Research On The Control Strategy And Optimal Sizing Of The Battery Storage In The Wind-storage Hybrid System

As the global environmental pollution and the energy cris is are becoming more and more severe,the wind energy generation is strongly promoted by the governments worldwide.The wind power penetration level is increas ing all along.However,comparing with the norma l generation units,the wind power has the characteristics of high intermittency,large fluctuation and has diffic ulty in dispatch.When large amount of wind power is directly injected into the grid,it may greatly affect the power quality and the unit commitment,even the safety operation of the power grid will be put in risk.This paper aims at investigating in the control strategy and optimal s izing of the battery storages under two different c ircumstances respectively.One is when large wind farm is directly connected to the high voltage network and the other is s mall distributed wind or solar generation is injected to the low voltage distribution network.The mean work done in this paper is:1.For the purpose of investigating in the optimal control strategy and sizing,two intelligent optimization algorithms are studied: partic le swarm algorithm and genetic algorithm,focus ing on their princ iples and app lic ations in power system.Meanwhile,they are implemented in Matlab and tested with some basic case.2.For the s ituation of large wind farm directly connected with the power grid: To achieve the integrated target of alleviating the wind power fluctuation and compensating the forecast error and thus make full utilization of the battery,a coordination control strategy of the hybrid wind-storage s ystem based on self-adaptive low pass filtering is proposed.At first,d ifferent wind-variation-modes will be identified according to the ultra-s hort term forecast of the wind power.An objec tive function,dedic ated for the optimization of the filter-coeffic ient applied in the hybrid system to smooth the wind power,should be modified in accordance with different types of modes and the SOC of the battery.Next,further amendment of the forecast error is performed depended on the s moothed wind power.A coordination control scheme with active energy feedbac k to rectify battery outp ut power and compens ate the forecast error is put forward.The methodology is verified through simulation.3.For the s ituation of battery storage cooperating with the distributed wind and solar generation to be injected into the low or middle voltage power grid: To make good use of the battery and reduce investment cost,a two-layer optimization algorithm to optimize the s itting and s izing of the storage is put forward: the outer layer aims at minimizing the system cost.An enhanced genetic algorithm is emplo yed to optimize the allocation while improving the computing effic iency.The inner layer applies sequential quadratic programming to acquire the optimized battery charging power when targeting at minimizing the power loss and rais ing the revenue of load shifting.Meanwhile,the inner layer modifies the power and capac ity of the storage to guarantee the cooperation of the two layers.Finally,a 17-bus distribution network is us ed to verify the proposed algorithm.The results show that,the method is capable of finding the optimal allocation of the storage.Moreover,the influence of the wind and solar rated power on the best allocation is analyzed.Finally,conc lus ion and prospection of the researc h of this paper is discussed.