genetic algorithm. The proposed method is used for monthly water supply and hydropower operation of Dez reservoir in Iran, and the results are presented and compared with those of a GA model. The efficiency and effectiveness of the method are tested for the short-, medium- and long-term operation periods assuming different target reliabilities. Comparison of the results with those of a GA model shows that the proposed method is both more effective and efficient compared to GA.