Reservoir operation using a robust evolutionary optimization algorithm

- Improved results for a real-world reservoir system scenario were achieved.
- Reservoir operation was improved by introducing a drawdown constraint.
- Good (i.e. near-optimal) solutions were found rapidly and reliably.
- The total annual imbalance between the supply and demand was reduced significantly.
- A state-of-the-art evolutionary optimization algorithm Borg MOEA was assessed.

In this research, a significant improvement in reservoir operation was achieved using a state-of-the-art evolutionary algorithm named Borg MOEA. A real-world multipurpose dam was used to test the algorithm's performance, and the target of the reservoir operation policy was to fulfil downstream water demands in drought condition while maintaining a sustainable quantity of water in the reservoir for the next year. The reservoir's performance was improved by increasing the maximum reservoir storage by 14.83 million m3. Furthermore, sustainable water storage in the reservoir was achieved for the next year, for the simulated low flow condition considered, while the total annual imbalance between the monthly reservoir releases and water demands was reduced by 64.7%. The algorithm converged quickly and reliably, and consistently good results were obtained. The methodology and results will be useful to decision makers and water managers for setting the policy to manage the reservoir efficiently and sustainably.