| Southwest China is the region with the most plentiful hydropower resources.The strategy of "West to East Power Transmission" has been implemented for more than 20 years,and the Jinsha River,Yalong River,Dadu River,Lancang River,Wujiang River,Hongshui River and other mainstream cascade hydropower stations with over 10 million kilowatts have been built in Southwest China.There are usually one or more large regulating reservoirs with significant controlling function on the mainstream of these river basins,which play dominant roles in the operation of cascade reservoirs.One of the core tasks of the hydropower system is to formulate the power generation operation strategies of cascade reservoirs reasonably,and the key is to make full use of the flexible regulation capacity of the mainstream controlling reservoirs.In this paper,the mainstream controlling reservoirs dominated by power generation in Southwest China are taken as the research objects,focusing on three key issues: water level control at key time points,drawdown control in dry season and water spillage risk in flood season.The main research contents are summarized as follows.(1)Multi-objective optimization method for water level control at key time points of cascade reservoirs.In view of the cascade hydropower system’s long-term energy storage control requirements,the key time points at the beginning of the flood season and the end of the year are selected for the reservoirs with one or over one year regulation capacity.The key water levels are set as optimization variables.A multi-objective optimization model with the objectives of maximum power generation and minimum water spillage is established.This model incorporates water level control strategies before,during and after the flood season and applies multi-objective optimization solution technologies to obtain the control ranges of the key water levels of the controlling reservoirs.Taking the cascade reservoirs on the mainstream of the Wujiang River as the engineering background,a case study was carried out and compared with the conventional chart control method to verify the effectiveness and practicability of the proposed method.(2)Drawdown operation of the controlling reservoir in dry season considering peak shaving benefits.Given the growing demand for flexibility of the power grid,the impact of peak shaving response on the drawdown operation of the controlling reservoir is analyzed.Firstly,the relationship between long-term power generation and short-term peak shaving benefits is explored based on the simulation method.Typical daily load scenarios are generated through cluster analysis,and the time-of-use electricity price strategy is used to determine the real-time electricity price of the load scenarios.A mixed integer linear programming model for short-term peak shaving optimization is established,and the peak shaving price curves are obtained through simulation calculations.Then,the peak shaving price curves are incorporated into the long-term optimization model.The case study results show that the proposed method can effectively balance the relationship between long-term power generation and peak shaving benefits,and the method can obtain a relatively balanced reservoir water level drawdown process.(3)Multi-objective optimization method for drawdown control of cascade reservoirs in dry season.Aiming at the problem of the drawdown control of mainstream cascade reservoirs during dry season,the "Gini coefficient" is introduced to describe the equilibrium degree of energy storage distribution,and a multi-objective optimization model to minimize the Gini coefficient and maximize power generation is constructed.A multi-scenario decision-making method that couples the "random principle" and "satisfaction principle" is proposed.This method is used to explore the impact of inflow uncertainty and initial energy storage,and then determine the control boundary of the Gini coefficient under different scenarios.Based on the Gini coefficient boundary,the water level drawdown scheme of the controlling reservoirs is determined.The application results indicate that the proposed method can balance the power generation benefits and the energy storage requirements,and the method can obtain practical drawdown control schemes for the cascade reservoirs.(4)Long-term power generation optimization method for controlling reservoir considering water spillage risk in flood season.Given the problem of water spillage risk,the flow rate of non-storage water spillage is selected as the risk indicator based on the minimum risk impoundment rule.The possible water spillage risk in each month of the impoundment period is obtained through historical daily inflow data statistics.Based on the Copula function,the joint distribution and conditional probability distribution between the monthly average inflow and water spillage risk are constructed,the water spillage risk when monthly average inflow is a specific value is quantitatively analyzed,and the risk confidence interval is determined.Finally,the water spillage risk is integrated into the objective function of the optimization model in the form of power loss to obtain a more practical operation scheme.The case study results show that the proposed method can effectively reduce the water spillage risk of reservoir power generation operation and improve the operability of optimization results. |