Study On Short Term Optimal Operation Of Giant Cascade Hydropower Stations Considering Runoff Forecast Of Meteorological Information

In recent years,with the completion of giant hydropower stations such as Xiaowan and Nuozhadu on Lancang River,Longtan,Tianyi,and Tianer on Hongshui River,the installed capacity of hydropower in Southwest China has experienced rapid growth.With the characteristics of large installed capacity,fast-tracking load,and flexible regulation,the giant cascade has become the main power source for peak shaving.However,the low accuracy of runoff prediction,sensitive hydraulic connection,and complex power connection make the giant cascade hydropower stations face two major problems in the short-term optimal operation: water spillage and peak shaving.Taking Southwest giant cascade hydropower stations as the research object,this paper studies three aspects: runoff forecast considering meteorological information,shortterm operation of giant hydropower stations with sharing a common tunnel,and peak shaving operation of one-reservoir and multi-cascade hydropower stations considering accurate modeling of water delay time,so as to obtain a precise modeling method for short-term operation of giant cascade hydropower stations to meet the actual requirements,so as to reduce water spillage and improve the peak-shaving effect of cascade hydropower stations.The main work and production of the paper are as follows:(1)A multi-step daily runoff forecasting framework based on meteorological information and gradient boosting regression tree is proposed.The proposed framework takes meteorological data and ground station monitoring data as input,uses the method based on maximum mutual information coefficient for input selection and the prediction model based on gradient boosting regression tree for runoff prediction,and uses six commonly used indicators to evaluate the prediction results.Taking Xiaowan Hydropower Station of Lancang River as an example,the multi-step daily runoff forecast of 1-10 days is carried out.The results show that the proposed daily runoff forecast framework can improve the runoff forecast accuracy of multi-step lead times,especially for the longer lead times.(2)A short-term optimal operation method for a giant hydropower station with multiple units sharing a common tunnel is proposed.A mixed integer linear programming(MILP)model is established to solve the short-term operation problem of giant hydropower station with multiple units sharing a common tunnel,and an iterative solution method based on gross head is proposed.This model not only considers the change of water head,the water head loss of multiple units sharing a common tunnel,the water consumption of unit startup and shutdown,but also considers the water consumption of crossing the vibration zones.In this paper,a new general modeling method of crossing the vibration zones is coupled to the model.The example of Tianshengqiao II hydropower station in Nanpanjiang River shows that this method can effectively solve the problems of short-term operation and crossing the vibration zones of the giant hydropower station with multiple units sharing a common tunnel,and it is a feasible solution.(3)A peak-shaving model of giant cascade hydropower stations by coupling accurate modeling method of water delay time is proposed.In order to solve the problem that the modeling accuracy of water delay time of giant hydropower cascade is insufficient,the accurate modeling method of water delay time and its parameter determination process are proposed,which coupled to the peak-shaving model.At the same time,three model improvement strategies are proposed,which are aggregation of nonlinear constraints,selection of linearization area and periodicity assumption,to improve the speed of solving the model and reduce the solution error.The case study of one-reservoir and six-cascade hydropower stations headed by the Longtan Hydropower Station in Hongshui River shows that the proposed modeling method can fully consider the dynamic characteristics of flow routing and reduce the deviation of peak-shaving operation.The proposed model can give full play to the role of hydropower peak shaving,and the peak shaving effect is obvious.Finally,the full text research work is summarized,and the direction of further research is prospected.