To quantify the uncertainty of reservoir inflow runoff and the load demand of power grids and analyze their influence on the optimization and dispatch processes of a reservoir,with the Three Gorges Reservoir as an example,a polyhedral uncertainty set of the runoff and load is established,which introduces the robust optimization theory.By combining with the k-means clustering algorithm,the randomly simulated inflow and load scenarios are clustered under various uncertainty scenarios.A multiobjective optimal dispatch model is established to minimize the devia-tion between the actual output and the planned output of the power station,maximize the total power generation,and minimize the change in the downstream suitable ecological flow.The results of the multiobjective particle swarm optimization algorithm show that considering the uncertainty of reservoir inflow and load demand,the fluctuation law of the reservoir water level in each scenario is basically the same as that of the actual water level.Moreover,com-pared with the actual water level,the reservoir can be operated while maintaining higher water levels for longer pe-riods,which improves the overall power generation level of the Three Gorges Power Station during the water storage period.
维普
万方数据