Multipurpose Rule Curves for Multipurpose Reservoir by Conditional Genetic Algorithm

摘要

Multipurpose rule curves are suitable guide lines to operate multipurpose reservoirs during a considered time period. This study applied the conditional genetic algorithm and reservoir simulation model by considering multipurpose objective functions. 25 years of historic inflow data (1992-2016) and 50 years of future inflow (2017-2066) under climate change scenario B2 data and land use change data for the Man Oon Reservoir in Thailand have been considered. The extreme inflow, the future inflow and the synthetic inflow of reservoirs were used in evaluating the obtained rule curves. The situations of water shortage water excess as well as stored water at the end of the wet season were shown. The results showed that the multipurpose rule curves with objective function of minimized average excess spill water, can mitigate flood situations more than the other rule curves, furthermore it is possible to have more stored water at the end of wet season than water stored when using the other rule curves.