摘要
Abstract
To address the reliability issues of off-grid power systems in remote areas,retrofitting existing cascade hydropower with pumped storage has been regarded as a novel effective solution.The variable-speed pumped storage,due to its adjustable pumping capacity,is found to be more advantageous compared to fixed-speed systems.Consequently,the formation of a regional 100%clean energy off-grid system through the integration of variable-speed pumped storage with wind and photovoltaic re-sources is enabled.In this context,a data-driven capacity optimization model for off-grid cascaded hydro-wind-photovoltaic complementary system considering pumped-storage retrofitting is developed in this paper.Firstly,with the minimum comprehensive cost on an equivalent year as the objective,the optimal capacity configuration is achieved by considering the operation of the pumped-hydro-wind-photovoltaic system,power supply reliability,and various constraints.Secondly,using the McCormick linearization method,the original mixed-integer nonlinear model is transformed into a mixed-integer linear optimization model.Moreover,to handle uncertainties of load,wind,photo-voltaic,and inflow,a data-driven two-stage distributionally robust optimization configuration model for capacity optimization is presented based on both 1-norm and ∞-norm.The column and constraint generation(C&CG)algorithm is used to solve the proposed model.Finally,the feasibility is verified.
The following conclusions can be drawn from the simulation analysis:(1)With the scale-up of wind power and photovoltaic units,in order to better accommodate the curtailed electricity generated during system operation,the use of data-driven distributed robust optimization methods compared to traditional stochastic optimization methods can better integrate water,wind,and solar resources,reduce curtailment rates,and lower system operating costs.(2)The capacity optimization configuration model established in this paper has good operational performance in terms of accuracy and solution efficiency.It can make investment decisions by considering load demand curves and historical scenario data of wind and photovoltaic output,resulting in a slightly higher equivalent annual cost compared to traditional stochastic optimization methods but with lower conservatism.It can better consider the impact of extreme adverse scenarios on system power supply reliability,thus improving the reliability of off-grid systems.Additionally,considering the stochastic nature of wind and photovoltaic output scenarios,using a mixed norm instead of a single norm can improve accuracy.Moreover,using a parallelizable column and constraint generation algorithm for solving avoids dual transformation and bilinear term computations,leading to higher computational efficiency.(3)Due to the limited reservoir capacity and inadequate regulation capacity of existing cascade hydropower stations,a model for off-grid hybrid variable-speed pumped storage retrofit has been established in this paper.Compared to only adding wind and photovoltaic resources on top of cascade hydropower stations with limited regulation capacity,the retrofit with optimal capacity variable-speed pumped storage units significantly reduces curtailment rates during operation.This reflects a substantial improvement in the flexible regulation capacity of hybrid cascade hydropower after the variable-speed pumped storage retrofit,enabling better integration of wind and photovoltaic resources.(4)Furthermore,the optimization model constructed by retrofitting variable-speed pumped storage units into cascade hydropower stations used in this paper results in lower equivalent annual costs and more curtailed electricity absorption compared to using traditional fixed-speed pumped storage units,demonstrating stronger regulation effects.关键词
混合式抽蓄改造/水-风-光-蓄/供电可靠性/分布鲁棒/列与约束生成Key words
Hybrid pumped storage retrofitting/hydro-wind-photovoltaic-storage/power supply reliability/distributinally robust/column and constraint generation分类
动力与电气工程
