Research On Hierarchical Coupling Forecast Generation Scheduling Model Of Hydropower Station Based On Feedback Mechanism

As a low-carbon,environmentally friendly,safe,efficient and renewable energy source,hydropower plays a pivotal role in safeguarding people's livelihood and promoting the healthy development of the national economy.With the gradual withdrawal of the large-scale development of China's hydropower energy system,reasonable reservoir scheduling strategies are becoming more and more important in improving the utilization of water energy resources and increasing the efficiency of power generation.In this paper,the research on the long-term and short-term power generation dispatching problems of the reservoir is carried out.Taking the Three Gorges Hydropower Station as the engineering background and considering the different forecasting scheduling scale factors,the coupling nesting relationship between the scheduling models based on different time scales is described.The time scale coupling nested power generation optimization scheduling model.In addition,the model introduces a real-time feedback correction mechanism to dynamically correct the scheduling decision process according to the real-time feedback of the water level process to improve the overall power generation efficiency.The main research content of this paper contains:(1)In order to fully consider the impact of random inflow runoff sequence on reservoir optimal scheduling,and verify the effectiveness of the nested forecasting optimization scheduling method based on feedback mechanism proposed in this paper,the hydropower station under the condition of uncertain water supply The optimization scheduling problem is studied.The reservoir scheduling process based on uncertain water conditions is transformed into a multi-period,multi-state,non-applicability Markov optimal dynamic decision stochastic process,using stochastic dynamic programming.The theory seeks the maximum expected power generation benefit under uncertain conditions of incoming water.(2)Fully excavate the advantages of various types of forecasting models,and make good use of the forecasting performance of different forecasting models under different conditions,such as forecasting period and different geographical areas,and the existing available water and rain information in the basin.In the study of this paper,the autoregressive wavelet prediction model was selected for the medium and long-term forecast,and the Xin'anjiang model was selected for the short-term forecast.(3)An optimal power generation scheduling model based on coupled nesting at different time scales is established.The model combines the long,medium and short-term scheduling models hierarchically and nested together.The short time scale scheduling model can better formulate the current The scheduling strategy at the moment,the long-term scheduling model can improve the rationality of global scheduling.The model introduces a system dynamics feedback correction mechanism,and the water level is corrected by the feedback mechanism to make the scheduling decision more reasonable.