Decision Support Model for Ecological Operation of Reservoirs Based on Dynamic Bayesian Network

In this study, a model was proposed based on the sustainable boundary approach, to provide decision support for reservoir ecological operation with dynamic Bayesian network. The developed in four steps: (1) calculating and verifying boundaries combination objectives of study area, (2) generating learning samples by establishing an optimal Monte Carlo simulation model, (3) training network examples (4) probability economic targets exceeding set threshold from time trained model. Using water drawing can be adjusted dynamically according during operation. applied middle reaches Heihe River, effect supply proportion target analyzed, response storage each period calculated. results show that risks analyzed Compared existing studies, provides guidance technical formulation chart. designed guaranteed rate, uncertainty reduces variation range flow shortage or overflow rate loss 5% 6%, respectively. Thus, it seen effectively reduce influence inflow rainfall uncertainties