Research On Capacity Optimal Configuration Of Thermal Energy Storage Within CSP For Wind-Photovoltaic-CSP Hybrid Power Generation System In Multi-Scenario

Wind and solar resources are abundant in Northwest China,but due to the single cooperation form between new energy power plants and the influence of the uncertainty and volatility of wind power and photovoltaic,large-scale grid-connection of wind power and photovoltaic has brought challenges to the planning and operation of electric power system and the consumption of new energy.While the concentrating solar power(CSP)plant with thermal energy storage(TES)has certain time-shift characteristics of energy,which can be combined with wind power and photovoltaic power generation with high volatility to realize energy complementarity.Therefore,in this paper,a typical scene generation method is used to transform the large-scale uncertain wind-solar output scene into a certain small-scale scene set with its typical characteristics.Based on this scenario set,a method for the allocation of TES capacity of CSP plant in Wind-Photovoltaic-CSP(W-PV-CSP)hybrid power generation system is proposed.It is of great significance to promote the local consumption of new energy and alleviate the phenomenon of wind and solar abandonment.The main research contents are as follows:(1)Firstly,the electric heater(EH)is used to make the TES of CSP plant and wind and photovoltaic power station effectively combined,and the operating framework and principle of W-PV-CSP hybrid power generation system are clarified.Then,mathematical models and operating characteristics of wind power generation and photovoltaic power generation are introduced.The energy conversion characteristics of CSP plant with TES combined with EH are analyzed,which provides a theoretical basis for the subsequent modeling in this paper.(2)An improved Ng-Jordan-Weiss spectral clustering algorithm(NJW algorithm)is proposed to generate the typical scene of wind/PV/load.This method combines the advantages that spectral clustering algorithm can converge to the global optimal solution on any shape sample space,and the characteristics of Kantorovich distance-based synchronous back subgeneration elimination method(KD algorithm)which considers both distance and probability in the clustering process.Through this algorithm,typical scenes are generated for the historical data of actual regional wind/PV/load.In addition,three internal evaluation indexes for clustering effectiveness will be adopted,together with K-means algorithm and Principal component analysis(PCA)+K-means clustering algorithm for comparison,analysis of the effectiveness of the typical scene generation method,for the subsequent TES capacity configuration model to provide basic data.(3)A dual-layer optimal configuration model of TES capacity of a CSP plant in a W-PV-CSP hybrid power generation system is established.The two-layer model includes configuration optimization layer and operation optimization layer.With the goal of minimum total scenario investment cost and maximum new energy consumption,the configuration optimization layer determines TES capacity configuration and transmits it to the operation optimization layer.The operation optimization layer aims to minimize the daily operation and maintenance cost,formulate the system operation strategy,and then return to the configuration optimization layer to update the configuration scheme of the TES.The configuration optimization layer and operation optimization layer are iterated repeatedly and optimized continuously to obtain the optimal TES rated capacity and thermal charging and releasing power configuration scheme.(4)Based on a typical scenario of wind/PV/load in a region of Northwest China,the two-layer TES capacity configuration model of CSP plant is solved by combining NSGA-Ⅱ and PSO algorithms,and the TES capacity configuration scheme of CSP plant in W-PV-CSP hybrid power generation system in the region is obtained.The results after configuration and the indexes such as wind and solar abandonment rate,electricity purchased from the grid and operation cost under each typical scenario operation scheme are analyzed.The results verify the importance of the research on optimal configuration of TES capacity parameters of photovoltaic power station.