عنوان مقاله [English]
One of the solutions for managing water resource systems is combining simulation and optimization models in order to extract optimal reservoir exploitation policies in the form of a specific optimization. In this research, the combination of the NSGA-II multi objective algorithm and the WEAP simulator model and The dam control curve or the amount of release from the reservoir was optimized in order to supply the downstream demands in these conditions. But the optimal solutions cannot be generalized for other possible inputs to the reservoir and in the event of a change in the input flow into the reservoir the optimal solutions is not effective and the operation of the system in the form of optimization algorithm should be optimized again. Therefore, in order to solve this problem, a new method was developed based on the integration of the Support vector machine method and the NSGA-II algorithm for real-time optimal system operation. The results showed that the average error of the optimal rules of the Support vector machines relative to the output of the NSGA-II algorithm at the verification stage were less than 10%, which indicates the effectiveness of this method in predicting the optimal dam control curve pattern in real time. The extended Support vector machine model has this capability to rapidly optimize the operation policies in a way that allows for optimal system management in real time, due to new inputs to the dam.